encevis is intended for the review, monitoring and analysis of EEG recordings made by electroencephalogram (EEG) devices using scalp electrodes and to aid neurologists in the assessment of EEG. This device is intended to be used by qualified medical practitioners who will exercise professional judgment in using the information.
The seizure detection component of encevis is intended to mark previously acquired sections of adult (greater than or equal to 18 years) EEG recordings that may correspond to electrographic seizures, in order to assist qualified clinical practitioners in the assessment of EEG traces. EEG recordings should be obtained with a full scalp montage according to the standard 10/20-system.
The spike detection component of encevis is intended to mark previously acquired sections of the patient's EEG recordings that may correspond to spikes, in order to assist qualified clinical practitioners in the assessment of EEG traces. The Spike detection component is intended to be used in adult patients greater than or equal to 18 years. encevis Spike detection performance has not been assessed for intracranial recordings.
encevis includes the calculation and display of a set of quantitative measures intended to monitor and analyze the EEG waveform. These include frequency bands, rhythmic and periodic patterns, burst suppression and spectrogram. These quantitative EEG measures should always be interpreted in conjunction with review of the original EEG waveforms.
The aEEG functionality included in encevis is intended to monitor the state of the brain.
encevis provides notifications on an on-screen display for seizure detection, electrographic status epilepticus detection, spike detection, quantitative EEG and aEEG that can be used when processing a record during acquisition (online) or based on stored EEG files (offline). Notifications can also be provided to external systems via the external interfaces to make them accessible to the user through the external system in a human-readable format. Delays of up to several minutes can occur between the beginning of a seizure, electrographic status epilepticus, the occurrence of a spike or detection of quantitative EEG features and when the encevis notifications will be shown to a user. encevis notifications cannot be used as a substitute for real time monitoring of the underlying EEG by a trained expert.
encevis PureEEG (Artifact Reduction) is intended to reduce EMG and electrode artifacts in a standard 10-20 EEG recording. PureEEG does not remove the entire artifact signal and is not effective for other types of artifacts. PureEEG may modify portions of waveforms representing cerebral activity. Waveforms must still be read by a qualified medical practitioner trained in recognizing artifact, and any interpretation or diagnosis must be made with reference to the original waveforms.
This device does not provide any diagnostic conclusion about the patient's condition to the user.
The encevis Component for Detection of Seizures and Electrographic Status Epilepticus is indicated for the detection of Seizures and Electrographic Status Epilepticus in patients greater than or equal to 18 years of age who are at risk for seizures. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis analyzes EEG waveforms and identifies patterns that may be consistent with seizures and electrographic status epilepticus as defined in the American Clinical Neurophysiology Society's Guideline 14. EEG recordings should be obtained with a full scalp montage according to the standard 10/20-system. The diagnostic output does also include a measure of seizure prevalence ("seizure burden") within a 10 minute (short-term seizure burden) and a 60 minute (hourly seizure burden) moving window. The output of the Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is intended to be used as a diagnostic output for determining patient treatment in acute-care environments. Detections from the Component for Detection of Seizures and Electrographic Status Epilepticus of encevis provide one input to the clinician that is intended to be used in conjunction with other elements of clinical practice to determine the appropriate treatment course for the patient. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is intended for detection of electrographic status epilepticus only. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis does not substitute for the review of the underlying EEG by a qualified clinician with respect to any other types of pathological EEG patterns. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is not intended for use in Epilepsy Monitoring Units.

Device Description

encevis combines several modalities for viewing and analyzing EEG data in one integrated software package. The software package can be used both as a standalone desktop application for opening and analyzing stored EEG files (offline mode) and as a module for integration into external EEG systems via the provided API interfaces, enabling the processing of real-time streaming data in online mode. encevis consists of the following modalities: encevis EEG-viewer, Artefact reduction encevis PureEEG, Seizure detection of encevis NeuroTrend, Detection of seizures and status epilepticus of encevis acute care, Spike detection encevis EpiSpike, Pattern detection and aEEG, Spectrogram, External Interface "encevis AITInterface", External Interface "encevis SeizureICUInterface".

AI/ML Overview

Here's a breakdown of the acceptance criteria and study details for the encevis (2.1) device, based on the provided text:

Acceptance Criteria and Device Performance

The document outlines acceptance criteria and performance for several components of the encevis (2.1) device.

Table 1: Acceptance Criteria and Reported Device Performance

Component / Metric	Acceptance Criteria (Implicit)	Reported Device Performance (encevis 2.1)	Predicate Device Performance (Persyst 12 where applicable)
Seizure Detection	Non-inferiority to Persyst 12 in PPA.	Patient-wise PPA: 97.6% (95% CI=[92.6, 99.5])	Patient-wise PPA: 83.7% (95% CI=[71.4, 91.7])
		Overall PPA: 93.8% (95% CI=[87.0%, 97.7%])	Overall PPA: 77.3% (95% CI=[67.7%, 85.2%])
	Higher sensitivity than Persyst 12.	Significant superiority in patient-wise PPA (p=0.003).
False Positive Rate	Accepted higher false positive rate due to sensitive operating point.	Average NDR: 33.7 false detections in 24 hours (95% CI=[25.5, 47.7])	Average NDR: 10.5 false detections in 24 hours (95% CI=[7.4, 15.4])
Seizure Detection (Acute Care)	Non-inferiority to Persyst 12 in sensitivity (PPA).	Event-based PPA: 71.6% [54.0 % - 86.9 %]	Event-based PPA: 41.5 % [23.3 % - 62.7 %]
False Positive Rate (Acute Care)	Accepted higher false positive rate for comprehensive seizure detection.	NDR: 2.0 / hour [1.1 - 3.7]	NDR: 0.26 / hour [0.049 - 0.84]
Status Epilepticus Detection (ESE)	PPA and NPA comparable to Ceribell Status Epilepticus Monitor.	PPA: 82.6% [CI 60.9%-95.7%]	Ceribell: 100% (various CIs), NPA: 94% [91%, 96%]
	NPA: 91.4% [CI 81.0%-96.6%]	Ceribell: 100% (various CIs), NPA: 94% [91%, 96%]
Hourly Seizure Burden (HSB) (>10% threshold)	High PPA and NPA.	PPA: 86.8% [Cl 75.5%-94.3%]	Not reported
		NPA: 87.7% [Cl 81.8%-92.2%]	Not reported
Short-time Seizure Burden (STSB) (>10% threshold)	High PPA and NPA.	PPA: 91.3% [Cl 82.6%-97.1%]	Not reported
		NPA: 85.5% [Cl 79.0%-90.6%]	Not reported
Short-time Seizure Burden (STSB) (>50% threshold)	High PPA and NPA.	PPA: 88.6% [C] 77.3%-95.5%]	Not reported
		NPA: 95.1% [Cl 90.8%-97.5%]	Not reported
Spike Detection (PPA)	Non-inferiority to Persyst 12 with a 3% margin.	Average PPA: 84.81% (95% CI=[78.5-91.1])	Average PPA: 8.7% (95% CI=[4.4-13.0])
Spike Detection (NPA)	Non-inferiority to Persyst 12 with a 3% margin.	Average NPA: 98.58% (95% CI=[98.1.-99.1])	Average NPA: 99.69% (95% CI=[99.4-99.9])
Spike Detection (PLPA)	Non-inferiority to Persyst 12 with a 3% margin.	Average PLPA: 95.63% (95% CI=[91.0-100.2])	Average PLPA: 93.97% (95% CI=[83.6-104.31])
Artifact Reduction (Relative Suppression of clean EEG)	Non-inferiority to Persyst with a 1dB margin.	95% delta CI=[-0.07, -0.02] (margin = 0.01)
Artifact Reduction (Signal-to-noise ratios after artifact removal)	Non-inferiority to Persyst with a 1dB margin.	95% delta Cl=[4.37, 5.88] (margin = 0.01)
Rhythmic and Periodic Patterns (ANY type)	High sensitivity and specificity.	Sensitivity: 81.86% (79.9 - 83.8), Specificity: 83.80% (83.1 - 84.5)	Not reported
Rhythmic and Periodic Patterns (PD type)	High sensitivity and specificity.	Sensitivity: 69.73% (67.2 - 72.3), Specificity: 95.89% (95.5 - 96.3)	Not reported
Rhythmic and Periodic Patterns (ARA type)	High sensitivity and specificity.	Sensitivity: 89.40% (84.2 - 94.6), Specificity: 94.85% (94.5 - 95.3)	Not reported
Rhythmic and Periodic Patterns (RDA type)	High sensitivity and specificity.	Sensitivity: 91.73% (86.4 - 97.1), Specificity: 86.05% (85.4 - 86.7)	Not reported

Study Details

2. Sample Size and Data Provenance

Seizure Detection:
- Test Set: 55 subjects (1603 hours of EEG data, max 30 hours per subject)
- Data Provenance: Retrospective, patients from an epilepsy monitoring unit. Countries of origin are not specified, but the context implies it is likely from a clinical setting.
Seizure Detection and Status Epilepticus (Acute Care):
- Test Set: 81 patients (62.4 hours of EEG data)
- Data Provenance: Retrospective, neurological/general intermediate care units or neurological/general intensive care units at two different sites in the US (31 patients) and outside of US (50 patients).
Spike Detection:
- Test Set: 23 patients
- Data Provenance: Retrospective, patients from an epilepsy monitoring unit. Countries of origin are not specified, but the context implies it is likely from a clinical setting.
Artifact Reduction:
- Test Set: 128 EEG data records (10 seconds each) from different patient groups (60 epilepsy monitoring, 65 ICU patients).
- Data Provenance: Retrospective, epilepsy monitoring units and ICU settings.
Rhythmic and Periodic Patterns:
- Test Set: 83 long-term EEGs from ICU patients, first minute of each hour, split into three 20-second segments (11935 common annotation segments).
- Data Provenance: Prospective, two different centers. Countries are not specified.
aEEG:
- Test Set: "Real EEG data" for comparison with Persyst. (Specific sample size not provided for this comparison). Also sinusoidal test data.
- Data Provenance: Not explicitly stated, but "real EEG data" implies clinical origin.
Frequency Bands:
- Test Set: Sinusoidal test data and manually selected EEG recordings from epilepsy/ICU patients. (Specific sample size not provided).
- Data Provenance: Not explicitly stated, but "manually selected EEG recordings from epilepsy- or ICU patients" implies clinical origin.
Burst Suppression:
- Test Set: 83 long-term EEGs from intensive care patients (3978 valid annotation segments from the first minute of each hour).
- Data Provenance: Retrospective?, two different centers. Countries are not specified.
Spectrogram:
- Test Set: Artificially created data and real EEG data. (Specific sample size not provided).
- Data Provenance: Not explicitly stated, but "real EEG data" implies clinical origin.

3. Number of Experts and Qualifications for Ground Truth

Seizure Detection: 3 independent neurologists, blinded review. Qualifications not explicitly stated beyond "independent neurologists".
Seizure Detection and Status Epilepticus (Acute Care): 6 experienced, board-certified, and independent Neurologists, blinded review. Qualifications specified as "experienced, board-certified".
Spike Detection: 3 independent neurologists, blinded review. Qualifications not explicitly stated beyond "independent neurologists".
Artifact Reduction: 3 independent epileptologists or neurologists, blinded review.
Rhythmic and Periodic Patterns: 2 clinical neurophysiologists, naive to the EEGs.
Burst Suppression: 2 clinical EEG experts.

4. Adjudication Method for Test Set

Seizure Detection: An event was considered a "true seizure" if the time interval of two out of three reviewers overlapped by at least 1 second.
Seizure Detection and Status Epilepticus (Acute Care): Reference standard for seizures derived from 6 independent neurologists. Reference standard for ESE and seizure burden derived from consensus seizure annotations. The specific voting rule for "consensus" is not explicitly stated, but implies agreement among experts.
Spike Detection: An event was considered a "true spike" if the time interval of two out of three reviewers overlapped.
Artifact Reduction: Not explicitly stated for artifact detection itself, but for identifying clean EEG patterns and artifacts, 3 independent epileptologists or neurologists were involved. Implies consensus or agreement.
Rhythmic and Periodic Patterns: Annotations had to be consistent between both reviewers to be used in sensitivity and specificity measurement. Cohens' kappa statistic (0.66) indicates substantial agreement.
Burst Suppression: The detection performance was analyzed for consensus annotations of the two reviewers. Consensus annotations only included segments where both reviewers showed the same decision.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

There is no explicit mention of an MRMC comparative effectiveness study where human readers improve with AI vs without AI assistance. The studies primarily focus on the standalone performance of the AI algorithms and compare them to predicate devices (other algorithms). In the case of "Rhythmic and Periodic Patterns", human reader agreement (inter-reader agreement) is used to establish ground truth, not to evaluate human performance with/without AI assistance.

6. Standalone Performance Study

Yes, standalone (algorithm only without human-in-the-loop performance) studies were done for all major components. The reported metrics like PPA, NPA, NDR, sensitivity, and specificity are all measures of the algorithm's performance against the established ground truth.

7. Type of Ground Truth Used

Expert Consensus: This is the predominant type of ground truth used across all evaluated components. Experts (neurologists, epileptologists, clinical neurophysiologists) retrospectively reviewed EEG recordings and marked events like seizures, spikes, ESE, and patterns.
Artificial Data: Used for validating aEEG, frequency bands, and spectrogram for initial functional verification.
Pre-calculated Values: Used for validating the quantitative measure of amplitude loss in burst suppression.

8. Sample Size for the Training Set

The document does not provide information on the sample size used for the training set for any of the encevis (2.1) components. The studies described are validation studies using a test set.

9. How the Ground Truth for the Training Set Was Established

Since the document does not provide information on the training set, it does not describe how the ground truth for the training set was established.

Summary

{0}------------------------------------------------

Image /page/0/Picture/0 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). On the left is the Department of Health & Human Services logo. To the right of that is the FDA logo, which is a blue square with the letters "FDA" in white. To the right of the blue square is the text "U.S. FOOD & DRUG ADMINISTRATION" in blue.

September 27, 2024

AIT Austrian Institute of Technology GmbH Tilmann Kluge Official Correspondent Giefinggasse 4 Vienna, Vienna 1210 Austria

Re: K240993

Trade/Device Name: encevis (2.1) Regulation Number: 21 CFR 882.1400 Regulation Name: Electroencephalograph Regulatory Class: Class II Product Code: OMB, OMA, OLT Dated: April 11, 2024 Received: April 11, 2024

Dear Tilmann Kluge:

We have reviewed your section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (the Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. Although this letter refers to your product as a device, please be aware that some cleared products may instead be combination products. The 510(k) Premarket Notification Database available at https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm identifies combination product submissions. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.

If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.

Additional information about changes that may require a new premarket notification are provided in the FDA guidance documents entitled "Deciding When to Submit a 510(k) for a Change to an Existing Device"

{1}------------------------------------------------

(https://www.fda.gov/media/99812/download) and "Deciding When to Submit a 510(k) for a Software Change to an Existing Device" (https://www.fda.gov/media/99785/download).

Your device is also subject to, among other requirements, the Quality System (QS) regulation (21 CFR Part 820), which includes, but is not limited to, 21 CFR 820.30. Design controls; 21 CFR 820.90. Nonconforming product; and 21 CFR 820.100, Corrective and preventive action. Please note that regardless of whether a change requires premarket review, the OS regulation requires device manufacturers to review and approve changes to device design and production (21 CFR 820.30 and 21 CFR 820.70) and document changes and approvals in the device master record (21 CFR 820.181).

Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting of medical device-related adverse events) (21 CFR Part 803) for devices or postmarketing safety reporting (21 CFR Part 4, Subpart B) for combination products (see https://www.fda.gov/combination-products/guidance-regulatory-information/postmarketing-safety-reportingcombination-products); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820) for devices or current good manufacturing practices (21 CFR Part 4, Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR Parts 1000-1050.

All medical devices, including Class I and unclassified devices and combination product device constituent parts are required to be in compliance with the final Unique Device Identification System rule ("UDI Rue"). The UDI Rule requires, among other things, that a device bear a unique device identifier (UDI) on its label and package (21 CFR 801.20(a)) unless an exception or alternative applies (21 CFR 801.20(b)) and that the dates on the device label be formatted in accordance with 21 CFR 801.18. The UDI Rule (21 CFR 830.300(a) and 830.320(b)) also requires that certain information be submitted to the Global Unique Device Identification Database (GUDID) (21 CFR Part 830 Subpart E). For additional information on these requirements, please see the UDI System webpage at https://www.fda.gov/medical-devices/device-advicecomprehensive-regulatory-assistance/unique-device-identification-system-udi-system.

Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to https://www.fda.gov/medical-device-safety/medical-device-reportingmdr-how-report-medical-device-problems.

For comprehensive regulatory information about mediation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/medicaldevices/device-advice-comprehensive-regulatory-assistance) and CDRH Learn (https://www.fda.gov/training-and-continuing-education/cdrh-learn). Additionally, you may contact the Division of Industry and Consumer Education (DICE) to ask a question about a specific regulatory topic. See the DICE website (https://www.fda.gov/medical-device-advice-comprehensive-regulatoryassistance/contact-us-division-industry-and-consumer-education-dice) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

{2}------------------------------------------------

Sincerely,

Image /page/2/Picture/3 description: The image shows the text "Patrick Antkowiak -S" to the right of the letters "FDA". The letters "FDA" are in a light blue color, while the name is in black. The letters "FDA" are much larger than the name.

for Jay Gupta Assistant Director DHT5A: Division of Neurosurgical, Neurointerventional, and Neurodiagnostic Devices OHT5: Office of Neurological and Physical Medicine Devices Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

{3}------------------------------------------------

Indications for Use

Submission Number (if known)

K240993

Device Name

encevis (2.1)

Indications for Use (Describe)

encevis is intended for the review, monitoring and analysis of EEG recordings made by electroencephalogram (EEG) devices using scalp electrodes and to aid neurologists in the assessment of EEG. This device is intended to be used by qualified medical practitioners who will exercise professional judgment in using the information.
The seizure detection component of encevis is intended to mark previously acquired sections of adult (greater than or equal to 18 years) EEG recordings that may correspond to electrographic seizures, in order to assist qualified clinical practitioners in the assessment of EEG traces. EEG recordings should be obtained with a full scalp montage according to the standard 10/20-system.
The spike detection component of encevis is intended to mark previously acquired sections of the patient's EEG recordings that may correspond to spikes, in order to assist qualified clinical practitioners in the assessment of EEG traces. The Spike detection component is intended to be used in adult patients greater than or equal to 18 years. encevis Spike detection performance has not been assessed for intracranial recordings.
encevis includes the calculation and display of a set of quantitative measures intended to monitor and analyze the EEG waveform. These include frequency bands, rhythmic and periodic patterns, burst suppression and spectrogram. These quantitative EEG measures should always be interpreted in conjunction with review of the original EEG waveforms.
The aEEG functionality included in encevis is intended to monitor the state of the brain.
encevis provides notifications on an on-screen display for seizure detection, electrographic status epilepticus detection, spike detection, quantitative EEG and aEEG that can be used when processing a record during acquisition (online) or based on stored EEG files (offline). Notifications can also be provided to external systems via the external interfaces to make them accessible to the user through the external system in a human-readable format. Delays of up to several minutes can occur between the beginning of a seizure, electrographic status epilepticus, the occurrence of a spike or detection of quantitative EEG features and when the encevis notifications will be shown to a user. encevis notifications cannot be used as a substitute for real time monitoring of the underlying EEG by a trained expert.
encevis PureEEG (Artifact Reduction) is intended to reduce EMG and electrode artifacts in a standard 10-20 EEG recording. PureEEG does not remove the entire artifact signal and is not effective for other types of artifacts. PureEEG may modify portions of waveforms representing cerebral activity. Waveforms must still be read by a qualified medical practitioner trained in recognizing artifact, and any interpretation or diagnosis must be made with reference to the original waveforms.
This device does not provide any diagnostic conclusion about the patient's condition to the user.

{4}------------------------------------------------

The encevis Component for Detection of Seizures and Electrographic Status Epilepticus is indicated for the detection of Seizures and Electrographic Status Epilepticus in patients greater than or equal to 18 years of age who are at risk for seizures. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis analyzes EEG waveforms and identifies patterns that may be consistent with seizures and electrographic status epilepticus as defined in the American Clinical Neurophysiology Society's Guideline 14. EEG recordings should be obtained with a full scalp montage according to the standard 10/20-system. The diagnostic output does also include a measure of seizure prevalence ("seizure burden") within a 10 minute (short-term seizure burden) and a 60 minute (hourly seizure burden) moving window. The output of the Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is intended to be used as a diagnostic output for determining patient treatment in acute-care environments. Detections from the Component for Detection of Seizures and Electrographic Status Epilepticus of encevis provide one input to the clinician that is intended to be used in conjunction with other elements of clinical practice to determine the appropriate treatment course for the patient. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is intended for detection of electrographic status epilepticus only. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis does not substitute for the review of the underlying EEG by a qualified clinician with respect to any other types of pathological EEG patterns. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is not intended for use in Epilepsy Monitoring Units.

Type of Use (Select one or both, as applicable)

Prescription Use (Part 21 CFR 801 Subpart D)

Over-The-Counter Use (21 CFR 801 Subpart C)

CONTINUE ON A SEPARATE PAGE IF NEEDED.

This section applies only to requirements of the Paperwork Reduction Act of 1995.

DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW!

The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:

Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov

"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."

{5}------------------------------------------------

510(k) summary encevis 2.1

1	Submission Sponsor and Application Correspondent	2
2	Date Prepared	2
3	Device Identification	2
4	Legally Marketed Predicate Devices	3
5	Device description	3
5.1	encevis EEG-viewer	3
5.2	Artefact reduction encevis PureEEG	3
5.3	Seizure detection of encevis NeuroTrend	3
5.4	Detection of seizures and status epilepticus of encevis acute care	4
5.5	Spike detection encevis EpiSpike	4
5.6	Pattern detection and aEEG	4
5.7	Spectrogram	4
5.8	External Interface „encevis AITInterface“	4
5.9	External Interface „encevis SeizureICUInterface“	5
6	Indication for Use Statement	5
7	Substantial Equivalence Discussion	6
7.1	Comparison to Primary and Secondary Predicate Device	6
7.2	Comparison to Tertiary and Quaternary Predicate Devices	11
7.3	Comparison of Intended Use and Technological Characteristics with the Predicate Devices	16
7.3.1	Indications for Use Comparison	16
7.3.1.1	General Use	16
7.3.1.2	Seizure detection component	16
7.3.1.3	Spike detection component	17
7.3.1.4	Calculation of quantitative measures	17
7.3.1.5	Calculation of aEEG	17
7.3.1.6	Notifications on an on-screen display	17
7.3.1.7	Artifact Reduction	18
7.3.1.8	Diagnostic output	18
7.3.1.9	Component for Detection of Seizures and Electrographic Status Epilepticus	18
7.3.2	Technological Characteristics Comparison	19
7.3.2.1	User	19
7.3.2.2	Use enivronment	19
7.3.2.3 Input Data	19
7.3.2.4 Patient registration method	20
7.3.2.5 Output	20
7.3.2.6 Mode of operation	20
8 Non-Clinical performance Data	20
9 Clinical Performance Data	21
9.1 Seizure detection performance testing	21
9.2 Detection of seizure and status epilepticus for acute care performance testing	22
9.3 encevis spike detection performance testing	27
9.4 encevis artifact reduction performance testing	28
9.5 encevis rhythmic and periodic patterns performance testing	32
9.6 encevis aEEG performance testing	34
9.7 encevis frequency bands performance testing	35
9.8 encevis burst suppression performance testing	35
10 encevis spectrogram performance testing	36
11 Statement of Substantial Equivalence	37

{6}------------------------------------------------

1 Submission Sponsor and Application Correspondent

AIT Austrian Institute of Technology GmbH Giefinggasse 4 1210 Vienna – Austria Phone: +43 50550-4203 +43 50550-4125 Fax: eMail: tilmann.kluge@ait.ac.at

2 Date Prepared

August 26, 2024

3 Device Identification

Trade/Proprietary Name:	encevis
Common Name:	Electroencephalograph
Classification Regulation:	21CFR882.1400
Product Code:	OMB, OLT, OMA
Class:	II
Panel:	Neurology

{7}------------------------------------------------

Primary Predicate:	K211452	encevis 1.12 Review and Analysis Software
AdditionalPredicate:	K132306	Persyst 12 EEG Review and Analysis Software
AdditionalPredicate:	K223504	Ceribell Status Epilepticus Monitor
AdditionalPredicate:	K191301	Ceribell Pocket EEG Device

4 Legally Marketed Predicate Devices

5 Device description

5.1 encevis EEG-viewer

The encevis EEG-viewer is intended for the review and the analysis of EEG-recordings that were recorded with an electroencephalography device using scalp electrodes. It shall aid the user in the examination of EEG-recordings. This includes the frequency filtering of the data, the scaling of the data in x and y direction, display of video together with the EEG data, and the visualization in different montages. In addition, the encevis EEG-viewer can also start modules that automatically analyze the EEG and present the results in form of notifications or in the form of modified EEG-curves. All included modules are intended for the user in the examination and monitoring of EEGrecordings.

5.2 Artefact reduction encevis PureEEG

The artefact reduction encevis PureEEG is an analysis module that automatically recognizes and reduces in the EEG-data that come from EMG and electrode artefacts. This modality is available within EEG viewer or via external software interface.

5.3 Seizure detection of encevis NeuroTrend

The seizure detection of encevis NeuroTrend is a module for the automatic marking of areas in the EEG that could correspond to epileptic seizures with electrographic correlate. The seizure detection of encevis NeuroTrend makes the results available to the user. This analysis can take place during the recording ("online"), which is recommended due to minimal notifiation delays. Alternatively the analysis can be done post-hoc with previoulsy stored EEG files ("offline"). To minimize notification delay for detected seizures, offline analyses should be done in regular, short time intervals.

{8}------------------------------------------------

5.4 Detection of seizures and status epilepticus of encevis acute care

Detection of seizures and electrographic status epilepticus of encevis acute care is a module for the automatic marking of areas in the EEG recorded in acute care patients that could correspond to epileptic seizures with electrographic correlate. It also identifies patterns that may be consistent with electrographic status epilepticus. The detection of seizures and status epilepticus of encevis acute care makes the results available to the user in form of seizure notifications and seizure burden over time. encevis recommends to perform this analysis during the recording ("online") to minimize notifiation delays by using an external EEG system that support the external interfaces for online processing. Alternatively the analysis can be done post-hoc with previously stored EEG files ("offline"). To minimize notification delay for detected seizures or detected electrographic status epilepticus, offline analyses should be done in regular, short time intervals.

5.5 Spike detection encevis EpiSpike

The spike detection encevis EpiSpike is a module for the automatic marking of areas in the EEG that could correspond to spikes. A graphical user interface presents the results to the user. This analysis can take place during the recording ("online") or post-hoc with previoulsy stored EEG files ("offline").

5.6 Pattern detection and aEEG

The encevis pattern detection of encevis NeuroTrend and encevis acute care automatically detects EEGpatterns defined in the Standardized Critical Care EEG Terminology of the American Clinical Neurophysiology Society and graphically presents the results to the user. Additionally, it detects and visualizes rhythmic patterns with frequencies of up to 12Hz and Burst-Suppression. It serves as a support during the examination of EEG-recordings in the EMU and acute care. This post-hoc analysis can take place in parallel to the recording or after the recording finished. In addition, encevis NeuroTrend and encevis acute care calculate and visualize continuous measures that describe the EEG. This includes the analysis of the frequency distribution and the aEEG. This analysis can take place during the recording ("online") or post-hoc with previoulsy stored EEG files ("offline").

5.7 Spectrogram

The encevis spectrogram graphically provides the user with a spectrogram for all or a selected number of EEG channels within a defined time range.

5.8 External Interface "encevis AlTInterface"

The external interface "encevis AlTInterface" enables the control of encevis by a software system of an EEG manufacturer. The interface allows to start the modules, to transmit EEG data to the modules, and to receive results at the calling software. This interface allows manufacturers of EEG systems to directly integrate encevis into their recording software. encevis viewer, encevis artifact reduction, and encevis spectrogram can not be controlled by the external AlTInterface. If encevis is used via the "AlTInterfaceDLL" external interface and without encevis graphical user interface, results and notifications are made available via the "AITInterface. In this scenario, the device using the external interface must ensure that the results and notifications are accessible to the user without avoidable delay.

{9}------------------------------------------------

5.9 External Interface "encevis SeizureICUInterface"

The external interface "encevis SeizurelCUnterface" allows the "encevis detection of seizures " to be integrated into a software system of an EEG manufacturer. The interface allows to start the module "encevis detection of seizures and status epilepticus", to transmit EEG data to the module and to receive results at the calling software. This interfacae allows manufacturers of EEG systems to directly integrate the module "encevis detection of seizures and status epilepticus" into their recording software. If encevis Seizure/CUInterface" external interface and without encevis graphical user interface, results and notifications are made available via the "SeizurelCUnterface" external interface. In this scenario, the external interface must ensure that the results and notifications are accessible to the user without avoidable delay,

6 Indication for Use Statement

encevis is intended for the review, monitoring and analysis of EEG recordings made by electroencephalogram (EEG) devices using scalp electrodes and to aid neurologists in the assessment of EEG. This device is intended to be used by qualified medical practitioners who will exercise professional judgment in using the information.
The seizure detection component of encevis is intended to mark previously acquired sections of adult (greater than or equal to 18 years) EEG recordings that may correspond to electrographic seizures, in order to assist qualified clinical practitioners in the assessment of EEG traces. EEG recordings should be obtained with a full scalp montage according to the standard 10/20-system.
The spike detection component of encevis is intended to mark previously acquired sections of the patient's EEG recordings that may correspond to spikes, in order to assist qualified clinical practitioners in the assessment of EEG traces. The Spike detection component is intended to be used in adult patients greater than or equal to 18 years. encevis Spike detection performance has not been assessed for intracranial recordings.
encevis includes the calculation and display of a set of quantitative measures intended to monitor and analyze the EEG waveform. These include frequency bands, rhythmic and periodic patterns, burst suppression and spectrogram. These quantitative EEG measures should always be interpreted in conjunction with review of the original EEG waveforms.
The aEEG functionality included in encevis is intended to monitor the state of the brain.
encevis provides notifications on an on-screen display for seizure detection, electrographic status epilepticus detection, spike detection, quantitative EEG and aEEG that can be used when processing a record during acquisition (online) or based on stored EEG files (offline). Notifications can also be provided to external interfaces to make them accessible to the user through the external system in a human-readable format. Delays of up to several minutes can occur between the beginning of a seizure, electrographic status epilepticus, the occurrence of a spike or detection of quantitative EEG features and when the encevis notifications will be shown to a user. encevis notifications cannot be used as a substitute for real time monitoring EEG by a trained expert.
encevis PureEG (Artifact Reduction) is intended to reduce EMG and electrode artifacts in a standard 10-20 EEG recording. PureEEG does not remove the entire artifact signal and is not effective for other types of artifacts. PureEEG may modify portions of waveforms representing cerebral activity. Waveforms must still be read by a qualified medical practitioner trained in recognizing artifact, and any interpretation or diagnosis must be made with reference to the original waveforms.
This device does not provide any diagnostic conclusion about the patient's condition to the user.
The encevis Component for Detection of Seizures and Electrographic Status Epileptious is indicated for the detection of Seizures and Electrographic Status Epilepticus in patients greater than or equal to 18 years of age who are at risk for seizures. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis

{10}------------------------------------------------

analyzes EEG waveforms and identifies patterns that may be consistent with seizures and electrographic status epiledicus as defined in the American Clinical Neurophysiology Society's Guideline 14. EEG recordings should be obtained with a full scalp montage according to the standard 10/20-system. The diagnostic output does also include a measure of seizure prevalence ("seizure burden") within a 10 minute (short-term seizure burden) and a 60 minute (hourly seizure burden) moving window.

The output of the Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is intended to be used as a diagnostic output for determining patient treatment in acute-care environments. Detections from the Component for Detection of Seizures and Electrographic Status Epilepticus of encevis provide one input to the clinician that is intended to be used in conjunction with other elements of clinical practice to determine the appropriate treatment course for the patient. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is intended for detection of electrographic status epilepticus only. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis does not substitute for the review of the underlying EEG by a qualified clinician with respect to any other types of pathological EEG patterns. The Component for Detection of Seizures and Electrographic Status Epilepticus of encevis is not intended for use in Epilepsy Monitoring Units.

7 Substantial Equivalence Discussion

The following table compares the encevis to the predicate device with respect to intended use, technological characteristics and principles of operation, providing more detailed information regarding the basis for the determination of substantial equivalence.

	Subject Device	Primary Predicate Device	Secondary Predicate Device
Device	encevis 2.1	encevis 1.12	Persyst 12EEG Review and Analysis Software
Device Identification
510k Reference	(subject device)	K211452	K132306
Product Code	OMB	OMB	OMB
Additional Codes	OLT, OMA	OLT, OMA	OLT, OMA
Class	II	II	II
RegulationNumber	21CFR882.1400	21CFR882.1400	21CFR882.1400
Regulation Name	Electroencephalograph	Electroencephalograph	Electroencephalograph
Manufacturer	AIT Austrian Institute ofTechnology GmbH	AIT Austrian Institute of TechnologyGmbH	Persyst Development Corporation
Device Description and Identification
General DeviceDescription	EEG Review and AnalysisSoftware	EEG Review and Analysis Software	EEG Review and Analysis Software
Indication forUseGeneral use	1. encevis is intended for thereview, monitoring and analysis ofEEG recordings made byelectroencephalogram (EEG)devices using scalp electrodes andto aid neurologists in theassessment of EEG. This device isintended to be used by qualifiedmedical practitioners who willexercise professional judgment inusing the information.	1. encevis is intended for thereview, monitoring and analysis ofEEG recordings made byelectroencephalogram (EEG)devices using scalp electrodes andto aid neurologists in theassessment of EEG. This device isintended to be used by qualifiedmedical practitioners who willexercise professional judgment inusing the information.	1. Persyst 12 EEG Review andAnalysis Software is intended for thereview, monitoring and analysis ofEEG recordings made byelectroencephalogram (EEG)devices using scalp electrodes andto aid neurologists in the assessmentof EEG. This device is intended to beused by qualified medicalpractitioners who will exerciseprofessional judgment in using theinformation.
Indication forUse: Seizuredetection	2. The seizure detectioncomponent of encevis is intendedto mark previously acquiredsections of adult (greater than orequal to 18 years) EEG recordingsthat may correspond toelectrographic seizures, in order toassist qualified clinical practitionersin the assessment of EEG traces.EEG recordings should be obtainedwith a full scalp montage accordingto the standard 10/20-system.	2. The seizure detection componentof encevis is intended to markpreviously acquired sections of adult(greater than or equal to 18 years)EEG recordings that maycorrespond to electrographicseizures, in order to assist qualifiedclinical practitioners in theassessment of EEG traces. EEGrecordings should be obtained witha full scalp montage according tothe standard 10/20-system.	2. The Seizure Detection componentof Persyst 12 is intended to markpreviously acquired sections of adult(greater than or equal to 18 years)EEG recordings that may correspondto electrographic seizures, in order toassist qualified clinical practitionersin the assessment of EEG traces.EEG recordings should be obtainedwith a full scalp montage accordingto the standard 10/20 system.
Indication forUse: Spikedetection	3. The spike detection componentof encevis is intended to markpreviously acquired sections of thepatient's EEG recordings that maycorrespond to spikes, in order toassist qualified clinical practitionersin the assessment of EEG traces.The Spike detection component isintended to be used in adultpatients greater than or equal to 18years. encevis Spike detectionperformance has not beenassessed for intracranialrecordings.	3. The spike detection component ofencevis is intended to markpreviously acquired sections of thepatient's EEG recordings that maycorrespond to spikes, in order toassist qualified clinical practitionersin the assessment of EEG traces.The Spike Detection component isintended to be used in adult patientsgreater than or equal to 18 years.encevis Spike Detectionperformance has not beenassessed for intracranial recordings.	3. The Spike Detection component ofPersyst 12 is intended to markpreviously acquired sections of thepatient's EEG recordings that maycorrespond to spikes, in order toassist qualified clinical practitionersin the assessment of EEG traces.The Spike Detection component isintended to be used in patients atleast one month old. Persyst 12Spike Detection performance has notbeen assessed for intracranialrecordings.
Indication forUse: Quantitativemeasures	4. encevis includes the calculationand display of a set of quantitativemeasures intended to monitor andanalyze the EEG waveform. Theseinclude frequency bands, rhythmicand periodic patterns, burstsuppression and spectrogram.These quantitative EEG measuresshould always be interpreted inconjunction with review of theoriginal EEG waveforms.	4. Persyst 12 includes the calculationand display of a set of quantitativemeasures intended to monitor andanalyze the EEG waveform. Theseinclude FFT, Rhythmicity, PeakEnvelope, Artifact Intensity,Amplitude, Relative Symmetry andSuppression Ratio. Automatic eventmarking is not applicable to thequantitative measures. Thesequantitative EEG measures shouldalways be interpreted in conjunctionwith review of the original EEGwaveforms.
Indication forUse: aEEG	5. The aEEG functionality includedin encevis is intended to monitorthe state of the brain.	5. The aEEG functionality included inPersyst 12 is intended to monitor thestate of the brain. The automatedevent marking function of Persyst 12is not applicable to aEEG.
Indication forUse: Notifications	6. encevis provides notifications onan on-screen display for seizuredetection, electrographic statusepilepticus detection, spikedetection, quantitative EEG andaEEG that can be used whenprocessing a record duringacquisition (online) or based onstored EEG files (offline).Notifications can also be providedto external systems via the externalinterfaces to make them accessibleto the user through the externalsystem in a human-readableformat.Delays of up to several minutescan occur between the beginning ofa seizure, electrographic statusepilepticus, the occurrence of aspike or detection of quantitativeEEG features and when theencevis notifications will be shownto a user. encevis notificationscannot be used as a substitute forreal time monitoring of theunderlying EEG by a trainedexpert.	6. Persyst 12 provides notificationsfor seizure detection, quantitativeEEG and aEEG that can be usedwhen processing a record duringacquisition. These include an onscreen display and the optionalsending of an email message.Delays of up to several minutes canoccur between the beginning of aseizure and when the Persyst 12notifications will be shown to a user.Persyst 12 notifications cannot beused as a substitute for real timemonitoring of the underlying EEG bya trained expert.
Indication forUse: Artifactreduction	7. encevis PureEEG (ArtifactReduction) is intended to reduceEMG and electrode artifacts in astandard 10-20 EEG recording.PureEEG does not remove theentire artifact signal and is noteffective for other types of artifacts.PureEEG may modify portions ofwaveforms representing cerebralactivity. Waveforms must still beread by a qualified medicalpractitioner trained in recognizingartifact, and any interpretation ordiagnosis must be made withreference to the originalwaveforms.	7. encevis PureEEG (ArtifactReduction) is intended to reduceEMG, eye movement, and electrodeartifacts in a standard 10-20 EEGrecording. PureEEG does notremove the entire artifact signal,and is not effective for other types ofartifacts. PureEEG may modifyportions of waveforms representingcerebral activity. Waveforms muststill be read by a qualified medicalpractitioner trained in recognizingartifact, and any interpretation ordiagnosis must be made withreference to the original waveforms.	7. Persyst AR (Artifact Reduction) isintended to reduce EMG, eyemovement, and electrode artifacts ina standard 10-20 EEG recording. ARdoes not remove the entire artifactsignal, and is not effective for othertypes of artifacts. AR may modifyportions of waveforms representingcerebral activity. Waveforms muststill be read by a qualified medicalpractitioner trained in recognizingartifact, and any interpretation ordiagnosis must be made withreference to the original waveforms.
Indication forUse: Diagnosticoutput	8. This device does not provide anydiagnostic conclusion about thepatient's condition to the user.	8. This device does not provide anydiagnostic conclusion about thepatient's condition to the user.	8. This device does not provide anydiagnostic conclusion about thepatient's condition to the user.
User	This device is intended to be usedby qualified medical practitionersonly who will exercise professionaljudgment in using the information.	This device is intended to be usedby qualified medical practitionersonly who will exercise professionaljudgment in using the information.	This device is intended to be used byqualified medical practitioners whowill exercise professional judgmentin using the information.
Use environment	Any professional healthcare facilityused by medical professionals inan appropriate environment.	Any professional healthcare facilityused by medical professionals in anappropriate environment.	-
Technology
Input Data	Display and calculation is based onEEG data recorded by externalEEG systems. They are either readfrom the EEG-file provided by theEEG system (offline mode) or canbe streamed to encevis using theinterfaces provided by AIT"AITInterfaceDLL" and/or"SeizureICUInterface" (onlinemode).	Display and calculation is based onEEG data recorded by external EEGsystems. They are either read fromthe EEG-file provided by the EEGsystem or can be send to encevisusing the interface provided by AIT(AITInterfaceDLL)	Display and calculation is based onEEG data recorded by external EEGsystems. They are read from theEEG-file provided by the EEGsystem
Compliance	No standard data format availablein the industry	No standard data format available inthe industry	No standard data format available inthe industry
PatientRegistrationmethod	encevis provides a database-basedpatient management system whereusers register a patient by a patientlabel, last name, first name andbirthdate. Results are stored in thisdatabase. There is no connectionto KIS system or HL7interface. Online mode using theexternal interfaces AITInterfaceDLLor SeizureICUInterface does nothandle PHI, the programmaticinterfaces do not use patientidentification.	encevis provides a database-basedpatient management system whereusers register a patient by a patientlabel, last name, first name andbirthdate. Results are stored in thisdatabase. There is no connection toKIS system or HL7 interface. Onlinemode using the external interfaceAITInterfaceDLL does not handlePHI, the programmatic interfaces donot use patient identification.	Results are stored in additional filesin the file system placed in the samefolder as the EEG file.No patient management available
Outputs	encevis can be used with its built-ingraphical user interfaces,which present markers in a list andgraphical plots showing resultsover time. The built-in displays canservice as a near-real time display(online) or show results from post-hoc analyses (offline). To use thebuilt-in displays, results must bestored in a database.The outputs and notifications arealso available in near-real time viaprogrammatical external interfaces"AITInterfaceDLL"or "SeizureICUInterface" to makethem accessible to theuser through the external systemin a human-readable format.Delays of up to several minutescan occur between events andwhen the encevis notifications willbe shown to a user.	encevis can be used with its built-ingraphical user interfaces,which present markers in a list andgraphical plots showing results overtime. The built-in displays canservice as a near-real time display(online) or show results from post-hoc analyses (offline). To use thebuilt-in displays, results must bestored in a database.The outputs and notifications arealso available in near-real time viaprogrammatical external interface"AITInterfaceDLL" to make themaccessible to the user through theexternal system in a human-readable format.Delays of up to several minutes canoccur between events and when theencevis notifications will be shownto a user.	Persyst can be used as a near-realtime display, when it is used duringacquisition with a medical EEGdevice. Delays of up to severalminutes can occur between eventsand when the Persyst notificationswill be shown to a user.Results are stored in additional filesin the file system placed in the samefolder as the EEG file. User output isgiven by graphical user interfaces
Mode ofoperation	encevis can be used a) in "onlinemode" through a direct connectionbetween the EEG recording systemand encevis via the"AITInterfaceDLL"or "SeizureICUInterface" externalinterface or encevis onlineinterfaces, or b) in "offline mode" byanalyzing stored EEG files.	encevis can be used a) in "onlinemode" through a direct connectionbetween the EEG recording systemand encevis via the"AITInterfaceDLL" external interfaceor encevis online interfaces, or b) in"offline mode" by analyzing storedEEG files.	Persyst 12 is used for bothmonitoring and trending of EEGrecordings during acquisition andreviewing of processed recordings.
CalibrationMethod	No calibration necessary	No calibration necessary	No calibration necessary
CompatibleEquipment andSoftware	Encevis can read and process EEGdata from several EEG vendors. Alist of compatible EEG systems canbe found onhttps://www.encevis.com/support/dataformats/	Encevis can read and process EEGdata from several EEG vendors. Alist of compatible EEG systems canbe found onhttps://www.encevis.com/support/dataformats/	Persyst can read and process EEGdata from several EEG vendors. Alist of compatible EEG systems canbe found on http://www.persyst.com

7.1 Comparison to Primary and Secondary Predicate Device

{11}------------------------------------------------

{12}------------------------------------------------

{13}------------------------------------------------

{14}------------------------------------------------

{15}------------------------------------------------

7.2 Comparison to Tertiary and Quaternary Predicate Devices

	Subject Device	Tertiary Predicate Device	Quaternary Predicate Device
Device	encevis 2.1	Ceribell Pocket EEG Device	Ceribell Status Epilepticus Monitor
Device Identification
510k Reference	(subject device)	K191301	K223504
Product Code	OMB	OMB	OMB
Additional Codes	OLT, OMA	OMC, GWQ	OMC, GWQ
Class	II	II	II
RegulationNumber	21CFR882.1400	21CFR882.1400	21CFR882.1400
RegulationName	Electroencephalograph	Electroencephalograph	Electroencephalograph
Manufacturer	AIT Austrian Institute of TechnologyGmbH	Ceribell, Inc.	Ceribell, Inc.
	Device Description and Identification
General DeviceDescription	EEG Review and Analysis Software	EEG Device and AnalysisSoftware	EEG Analysis Software
	9. The encevis Component forDetection of Seizures andElectrographic Status Epilepticus isindicated for the detectionof Seizures and ElectrographicStatus Epilepticus in patientsgreater than or equal to 18 years ofage who are at risk for seizures.The Component for Detection ofSeizures and Electrographic StatusEpilepticus of encevis analyzesEEG waveforms and identifiespatterns that may be consistentwith seizures and electrographicstatus epilepticus as defined in theAmerican Clinical NeurophysiologySociety's Guideline 14. EEG		The Ceribell Status EpilepticusMonitor software is indicated for thediagnosis of Electrographic StatusEpilepticus in patients greater than orequal to 18 years of age who are at
	recordings should be obtained witha full scalp montage according tothe standard 10/20-system. Thediagnostic output does also includea measure of seizure prevalence("seizure burden") within a 10minute (short-term seizure burden)and a 60 minute (hourly seizureburden) moving window.The output of the Component forDetection of Seizures andElectrographic Status Epilepticus ofencevis is intended to be used as adiagnostic output for determiningpatient treatment in acute-careenvironments. Detections from theComponent for Detection ofSeizures and Electrographic StatusEpilepticus of encevis provide oneinput to the clinician that is intendedto be used in conjunction with otherelements of clinical practice todetermine the appropriatetreatment course for the patient.The Component for Detection ofSeizures and Electrographic StatusEpilepticus of encevis is intendedfor detection of electrographicstatus epilepticus only. TheComponent for Detection ofSeizures and Electrographic StatusEpilepticus of encevis does notsubstitute for the review of theunderlying EEG by a qualifiedclinician with respect to any othertypes of pathological EEG patterns.The Component for Detection ofSeizures and Electrographic StatusEpilepticus of encevis is notintended for use in EpilepsyMonitoring Units
Indication forUse:Statusepilepticusdetection /seizure burden
Indication forUse:Notifications	6. encevis provides notifications onan on-screen display for seizuredetection, electrographic statusepilepticus detection, spikedetection, quantitative EEG andaEEG that can be used whenprocessing a record duringacquisition (online) or based onstored EEG files (offline).Notifications can also be providedto external systems via the externalinterfaces to make them accessibleto the user through the externalsystem in a human-readableformat.Delays of up to several minutes canoccur between the beginning of aseizure, electrographic statusepilepticus, the occurrence of aspike or detection of quantitativeEEG features and when theencevis notifications will be shownto a user. encevis notificationscannot be used as a substitute forreal time monitoring of theunderlying EEG by a trained expert.	Notifications include an on-screendisplay on the Pocket EEG Deviceand the optional sending of an e-mail message to a clinician. Delaysof up to several minutes can occurbetween the beginning of a seizureand when the Seizure Detectionnotifications will be shown to a user.
Indication forUse:Diagnosticoutput	8. This device does not provide anydiagnostic conclusion about thepatient's condition to the userThe output of the Component forDetection of Seizures andElectrographic Status Epilepticus ofencevis is intended to be used as adiagnostic output for determiningpatient treatment in acute-careenvironments.Detections from the Component forDetection of Seizures andElectrographic Status Epilepticus ofencevis provide one input to theclinician that is intended to be usedin conjunction with other elementsof clinical practice to determine theappropriate treatment course forthe patient.The Component for Detection ofSeizures and Electrographic StatusEpilepticus of encevis does notsubstitute for the review of theunderlying EEG by a qualifiedclinician with respect to any othertypes of pathological EEGpatterns.	The Pocket EEG Device does notprovide any diagnostic conclusionabout the subject's condition andSeizure Detection notificationscannot be used as a substitute forreal time monitoring of theunderlying EEG by a trained expert.	The diagnostic output of the CeribellStatus Epilepticus Monitor is intendedto be used as an aid for determiningpatient treatment in acute-careenvironments.The device's diagnosis ofElectrographic Status Epilepticusprovides one input to the clinician thatis intended to be used in conjunctionwith other elements of clinical practiceto determine the appropriatetreatment course for the patient.The device does not substitute for thereview of the underlying EEG by aqualified clinician with respect to anyother types of pathological EEGpatterns.
User	This device is intended to be usedby qualified medical practitionersonly who will exercise professionaljudgment in using the information.	The visual and audible signals assisttrained medical staff to makeneurological diagnoses.	The device's diagnosis ofElectrographic Status Epilepticusprovides one input to the clinician.
Use environment	The device is intended to be usedin a professional healthcare facilityby medical professionals in anappropriate environment. Thecomponent for detection of seizuresand electrographic statusepilepticus is intend to be used inthe acute care environments.	The Pocket EEG Device is intendedto be used in a professionalhealthcare facility environment.	The diagnostic output of the CeribellStatus Epilepticus Monitor is intendedto be used as an aid for determiningpatient treatment in acute-careenvironments.
Input Data	Display and calculation is based onEEG data recorded by externalEEG systems. They are either readfrom the EEG-file provided by theEEG system or can be send toencevis using the interfacesprovided by AIT (AITInterfaceDLLand/or SeizureICUInterface)	Display and calculation is based onEEG data recorded by the deviceitself.	The device software is intended foruse only with the Ceribell Pocket EEGDevice (K191301)
Compliance	No standard data format availablein the industry	No standard data format available inthe industry	No standard data format available inthe industry
PatientRegistrationmethod	encevis provides a database-basedpatient management system, whereusers register a patient by a patientlabel, last name, first name andbirthdate. Results are stored in thisdatabase. There is no connectionto KIS system or HL7 interface.Online mode using the externalinterfaces AITInterfaceDLLor SeizurelCUInterface does nothandle PHI, the programmaticinterfaces do not use patientidentification.	unknown	unknown
Outputs	encevis can be used with its built-ingraphical user interfaces,which present markers in a list andgraphical plots showing results overtime. The built-in displays canservice as a near-real time display(online) or show results from post-hoc analyses (offline). To use thebuilt-in displays, results must bestored in a database.The outputs and notifications arealso available in near-real time viaprogrammatical external interfaces"AlTInterfaceDLL"or "SeizureICUInterface" to makethem accessible to the user throughthe external system in a human-readable format.Delays of up to several minutes canoccur between events and whenthe encevis notifications will beshown to a user.	The Ceribell Pocket EEG Device isintended to record and store EEGsignals, and to present the EEGsignals in visual and audible formatsin real time.Delays of up to several minutes canoccur between the beginning of aseizure and whenthe Seizure Detection notificationswill be shown to a user.Notification in form of on-screendisplay and optional sending ofemail messages.	The device is intended to be used withthe Pocket EEG Device to providefeedback for electrographic statusepilepticus during acquisition.Notifications are given via CeribellPocket EEG device.
Mode ofoperation	encevis can be used a) in "onlinemode" through a direct connectionbetween the EEG recording systemand encevis via the"AITInterfaceDLL"or "SeizureICUInterface" externalinterface or encevis onlineinterfaces, or b) in "offline mode" byanalyzing previously stored EEGfiles.	The Ceribell Pocket EEG Device isintended to record and store EEGsignals, and to present the EEGsignals in visual and audible formatsin real time.	The device is intended to be used withthe Ceribell Pocket EEG Device toprovide feedback for electrographicstatus epilepticus during acquisition.
CalibrationMethod	No calibration necessary	No calibration necessary	No calibration necessary
CompatibleEquipment andSoftware	Encevis can read and process EEGdata from several EEG vendors. Alist of compatible EEG systems canbe found onhttp://www.encevis.com	Stand-alone EEG device	In combination with Ceribell PocketEEG Device only

{16}------------------------------------------------

ditionally, the EEG Recording wer Software component of the cket EEG Device incorporates a zure Detection component that is ended to mark previously uired sections of EEG recordings atients greater than or equal to years of age that may correspond lectrographic seizures in order to sist qualified clinical practitioners he assessment of EEG traces. e Seizure Detection component vides notifications to the user en detected seizure prevalence is equent," "Abundant," or ntinuous," per the definitions of American Clinical urophysiology Society Guideline

risk for seizure. The Ceribell Status Epilepticus Monitor software analyzes EEG waveforms and identifies patterns that may be consistent with electrographic status epilepticus as defined in the American Clinical Neurophysiology Society's Guideline 14.

The diagnostic output of the Ceribell Status Epilepticus Monitor is intended to be used as an aid for determining patient treatment in acute-care environments. The device's diagnosis of Electrographic Status Epilepticus provides one input to the clinician that is intended to be used in conjunction with other elements of clinical practice to determine the appropriate treatment course for the patient. The Ceribell Status Epilepticus Monitor is intended for diagnosis of Electrographic Status Epilepticus only. The device does not substitute for the review of the underlying EEG by a qualified clinician with respect to any other types of pathological EEG patterns. The device is not intended for use in Epilepsy Monitoring Units.

{17}------------------------------------------------

{18}------------------------------------------------

{19}------------------------------------------------

{20}------------------------------------------------

7.3 Comparison of Intended Use and Technological Characteristics with the Predicate Devices

The subject device is substantially equivalent to the predicate devices in terms of operating principles and design. They are intended to be used by qualified medical practitioners only who will exercise professional judgment in using the information.

7.3.1 Indications for Use Comparison

7.3.1.1 General Use

Similarities

The general indications for use are identical for the subject devices. All devices are intended for the review, the monitoring, and analysis of EEG recordings using scalp electrodes. This includes the frequency filtering of the data, the scaling of the data in x and y direction and the visualization in different montages. The display of video together with the EEG data is substantial equivalent to the secondary predicate device. The systems can start additional analysis of the EEG and present the results to the user. They aid neurologists in the assessment of EEG and are intended to be used by qualified medical practitioners only.

Differences

There are no differences between the devices.

7.3.1.2 Seizure detection component

Similarities

The subject device and primary predicate device have both a seizure detection with identical intended use and operation principles.

Differences

The seizure detection algorithm of the subject device was shown to yield higher sensitivity compared to the secondary predicate device, but also higher false positive rates. Compared to its predicate

{21}------------------------------------------------

device (encevis 1.12), the seizure detection algorithm of the subject device (encevis 2.1) combines detections of the algorithm in encevis 1.12 and an additional Al-model to achieve high sensitivity. This leads to a higher overall detection sensitivity but also to a higher false positive rate compared to encevis 1.12.

7.3.1.3 Spike detection component

Similarities

The subject device, the primary- and the secondary predicate devices are intended for the detection of spikes in order to assist qualified clinical practitioners in the assessment of EEG traces. In both the subject device and primary predicate device, the design of these algorithms are the same.

Differences

The subject device and the primary predicate device restrict the use to adult patients greater than or equal to 18 years while the secondary predicate device allows the use for patients at least one month of age. This raises no new concerns as the subject device is more restrictive in the patient population.

7.3.1.4 Calculation of quantitative measures

Similarities

The subject device and the primary predicate device both have frequency bands, rhythmic and burst suppression as quantitative measure. In both devices these algorithms are designed equally.

Differences

The subject device and the secondary predicate device both include quantitative "spectrogram" measures, while the primary device does not. While the secondary predicate device uses an FFT spectrogram, the subject device uses a wavelet spectrogram, which can show higher frequency components with increased temporal resolution. Both devices are intended to show a time-frequency representation of the EEG to the user and are compared in a clinical evaluation. There is no significant difference between the subject device and the secondary predicate device that could raise new concerns.

7.3.1.5 Calculation of aEEG

Similarities

The indications for use are identical. In both devices the design of the algorithm is the same.

Differences

There are no differences between the devices

7.3.1.6 Notifications on an on-screen display

Similarities

Notifications for seizure detection, quantitative EEG (frequency bands, rhythmic and periodic patterns, burst suppression) and aEEG are very similar for the subject device and for the primary and secondary predicate devices.

If the subject device is used in "online mode" with a direct connection to a medical EEG device via external interfaces, notifications for seizure detection, frequency bands, rhythmicheriodic patterns, burst suppression, and aEEG are very similar notification delays can be expected with the subject device, the tertiary predicate device, or with the primary and secondary predicate devices, when being used online (during acquisition). If furthermore encevis' graphical user interface is employed, notifications are displayed on-screen, similar to the functionality of the tertiary predicate device.

{22}------------------------------------------------

Differences

Only the tertiary predicate device provides optional email notifications. However, if implemented by an external system, encevis 2.1 can provide notifications to external systems via the external interfaces to make them accessible to the user through the external system in a human-readable format.

7.3.1.7 Artifact Reduction

Similarities

The indications for use are identical in both devices the design of the algorithm is the same.

Differences

There are no differences between the devices

7.3.1.8 Diagnostic output

Similarities

The subject device and all four predicate devices provide diagnostic outputs, but not any diagnostic conclusions. For the detection of electrographic status epilepticus, the outputs of both, the subject device encevis 2.1 and the quaternary predicate device Ceribell Status Epilepticus Monitor are intended to be used as a diagnostic output for determining patient treatment and provide one input to the clinician that is intended to be used in conjunction with other elements of clinical practice and review of the underlying EEG.

Differences

No differences.

7.3.1.9 Component for Detection of Seizures and Electrographic Status Epilepticus

Similarities

The subject device and the tertiary predicate device (Ceribell Pocket EEG Device) share intended use for detecting electrographic seizures and measuring seizure burden", indicating the prevalence of seizures within moving time windows) of patients in acute care environments who are 18 years of age or older.

The subject device and the quaternary predicate device (Ceribell Status Epilepticus Monitor Software) share the same intended use for detecting and providing diagnostic output for electrographic status epilepticus (ESE) of patients in acute care environments who are 18 years of age or older, as defined in the American Clinical Neurophysiology Society's Guideline 14, as an aid for determining patient treatment in acute-care environments. The outputs of both devices are intended to be used in conjunction with other elements of clinical practice to determine the appropriate treatment course for the patient.

The subject device, the tertiary predicate device (Ceribell Pocket EEG Device) and the quaternary predicate device (Ceribell Status Epilepticus Monitor Software) can be used online, analyzing EEG data in near-real time in parallel to the recording.

Differences

The subject device uses a full 10-20 montage for the detection of seizures and ESE, whereas the tertiary predicate device as well as the quaternary predicate device use reduced electrode montages with 10 electrodes. This raises no new concerns as the 10-20 system is the most widely used electrode configuration and is state of the art in EEG recordings.

The subject device includes a quantitative measure "seizure burden", which measures seizure within moving time windows of either 10 minutes ("short-time seizure burden") or 60 minutes ("hourly seizure burden"). The

{23}------------------------------------------------

tertiary predicate device (Ceribell Pocket EEG Device) has a similar seizure burden display, which however is based on a 5-minute moving window and provides notifications for seizure burden exceeding three different threshold levels 10% ("Frequent"), 50% ("Abundant"), and 90% ("Continuous"). The subject device in contrast creates detections/notifications of electrographic status epilepticus, if the seizure burden level exceeds thresholds corresponding to continuous seizures of duration >10 minutes, or if the hourly seizure burden level exceeds thresholds corresponding to more than 12 minutes of seizure activity within one hour. This corresponds to the definitions established by the ACNS in their Guideline 14, which state that "ESE is defined as an electrographic seizure for > 10 continuous minutes or for a total duration of >20% of any 60-minute period of recording." By deviating from the time windows chosen by the tertiary predicate device, we more closely adhere to the ACNS standard while maintaining the same design as Ceribell to provide on-screen notifications for different seizure burden levels. While encevis output provides a higher granularity by expressing seizure burden in percentage, the output also corresponds to three different threshold levels 10% ("Abundant"), and 90% ("Continuous") per the definitions of the American Clinical Neurophysiology Society Guideline 14. The seizure burden measures of both devices are suitable for measuring seizure prevalence and can be seen as substantially equivalent.

In contrast to the quaternary predicate device, in addition to the operation during EEG recording ("online mode") the subject device can also be used in an "offline mode", where previously recorded EEGs can be analyzed. Although online operation is preferable to provide notifications with low latency, this additional operational mode can enable detection of ESE in situations where existing EEG devices without an integrated detection of seizures or status epilepticus are in place.

In contrast to the tertiary predicate device (Ceribell Pocket EEG Device) and the quaternary predicate device (Ceribell Status Epilepticus Monitor Software), the subject device can also be used to detect ESE in offline mode, based on recorded and stored EEG files. Compared to online use, this may increase the time to treatment in case of detected seizures or ESE. Therefore, online use is clearly recommended in the instructions for use.

The offline use on the other hand operates similarly to offline seizure detection with the primary and secondary predicate devices (encevis 1.12 and Persyst 12). Therefore, compared to these predicate devices, the offline detection of ESE, in addition to seizure detection, enhances the benefit for this intended use by automatically assessing critical values for seizure prevalence, as defined for electrographic status epilepticus by the ANCS.

7.3.2 Technological Characteristics Comparison

7.3.2.1 User

The subject device and the four predicate devices are intended to be used by trained professionals.

7.3.2.2 Use enivronment

The general use environment of the subject device encevis 2.1, the primary predicate device encevis 1.12, and the secondary predicate device Persyst 12 are professional healthcare environments. The subject devices' Component for Detection of Seizures and Electrographic Status Epilepticus are intended for the use in acute-care environments.

7.3.2.3 Input Data

The subject device encevis 2.1. the primary predicate device encevis 1.12. and the secondary predicate device Persyst 12 rely on data recorded by medical EEG devices as input data. The software analyses of the tertiary predicate device Ceribell Pocket EEG and the quarternary predicate device Ceribell Status Epilepticus Monitor use EEG recorded by the Ceribell Pocket EEG device.

{24}------------------------------------------------

7.3.2.4 Patient registration method

Only the subject device encevis 2.1 and the primary predicate device encevis 1.12 provide a patient management system, where users register a patient label, last name, first name and birth date. Results are stored in this database.

7.3.2.5 Output

All compared devices can provide outputs and notifications via graphical user interfaces. The subject device encevis 2.1 and the primary predicate device encevis 1.12 can provide output to external systems via programmatical external interfaces. With all devices and all operational modes, there may be delays of up to several minutes in receiving notifications.

7.3.2.6 Mode of operation

The subject device encevis 2.1, the primary predicate device encevis 1.12, and the secondary predicate device Persyst 12 can either be used as online devices, analyzing data recorded by medical EEG devices in near-real time, or as offline devices, analyzing previously recorded EEG data. The analysis software of the tertiary predicate device Ceribell Pocket EEG and the quarternary predicate device Ceribell Status Epilepticus Monitor can be operated online during data acquisition with the Ceribel Pocket EEG device.

8 Non-Clinical performance Data

Software verification and validation testing was conducted, and documentation provided by the FDA Guidance for Industry and FDA Staff, Guidance for the Contained in Medical Devices. Traceability has been documented between all system specification test protocols. Verification and validation testing includes:

1. Code inspections
1. Unit level testing
1. Integration level testing
1. System level testing

In addition, tests according to "IEC 62366-1:2015, Medical Devices Part 1—Application of usability engineering to medical devices" have been performed.

Verification and validation activities established the safety and performance characteristics of the with respect to the predicate device. For bench tests, detection results of the modules were compared to annotations set by clinical EEG experts using large amount of EEG data from different centers. Where possible, the results of encevis were directly compared with the results of the predicate device. Suitable statistical measures like sensitivity and specificity were calculated.

The encevis (stand-alone software) meets all the stated requirements for overall design, performance, biocompatibility and electrical safety and passed all the testing noted above.

{25}------------------------------------------------

9 Clinical Performance Data

9.1 Seizure detection performance testing

For performance evaluation of the encevis seizure detection device we measured positive percentage agreement (detection sensitivity based on the reference standard) and negative disagreement rate (false detections per 24 hours based on reference standard) by comparing seizure detections to consensus annotations from three independent reviewers. Second, to define the acceptable performance level of the encevis seizure detection device we also measured positive percentage agreement and negative disagreement rate of the predicate device Persyst using the same study population and the same gold standard annotations. A statistical test is then used to show that the encevis seizure detection sensitivity is non-inferior to the performance of the predicate device.

Study population

We included scalp-EEG recordings of 55 subjects that underwent video-EEG monitoring in an epilepsy monitoring unit for the purpose of differential diagnosis or pre-surgical evaluation. All patients where 18 years of age or older. 50 patients where included that showed seizure events during recording and were diagnosed of having epilepsy. Further, we included the five subjects that were diagnosed of not having epilepsy (Subject-ID 30-34).

Reference Standard

To define the reference standard, a total of 1603 hours of EEG from these 55 subjects were presented to three independent neurologist for blinded review. The goal of the review sessions was to identify the start and end times of epileptic seizures to define "true seizure" epochs for later performance evaluation of the automatic seizure detection algorithm. The 1603 hours of EEG consisted of a maximum of 30 hours of continuous EEG data from each subject. For subjects without epilepsy the first available 30 hours of recording were included. The EEG experts were asked to mark the time positions of the seizure onset and end. An event was considered as "true seizure" only if the time interval of two out of three reviewers overlapped by at least 1 second. A seizure epoch was the overlapping time range of two reviewers.

Detection Performance

To define positive percentage agreement (PPA) and negative disagreement rate (NDR, given as false detections in 24 hours) for each patient the seizure epochs defined by consensus annotations of two out of three reviewers were compared to automatically calculated seizure time points of the encevis seizure detection device and the predicate device Persyst. The encevis seizure detection device results in a single time point for each detection that is used in this validation. The predicate device Persyst was used with default settings (perception score = 0.5) and the given start time point was used in this validation. The logical variables true positive (TP), false positive (FP), and false negative (FN) are defined as follows: seizure epochs are counted as TP if at least one detection occurred within the time range of the consensus annotation plus/minus 30 seconds. Detections outside of seizure epochs were defined as false positives (FP). Seizure epochs without a matching detection were defined as false negative (FN).

Results

The average patient-wise positive percentage agreement of the 46 subjects with at least one "true seizure" event resulted in 97.6% (95% Cl=[92.6, 99.5]) for encevis seizure detection and in 83.7% (95% Cl=[71.4, 91.7]) for the predicate device Persyst. The overall positive percentage agreement (the sum of all TP divided by the total number of seizures in the reviewer reference standard) is PPA = 91/97 = 93.8% (95% CI=[87.0%, 97.7%]) for encevis 2.1 and 75/97 = 77.3% (95% Cl=[67.7%, 85.2%]) for Persyst 12.

The average negative disagreement rate (NDR) was 33.7 false detections in 24 hours (95% Cl=[25.5, 47.7]) for the encevis seizure detection and 10.5 false detections in 24 hours (95% Cl=[7.4, 15.4]) for predicate device Persyst.

To show the non-inferiority of encevis in terms of positive percentage agreement with the clinical reference standard as defined by the consensus expert review, we compared the differences of patient-wise estimates for PPA (defined

{26}------------------------------------------------

as TP/(TP+FN) within single patients) between encevis and Persyst 12 using one-sided Wilcoxon signed-rank test for paired samples (Wilcoxon, 1945).

The resulting rank-sum of 45 showed, that the patient-wise positive percentage agreement of encevis was higher compared to Persyst (p=0.003). This shows that the number of seizures encevis detected within a single patient recording is higher compared to Persyst 12. Furthermore, the lower bound of the 95% confidence interval of the overall PPA for encevis (87.0%) exceeds the overall PPA point estimate of Persyst (77.3%). This shows the noninferiority of encevis 2.1 compared to Persyst 12 in terms of patient-wise PPA and overall PPA. Beyond noninferiority, the results are even suitable to show the superiority of encevis 2.1 compared to Persyst 12 in terms of patient-wise PPA and overall PPA.

9.2 Detection of seizure and status epilepticus for acute care performance testing

Study population

We included scalp-EEG recordings from 81 patients which were recorded in neurological/general intermediate care units or neurological/general intensive care units at two different sites in the US (31 patients) and outside of US (50 patients). In both sites, patients with and without frequent recurrent focal electrographic seizures and/or status epilepticus in the EEG were randomly selected from the clinical EEG databases.

Reference Standard

To define the reference standard for seizures, a total of 62.4 hours of EEG from the 81 subjects were presented to six experienced, board-certified, and independent Neurologists for blinded review to annotate beginning and end of electrographic seizures according to the ACNS criteria.

To define a reference standard for electrographic status epilepticus (ESE) we followed the definition in the ACNS Standardized Critical Care EEG Terminology and determined presence of ESE. The reference standard for seizure burden was derived from consensus seizure annotations. Within moving windows over 10 minutes and over 60 minutes, the total seizure duration was accumulated and normalized by the window lengths.

Detection Performance

Seizure Detection

To define positive percentage agreement (PPA, sensitivity) and negative disagreement rate (NDR, given as false detections in one hour) for each patient the seizure epochs defined by positive consensus annotations were compared to automatically calculated seizure detections of encevis and the predicate device Persyst 12. The logical variables true positive (TP), false positive (FP), and false negative (FN) are defined as follows: positive consensus seizures are counted as TP if at least one seizure detection overlaps with the time range of the annotation. Positive consensus seizures without a matching detection were defined as false negative (FN). Detections outside the time range of any negative consensus annotation were defined as false positives (FP).

Status Epilepticus

To assess the effectiveness and clinical utility of status epilepticus detection, EEGs with ESE according to the consensus reference standard were counted as true positives (TP), if at least one seizure with a positive status epiledicus flaq was detected by the software, and as false neqative (FN) otherwise. EEGs with no ESE according to the consensus reference standard were counted as true negative (TN), if no seizure with a positive status epilepticus flag was detected by the software, and as a false positive (FP) otherwise. PPA was calculated, and negative percentage agreement (NPA) was calculated as the number of true negatives (TN) divided by the sum of TN and false positives (FP). PPA and NPA were calculated together with 95% confidence intervals.

Seizure Burden

To assess effectiveness and clinical utility of seizure burden measures, short-time seizure burden (STSB) and hourly seizure burden (HSB) at distinct levels, positive percentage agreement (PPA, sensitivity) and negative percentage

{27}------------------------------------------------

agreement (NPA, specificity) with respect to the reference standard were assessed. Within the 15-minute periods, which were used to calculate the reference standard, maximum values of STSB and HSB were used for comparison with the reference standard. PPA and NPA were calculated together with 95% confidence intervals.

Results

Seizure Detection

Based on the 183 seizure events, values for PPA and NDR, average values were calculated together with 95% confidence intervals by means of clustered bootstrapping using the method of bias-corrected and accelerated intervals (BCa). To preserve the intra-patient correlation structure, we resampled all patients ("clusters") from our dataset with replacement. All seizure events for each of the sampled patients were included to calculate a single PPA value, forming one sample within the bootstrap distribution. The results are shown in the table below. The confidence intervals for PPA of encevis and Persyst show the non-inferiority of encevis to Persyst 12 in terms of sensitivity. This is different for specificity, here Persyst is superior to encevis. This shows that the two seizure detectors are set to different operating points. The reason for the choice of our sensitive operating point is that this seizure detector is used as the basis for seizure burden and status epilepticus detection, where it is important to not only detect some parts of a seizure, but rather the full duration of seizures. We therefore accept a higher false positive rate.

	encevis 2.1	Persyst 12
Positive percentage agreement [95%CI]Event-based, 42 subjects	71.6 %[54.0 % - 86.9 %]	41.5 %[23.3 % - 62.7 %]
Negative disagreement rate[95% CI]81 subjects	2.0 / hour[1.1 - 3.7]	0.26 / hour[0.049 - 0.84]

In a subgroup analysis PPA and NDR were calculated for patients from the ICUs in comparison to patients from acute care environments without ICUs and are shown in the table below. Event-based PPA for seizure detection was 76.6% in the ICU environment subgroup and 60.0% in acute care environment without ICU. Patient-based NDR was 2.3/h and 1.9/h for the ICU and AC environments, respectively.

Environment	Patients withseizures	Patients w/oseizures	PPA [95% CI]	NDR [95% CI]
ICU	25	29	76.6 %[53.2% - 91.3%]	2.3[1.0 - 5.1]
AC (w/o ICU)	17	10	60.0 %[36.3% - 91.7%]	1.7[0.75 - 3.6]

In a subgroup analysis of event-based PPA and NDR we calculated for 4 groups of different seizure duration (<1 min, 1-3 min, 3-10 min, >10 min), for the complete study cohort (81 patients) as well as for the environments of use ICU (54 patients, 128 seizures) and acute care environments without ICU (27 patients, 55 seizures). To calculate the confidence intervals, cluster bootstrapping was used with resampling on patient-level, i.e., all seizure events belonging to a seizure duration group of a patients were included into bootstrapping samples. The results in the table below show that seizures lasting for at least 3 minutes are detected with high sensitivity, whereas seizures that are shorter than 1 min are detected less reliably.

Environment	Seizure duration[min]	Number of patients	Number ofseizures	PPA [%] with95% CI
-------------	---------------------------	--------------------	-----------------------	------------------------

{28}------------------------------------------------

All81 patients183 seizures	< 1	16	72	54.2 %[32.6 % - 78.2 %]
	1-3	16	70	74.3 %[45.6 % - 92.5 %]
	3-10	13	22	95.5 %[69.2 % - 100.0 %]
	10+	18	19	100.0 %
ICU54 patients128 seizures	< 1	11	49	53.1 %[22.3 % - 81.0 %]
	1-3	10	49	87.8 %[60.7 % - 98.2 %]
	3-10	10	19	94.7 %[61.9 % - 100.0 %]
	10+	11	11	100.0 %
AC w/o ICU27 patients55 seizures	< 1	5	23	56.5 %[32.1 % - 91.7 %]
	1-3	6	21	42.9 %[6.3 % - 90.0 %]
	3-10	3	3	100.0 %
	10+	7	8	100.0 %

Status Epilepticus

Positive percentage agreement (PPA) and negative percentage agreement (NPA) were calculated together with 95% confidence intervals by bootstrapping. The PPA was 82.6% [Cl 60.9%-95.7%], whereas the NPA was 91.4% [Cl 81.0%-96.6%],

ElectrographicStatus Epilepticus (ESE)per patient		Reference standard
		Non-ESE	ESE
encevisdetections	Non-ESE	53	4
	ESE	5	19
		NPA91.4%CI95 [81.0%-96.6%]	PPA82.6%CI95 [60.9%-95.7%]

In a subgroup analysis PPA and NDR were calculated for patients from the ICUs im comparison to patients from acute care environments without ICUs and are shown in the table below.

Environment	Patients with ESE	Patients w/o ESE	PPA [%]	NPA [%]
ICU	15	39	86.7 %[60.0 % - 100.0 %]	92.3 %[79.5 % - 97.4 %]
AC (w/o ICU)	8	19	75.0 %[37.5 % - 100.0 %]	89.5 %[68.4 % - 100.0 %]

Hourly Seizure Burden

We tested the HSB for exceeding the threshold of 10%, corresponding to "Frequent hourly seizures" occurring within one hour of time.

{29}------------------------------------------------

Hourly Seizure BurdenFrequent (< 10%) cut-off		Total
		Reference standard
		<10%	>= 10%
encevisdetections	< 10%	135	7
	>= 10%	19	46
		NPA	PPA
		87.7%Cl95 [81.8%-92.2%]	86.8%Cl95 [75.5%-94.3%]

In 46 of 53 segments (15 minutes duration), where the reference standard exceeded the level of 10%, the HSB values calculated by the seizure burden component did also exceed the 10% level. This leads to a PPA of 86.8% [Cl 75.5%-94.3%]. In the remaining 154 segments, where the reference standard did not exceed the level of 10%, the HSB values calculated by the seizure burden component were also below 10%. The resulting NPA therefore is 87.7% [Cl 81.8%-92.2%].

Short-time Seizure Burden

We tested the STSB for exceeding the threshold of 10%, corresponding to "Frequent seizures" and the threshold of 50%, corresponding to "Abundant seizures", occurring within 10 minutes of time. The results are shown in the first of the following tables for both thresholds simultaneously, and in in the second and third of the following tables for the test of both thresholds separately.

Short-time Seizure Burden15-minute segments	Reference standard
	< 10%	10% - 49%"Frequent"	> 50%"Abundant"
	encevisoutput	< 10%	118	5
10% – 49%		15	17	4
> 50%		5	3	39

Short-time Seizure BurdenFrequent (< 10%) cut-off		Reference standard
		< 10%	>= 10%
encevisoutput	< 10%	118	6
	>= 10%	20	63
		NPA	PPA
		85.5%	91.3%
		Cl95 [79.0%-90.6%]	Cl95 [82.6%-97.1%]

In 63of 69 segments, where the reference standard exceeded the level of 10%, the STSB values calculated by the seizure burden component did also exceed the 10% level. This leads to a positive percentage agreement of 91.3% [Cl 82.6%-97.1%]. In the remaining 138 segments, where the reference standard did not exceed the level of 10%, the STSB values calculated by the seizure burden component were also below 10%. The resulting NPA therefore is 85.5% [Cl 79.0%-90.6%].

Short-time Seizure BurdenAbundant (< 50%) cut-off10-minute segments		Reference standard
encevisoutput	< 50%	>= 50%
< 50%	155	5
>= 50%	8	39
	NPA95.1%Cl95 [90.8%-97.5%]	PPA88.6%Cl95 [77.3%-95.5%]

In 39 of 44 segments, where the reference standard exceeded the level of 50%, the STSB values calculated by the seizure burden component did also exceed the 50% level. This leads to a positive percentage agreement of 88.6% [C] 77.3%-95.5%]. In the remaining 163 segments, where the reference standard did not exceed the level of 50%, the STSB values calculated by the seizure burden component were also below 50%. The resulting NPA therefore is 95.1% [Cl 90.8%-97.5%].

Benefit Risk Analysis

{30}------------------------------------------------

Analysis of the benefits and risks of the software is performed according to the FDA quidance document "Benefit-Risk Factors to Consider When Determining Substantial Equivalence in Premarket Notifications (510(k)) with Different Technological Characteristics." (September 2018). A similar benefit and risk analysis had been performed by Ceribell in their analysis of substantial equivalence with the Ceribell Status Epilepticus Monitor (K223504). We believe that the submitted performance validation data clearly demonstrates significant benefit of the encevis detection of seizures and status epilepticus due to the maqnitude and the probability of the benefit of faster detection of ESE. At the same time, the identified risks of the software are of low probability and low severity, post-mitigation.

Benefits

The greatest benefits of the software are specifically tied to the intended use of detecting electrographic status epiledticus. The maximum benefit of the software occurs when ESE is recognized and detected during the time prior to the availability of a qualified neurologist to perform a full review of the underlying EEG. Initiation of treatment for Status Epilepticus is highly time sensitive, yet in the standard-of-care workflow it can take 12-24 hours just to get the full EEG read by a qualified neurologist. Automatic detection of ESE provided by the software enables early diagnosis of status epilepticus. In the intended use of the software, EEG recordings are analyzed either in parallel with the recording (online) or in regular, preferably short intervals based on stored EEG files. In case of detected ESE, a Neurologist can be called to review EEGs indicated by the software and decide for appropriate treatment. Administration of first line antiseizure medications (ASMs) and initiation of other time-sensitive actions consequently can be performed as quickly and as accurately as possible. At the software does not replace the full review of the underlying EEG by a qualified neurologist because pathologies other than ESE may be present in the EEG.

Risks

Risks of the software can be categorized into risks associated with false-positive detections, false-negative detections, device malfunctions, or device misuse. In general, these risks are all low in part because in all potential cases of failure of the software, the patient remains no worse off compared to the current standard-of-care, where the intensive care physician is forced to make a treatment decision without having EEG data available. Table 16 and Table 17 in the Appendix provide a detailed analysis of the benefits and risks of the software.

Comparison with the predicate devices

A comparison of the encevis ESE detection with the predicate device Ceribell Status Epilepticus Monitor shows that performance values obtained for encevis in this study are comparable to the values reported in the performance evaluation of the Ceribell device. They reported a sensitivity of 100%, 95%-confidence intervals [100%, 100%], [72%, 100%], or [78%, 100%] depending on three different calculation methods, which they used due to only 10 ESE positive samples. The PPA for ESE detection with encevis 2.1 is 82.6% [C] 60.9%-95.7%], which is within at least two of the three reported confidence intervals for sensitivity. They furthermore reported a specificity of 94%, with a 95%confidence interval [91%, 96%]. The NPA for ESE detection with encevis 2.1 is 91.4% [Cl 81.0%-96.6%], which is also within the reported confidence intervals. However, this is a comparison of performance values from different studies with different subjects and data.

In the light of similar performance values, the benefit risk profiles for both devices are equivalent if the software is used in online mode (in parallel with the EEG recording by a medical EEG device encevis 2.1 and the predicate device Ceribell Status Epilepticus Monitor thus can be seen substantially equivalent in the detection of electrographic status epilepticus.

If the software is used in offline mode, i.e., if previously recorded EEG files are analyzed, the risks are not different compared to online use. When the software correctly identifies that ESE is present, the benefit of faster treatment is still increased compared to the current standard of care. It is, however, not as high as in the use. In the benefit analysis, the magnitude of this benefit was therefore estimated to be moderate-to-high in the offline use case as compared to high in the online use case. Consequently, online use should be encouraged in the instructions of use if an EEG with this capability is available. Otherwise, it should be recommended to make post-hoc analyses ("offline") in regular and short intervals.

{31}------------------------------------------------

9.3 encevis spike detection performance testing

The reference standard was determined based on the results of blinded review sessions from three neurologists. The "true spike" events (reference standard) were then compared to automatically calculated spike time points of the encevis spike detection device to define true positives (FP), false negatives (FN), and true negatives (TN) for each patient. With these values the positive percentage agreement (PPA) and negative percentage agreement (NPA) for each patient are calculated. In addition, "true spike" events were compared to the automatic detections of Persyst resulting in PPA and NPA values for the predicate device. Furthermore, the localization performance of both systems encevis spike detection and Persyst was evaluated based on the localization information given by the detection systems and the spatial information provided by the reviewer (clinical truth). We define a positive localization percentage agreement (PLPA) which is calculated for each patient.

Study population

To prove the validity of the spike detection system, encevis spike detection was tested with the EEG of 23 patients. For clinical validation, we included scalp-EEG recordings of 23 subjects that underwent video-EEG monitoring in an epilepsy monitoring unit for differential diagnosis or pre-surgical evaluation. 18 subjects of 18 years of age or older that showed spike events during recording based on initial clinical information where included. In addition, five subjects of 18 years of age or older that were diagnosed of not having epilepsy were included. No further selection of subjects were made.

The statistical parameters PPA, NPA and PLPA were used in a two one-sided test (TOST, (Walker E et. al.) using paired samples in order to show the non-inferiority of encevis spike detection device compared to the predicate of Persyst.

Reference standard

To define the reference standard the EEG from all subjects were presented to three independent Neurologists for blinded review. The goal of the review sessions was to identify all "true focal spikes" for later performance evaluation of the automatic spike detection algorithm. The EEG experts were asked to mark the beginning and the end of the spike. Furthermore, the reviewers were asked to specify the electrode which is next to the spike maximum (phase reversal).

An event was considered as "true spike" only if the time interval of two out of three reviewers overlapped. For the determination of the localization performance, the 3D-coordinates of the electrode which is next to the spike maximum averaged over reviewers was used. The determined average position is considered as the reference standard with respect to the localization and is used to evaluate the localization performance of encevis spike detection and the predicate Persyst.

Performance evaluation

Data of all 23 subjects was processed with encevis spike detection. In order to compare the obtained results of encevis spike detection with the predicate Persyst, the same data was processed with the spike detector of Persyst 12. The detection systems were evaluated by means of suitable performance measures like positive percentage agreement (PPA) and negative percentage agreement (NPA). For measuring the localization performance, we defined a positive localization percentage agreement (PLPA). The basis for the performance evaluation are the annotations of the EEG experts which were placed at the onset and the spike. Comparison between the time instances of the annotations and the time instances of automatic detections allows assessing the performance. The detection resolution of both systems, encevis spike detection and the Persyst spike detection was one microsecond.

Results of the performance measures

The average positive percentage agreement of the 15 subjects with at least one "true spike" in 84.81% (95% Cl=[78.5-91.1]) for encevis spike detection and in 8.7% (95% Cl=[4.4-13.0]) for the predicate device Persyst.

{32}------------------------------------------------

The average negative percentage agreement of all 23 subjects was 98.58 (95% Cl=(98.1.-99.1)) for the encevis spike detection and 99.69 (95% Cl=[99.4-99.9]) for predicate device Persyst.

The average correct localization ratio of the 12 subjects with at least one "true positive" event was 95.63 (95% Cl=(91.0-100.2)) for the encevis spike detection and 93.97 (95% Cl=(83.6-104.31) for predicate device Persyst.

A Two One-Sided Test (TOST) for paired samples (Walker et al) was used to test the non-inferiority of the encevis spike detection device to the predicate device. For statistical comparison. a type I error of 0.05 and non-inferiority margins of 3% for positive percentage agreement (PPA), the negative percentage agreement (NPA) and the positive localization percentage agreement (PLPA) are used. The three performance measures PPA, NPA and PLPA were tested independently to measure the non-inferiority of all device parameters separately.

9.4 encevis artifact reduction performance testing

The quality of an artifact removal algorithm is determined by two aspects.

1. The method should not significantly modify true, clean EEG pattern that are not disturbed by artifacts. To quantify the performance of the algorithms with regard to this aspect, changes in clean EEG patterns due to the algorithms are evaluated.
1. The method should suppress artifacts that are superimposed on the true EEG as far as possible, revealing the underlying, pure EEG patterns. To quantify the ability of the algorithms to remove artifacts, signal-to-noise ratios will be measured before and after artifact removal.

For these measurements we need clean, pure EEG patterns and artifacts of different types. In order to identify these patterns, three EEG experts Neurologists are engaged as independent reviewers.

Validation data

For the validation study, 128 EEG data records from different patient groups are used, covering all intended use populations of encevis, i.e., adult patients in epilepsy monitoring and in critical care. Each record consisted of 10 seconds of data to be evaluated. The datasets include 60 patients from epilepsy monitoring units and 65 from ICU patients. These data were selected as follows:

Epilepsy monitoring – seizure EEGs: We include 31 EEG segments from 31 subjects of 18 years of age or older that underwent video-EEG monitoring in an epilepsy monitoring unit for the purpose of differential diagnosis or presurgical evaluation and that showed seizure events during recording and were diagnosed of having epilepsy.

Epilepsy monitoring – spikes: We include 33 EEG segments from 6 subjects of 18 years of age or older that underwent video-EEG monitoring in an epilepsy monitoring unit for the purpose of differential diagnosis or presurgical evaluation that showed spikes during recording.

Intensive care unit: We include 65 EEG segments from 65 subjects of 18 years of age or older that have been admitted to an intensive care unit due to severe neurological disorders (cerebral ischemia, cerebral hemorrhage of different genesis, cerebral tumors, status epilepticus, toxidromes, encephalopathies of different genesis, cerebral malformations and craniocerebral traumas) on a systemic or localized basis. The random selection includes 9 segments with seizures, 10 segments with rhythmic activity, 11 segments with periodic discharges, 17 segments with burst-suppression and 18 segments without any pattern.

Expert review

For this validation study we need annotations of clean EEG recordings without any artifacts, and moreover annotations of artifacts that can be superimposed to the clean recordings. We engage three independent epileptologists or neurologists for blinded review of the EEG data from EMU and ICU.

Statistical testing

{33}------------------------------------------------

A Two One-Sided Test (TOST) procedure for paired samples (Walker E, Nowacki AS, J Gen Intern Med. 2011 Feb; 26(2):192-6) is used to test the non-inferiority of the encevis artifact reduction compared to the predicate device. For statistical comparison, a type I error of 0.05 and non-inferiority margins of 1dB.

The hypothesis to test non-inferiority of the relative suppression of true EEG in dB is defined as:

H0: The relative suppression of true EEG in dB of the encevis artifact removal is higher than the suppression of true EEG in dB of the predicate device.
H1: The relative suppression of true EEG in dB of the encevis artifact removal is lower than or equal to the suppression of true EEG in dB of the predicate device.

The hypothesis to test the signal-to-noise ratio after artifact removal is defined as:

H0: The signal-to-noise ratio after artifact removal by encevis is lower than the signal-to-noise ratio after artifact removal by of the predicate device.
H1: The signal-to-noise ratio after artifact removal by encevis is higher than or equal to the signal-to-noise ratio after artifact removal by of the predicate device.

The results of the evaluation of relative suppression of clean EEG are summarized in the following table in %. This number means, that the variance of the clean EEG activity has been suppressed by this relative value, i.e., low values are desired. Due to technical reasons, only 127 out of 131 test cases could be evaluated: in the remaining 4 cases, Persyst produced zero lines in all channels.

{34}------------------------------------------------

Image /page/34/Figure/1 description: The image is a bar graph comparing the relative suppression of clean EEG in percentage for two different systems, encevis and Persyst. The y-axis lists numbers from 1 to 131, while the x-axis shows the percentage of relative suppression, ranging from 0% to 80%. Each number on the y-axis has two bars, one representing encevis and the other representing Persyst, indicating the relative suppression percentage for each system. The graph provides a visual comparison of the performance of the two systems in suppressing clean EEG.

Relative suppression of clean EEG by encevis and Persyst

The results of the evaluation SNR prior and post artifact removal are summarized in the following table in dB. This number means, that the signal-to-noise ratio (noise=artifacts) has been achieved due to artifact removal, i.e., high values are desired. Eleven out of 104 test cases could not be evaluated, since the artifacts in these cases were on channels, where the initial EEG was not undistorted according to reviewers. The remaining 93 cases have been evaluated.

{35}------------------------------------------------

Image /page/35/Figure/1 description: The image is a bar graph titled "SNR prior and post artifact removal [dB]." The x-axis is labeled from -60 to 40. The y-axis is labeled with numbers from 1 to 104. The graph shows the initial SNR, encevis, and Persyst values for each number on the y-axis.

SNR prior and post artifact removal by encevis and Persyst

Results of Statistical testing

The results of the Two One-Sided Test for relative Suppression of clean EEG (Test-Control) are (95% delta CI=[-0.07, -0.02], margin = 0.01):

$$\bullet \bullet \bullet \bullet \bullet \bullet \bullet \bullet \bullet \bullet$$

The results of the Two One-Sided Test for signal-to-noise ratios after artifact removal are (95% delta Cl=[4.37, 5.88], margin = 0.01):

{36}------------------------------------------------

Both device parameters, "relative Suppression of clean EEG" and "signal-to-noise ratios after artifact removal" of the encevis artifact reduction are therefore non-inferior to the parameters of predicate device Persyst.

In the statistical evaluation of both device parameters, "relative Suppression of clean EEG" and "signal-to-noise ratios after artifact removal" of the encevis artifact reduction are shown to the parameters of predicate device Persyst. Moreover in 73 out of 127 test cases, the suppression of clean EEG by encevis was lower compared to Persyst. And in 83 out of 93 test cases, the SNR after artifact removal by encevis was higher compared to Persyst. It can be concluded that the encevis artifact reduction "PureEEG" does not perform worse that the artifact reduction by the predicate device.

9.5 encevis rhythmic and periodic patterns performance testing

The detection of rhythmic and periodic patterns in encevis NeuroTrend and encevis acute care is used to visually mark EEG segments with rhythmic or periodic signal content. The definition of rhythmic and periodic patterns follow the quidelines of the ACNS (American Clinical Neurophysiology Society) ICU EEG Terminology (Hirsch et al., 2013). NeuroTrend and encevis acute care display all detected rhythmic and periodic patterns in plots called "Pattern Localization" and "Pattern Frequency".

For the validation we compared and statistically analyze annotations of two human EEG-readers with the detections of the rhythmic and periodic pattern detection of NeuroTrend and encevis acute care. We showed that the detected patterns have a high sensitivity compared to manually annotated EEG segments. We prospectively recorded 83 long term EEGs from ICU-patients at two different centers using the international 10-20 electrode system with a sampling rate of 256Hz.

EEGs were annotated by two clinical neurophysiologists that were naive to these EEGs. The annotation procedure included the first minute of each hour, were each minute was split into three independent segments of 20 seconds resulting in 11935 common annotation segments. Several non-overlapping annotations were allowed in each annotation segment. Each annotation may have an arbitrary start and end position but has to be fully included in the annotation segment. For each annotation, the reviewer was allowed to choose between one of the following pattern types:

1. PD: periodic pattern
1. RDA: rhythmic delta activity
1. RTA: rhythmic theta activity
1. RAA: rhythmic alpha activity
1. SW: rhythmic spike-and-wave activity
1. BS: burst suppression pattern
1. No annotation (short NOPA).

In addition to the type of the pattern the localization property had to be set by the human reviewers. This property was defined in (Hirsch et al., 2013) as main term 1:

1. G: generalized pattern
1. L: lateralized pattern

The annotations from the two reviewers were then used as gold standard condition to test sensificity of the rhythmic and periodic pattern detection of NeuroTrend and encevis acute care. Annotations had to be consistent between both reviewers to be used in the sensitivity and specificity measurement.

The detection performance was defined by assigning one of four possible test conditions to each of the 1 minute annotation segments: true positive (TP), false positive (FP), true negative (TN), and false negative (FN), An annotation segment was counted as TP if a detection and an annotation seqment with a

{37}------------------------------------------------

gold standard annotation but without any detection will be counted as FN. An annotation segment with detections but without annotations will be counted as FP. An annotation segment without gold standard annotation and without detections will be counted as TN.

The sensitivity is defined as:

SE [%] = #TP/(#TP+#FN) * 100

The specificity is defined as:

SP [%] = #TN/(#TN+#FP) * 100

The # symbol stands for "number of". The symbol "#TP" represents the number of true positive annotation segments.

The localization information will be validated by comparing the concise annotations of the two human reviewers for all correctly detected markers (the TP detections).

The result of the manual annotation procedure was evaluated using the Cohens' kappa statistic measures the level of agreement between two reviewers. A kappa value of 0.66 was measured between reviewer 1 and reviewer 2.

	Reviewer 2
Reviewer 1	NOPAT	PD	RAA	RDA	RTA	RDA+S
NOPAT	10757	311	14	78	23	4
PD	588	1290	0	63	14	0
RAA	1	1	6	1	10	0
RDA	135	5	0	119	1	4
RTA	50	25	1	23	107	0
RDA+S	10	0	3	0	0	20
Cohens Kappa:	0.66 (CI=0.64-0.67)		Substantial agreement

Cohens' kappa statistic for the evaluation of the pattern detection

The results of the validation are given in Table 3. The overall detection performance measures the sensitivity and specificity of the rhythmic and periodic pattern detection without evaluating the pattern type. The result is marked with the label "ANY" in the result file. This result proofs the ability of the rhythmic and periodic pattern to detect any relevant pattern and ignores pattern type mismatches. The result of the periodic pattern group is labeled as "PD". This result shows the sensitivity and specificity of the periodic pattern detections. The rhythmic delta activity pattern detections is labeled as "RDA". The result of the ARA group shows the result of aggressive rhythmic activity, including the pattern types RTA, RAA, and RDA+S. Reviewer annotations of SW and RDA+S are considered equivalent

Pattern Type	Sensitivity[%]	Specificity[%]
ANY	81.86 (79.9 - 83.8)	83.80 (83.1 - 84.5)
PD	69.73 (67.2 - 72.3)	95.89 (95.5 - 96.3)

{38}------------------------------------------------

ARA (including RTA, RAA, RDA+SW)
89.40 (84.2 - 94.6)	94.85 (94.5 - 95.3)
RDA
91.73 (86.4 - 97.1)	86.05 (85.4 - 86.7)

Sensitivity and specify for encevis pattern detection

The inter reader agreement table of the localization information (ACNS Main Term 1) compares the consistent annotations of two EEG experts to the localization shown in NeuroTrend and encevis acute care. The result is shown in the following table:

	NeuroTrend/encevis acute care
Reviewer 1+2	G	L
G	891	86
L	130	175
Cohens Kappa:	0.51, CI=0.45-0.57(Moderate agreement)

Inter-reader agreement between reviewers and NeuroTrend and encevis acute care pattern localization

9.6 encevis aEEG performance testing

Amplitude-integrated EEG (aEEG) is a popular method for monitoring by displaying the amplitude trend of brain activity. It is the boundary of the EEG waveform (i.e. the envelope) and not the EEG itself (i.e. the carrier) that characterizes the tendency of amplitude changes (Zhang and Ding, 2013).

The aEEG module of NeuroTrend and encevis acute care estimates and visualizes the temporal evolution (trend) of the EEG amplitude. The implementation is oriented on the proposed method of (Zhang and Ding, 2013)

In the first step the frequency response of the module is checked for equality with the proposed method of (Zhang and Ding, 2013). This test only considers the correct slope (dB loss per decade) not the correct filter gain factor. In this test, sinusoidal one-channel test data with increasing frequencies from 0.5Hz to 32Hz and amplitude of 40µV are generated, one test case for each hemisphere. With the results of the frequency response is determined and checked if the dB loss per decade within the band pass (cut-off frequencies of 2 and 15Hz) is -12db/dec and the maximum gain factor in the stop band is not greater than -30dB. This step validates the correct implementation of the filters and its characteristics (expect the gain factor) within the module.

In the second step the results of the module are compared with the aEEG results of Persyst (CE certified and FDA approved software; http://www.persyst.com/) using real EEG data. The configuration of Persyst is set in a way to allow an adequate comparison.

After successful validation according to the description above we will have shown that the aEEG module correctly determines the averaged EEG amplitude of the left and right hemisphere according to the proposes method of (Zhang and Ding, 2013)

In the first validation step the frequency response of the aEEG module is checked for equality with the pro-posed method of (Zhang and Ding, 2013). The following conclusions were drawn from the results:

· The determined characteristic is very similar to the published version in (Zhang and Ding, 2013). Only the absolute shift of the complete frequency response is different but because only changes in aEEG values are of clinical relevance this detail is irrelevant.

{39}------------------------------------------------

Both hemispheres show the same characteristic
· In the stop band there is a suppression of -30dB and higher
The slope in the pass band is approximately -12dB/decade

In the second step the results of the aEEG module are compared with the aEEG results of Persyst. For this test, real EEG data were used. The aEEG of the same EEG segment using either Persyst or the aEEG module of encevis NeuroTrend or encevis acute care were compared. The test cases showed that the aEEG module of encevis NeuroTrend and encevis acute care and Persyst are in good accordance. Furthermore, the aEEG values are in good accordance with the corresponding raw EEG.

9.7 encevis frequency bands performance testing

The background-frequency module of NeuroTrend and encevis acute carte estimates and visualizes the temporal evolution (trend) of relative proportions of dominant EEG-waveform-frequencies. The result is graphically presented using a plot (of. Figure 10), where the x-axis represents the time-axis, and four stacked areas in different colors and widths represent the relative proportions of the four frequency bands Delta, Theta, Alpha, and Beta for subsequent time windows with lengths of 15 seconds. The intensity of the colors furthermore corresponds to the amplitudes in these four frequency bands. This representation allows the user to identify time epochs that are dominated by a specific frequency band. E.g., EEG-slowing or, in other words, an epoch with dominant delta- or theta-wave can be recognized in the graphical representation by broad stretches of the corresponding areas.

Image /page/39/Figure/7 description: The image shows a stacked proportion chart of frequency bands. The y-axis ranges from 0% to 100%, and the x-axis represents the frequency bands. The chart displays the proportions of different frequency bands, including beta, alpha, theta, and delta. The delta frequency band appears to be the most prominent.

Graphical representation of the Background-EEG-Frequency evaluation results.

In order to proof the validity of the Background-EEG-frequency module we followed a two-step approach. In the first step it was shown that the assignment of sinusoidal test data to frequency bands (Delta, Theta, Alpha, or Beta) is correct according to the above definitions of frequency borders. In this test, sinusoidal test data with frequencies across all four bands and amplitudes ranging from 2 µV to 200 µV were generated. Then it was verified, that the algorithm correctly assigns each test signal to the corresponding frequency band, and that the measurement error for amplitudes are below 5 %. This validates the correct assignments of single, 3-second EEG epochs to a frequency band and amplitude.

In a seconds step it is shown that the globally dominant background frequency within a 15-seconds window is correctly identified. This is done using manually selected EEG recordings from epilepsy- or ICU patients. Each of these EEG samples is representative for a specific background-EEG-frequency band, i.e., it is mainly dominated by delta-, theta-, alpha-, or beta-waves. For these samples the background-EEG-frequency module calculates the proportional composition of frequency bands. The one frequency band with the largest proportion can be seen as the globally dominant background frequency, if this proportion is particularly high. Thus it is verified for each of these representative examples that the relative proportion corresponding to the true frequency band is greater than 50 %.

9.8 encevis burst suppression performance testing

The detection of burst suppression patterns and quantitative measure for the EEG shown in NeuroTrend and encevis acute care was validated using the following approach:

1. The time point of the detected burst suppression patterns will be compared to annotations defined by two clinical EEG experts using EEG data from a multicenter study. Sensitivity and specificity will be calculated.
1. The quantitative measure of the amplitude loss of the suppression time in percent will be validated using an artificial EEG. The EEG file includes a set burst suppression patterns with different values

{40}------------------------------------------------

for suppression time and suppression amplitude loss. The calculation results of the quantitative burst suppression plots shown in NeuroTrend and encevis acute care will be compared to pre-calculated values.

We recorded 83 long term EEGs from intensive care patients from two different centers using the international 10-20 electrode system with a sampling rate of 256Hz. EEGs were annotated by two clinical neurophysiologists that were naive to these EEGs. The annotation procedure included the first minute of each hour resulting in 3978 valid annotation segments. The reviewers were allowed to assign two categories for each annotation segment:

1. EEG with burst suppression patterns ( BS )
1. EEG without burst suppression patterns ( BS )

Statistical analysis of the detection performance was done by defining the annotations of the reviewers as qold standard and by comparing these annotations to the detection results of the computational method. Each one minute EEG segments annotated as "EEG with burst suppression" with an overlapping burst suppression detection segment of 15 seconds was defined as true positive (TP) event. One minute EEG segments annotated as "EEG with burst suppression patterns" without any overlapping burst suppression detection result were defined as false negatives (FN). One minute segments annotated as "EEG without burst suppression patterns" and with an overlapping burst suppression detection result are defined as false positives (FP), all other segments are defined as true negatives (TN).

The following table shows the evaluation results of the automatic burst suppression detection method using 3978 seqments annotated by two reviewers. The results of the automatic burst suppression detection method were compared to the manual annotations of the reviewers. The detection performance was analyzed for consensus annotations of the two reviewers. The consensus annotations only include annotation segments where both reviewers showed the same decision about Burst Suppression pattern. The measured values for sensitivity (SE), specificity (SP), positive predictive value (PPV), and negative predictive value (NPV) prove the validly of the detection algorithm. The very large sample size does imply a statistically high confidence.

Rev. (N)	SE (%)	SP (%)	PPV (%)	NPV (%)
2	87	92	61	98

Performance of the automatic burst suppression detection method

10 encevis spectrogram performance testing

The spectrogram of the module was verified using artificially created data with varying frequencies) and using real EEG data in a comparison with Persyst 12.

In the verification with artificially created EEG data we first used constant sinusoidal waveforms. The detected peak in the frequency of the encevis spectrogram was verified to equal to the frequency of the sinus wave.

In the second step the spectrogram of the module was checked for equality using artificially created EEG data with modulated sinusoidal waveforms. The detected increase in activity in the encevis spectrogram was verified to follow the frequency modulations of the test data.

In the third step the results of the module are compared with the spectrogram results of Persyst. The two test cases showed that the spectrogram visualization of the Spectrogram module of encevis and Persyst are in good accordance. Furthermore, the spectrogram values are in qood accordance with the corresponding raw EEG.

None of the tests failed, the validation of the Spectrogram module was successful.

{41}------------------------------------------------

11 Statement of Substantial Equivalence

The subject device encevis 2.1 is substantially equivalent in design and intended use to the primary predicate device encevis 1.12 and the secondary predicate device Persyst 12 with respect to all components except for the component for detection of seizures and electrographic status epilepticus. With respect to the component for detection of seizures and electrographic status epilepticus the subject device encevis 2.1 is substantially equivalent in design and intended use to the EEG analysis software components of the tertiary and quarternary predicate devices Ceribell Pocket EEG device and Ceribell Status Epilepticus MonitoAny differences between the subject and predicate devices have no significant influence on safety or effectiveness as established through performance testing. Therefore, the encevis raises no new issues of safety or effectiveness when compared to the predicate devices.

Regulation Number and Section

§ 882.1400 Electroencephalograph.

(a)
Identification. An electroencephalograph is a device used to measure and record the electrical activity of the patient's brain obtained by placing two or more electrodes on the head.(b)
Classification. Class II (performance standards).

Acceptance Criteria and Device Performance

Study Details

2. Sample Size and Data Provenance

3. Number of Experts and Qualifications for Ground Truth

4. Adjudication Method for Test Set

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

6. Standalone Performance Study

7. Type of Ground Truth Used

8. Sample Size for the Training Set

9. How the Ground Truth for the Training Set Was Established

Indications for Use

Indications for Use (Describe)

CONTINUE ON A SEPARATE PAGE IF NEEDED.

DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW!

510(k) summary encevis 2.1

Table of contents

1 Submission Sponsor and Application Correspondent

2 Date Prepared

3 Device Identification

4 Legally Marketed Predicate Devices

5 Device description

5.1 encevis EEG-viewer

5.2 Artefact reduction encevis PureEEG

5.3 Seizure detection of encevis NeuroTrend

5.4 Detection of seizures and status epilepticus of encevis acute care

5.5 Spike detection encevis EpiSpike

5.6 Pattern detection and aEEG

5.7 Spectrogram

5.8 External Interface "encevis AlTInterface"

5.9 External Interface "encevis SeizureICUInterface"

6 Indication for Use Statement

7 Substantial Equivalence Discussion

7.1 Comparison to Primary and Secondary Predicate Device

7.2 Comparison to Tertiary and Quaternary Predicate Devices

7.3 Comparison of Intended Use and Technological Characteristics with the Predicate Devices

7.3.1 Indications for Use Comparison

7.3.1.1 General Use

Similarities

Differences

7.3.1.2 Seizure detection component

Similarities

Differences

7.3.1.3 Spike detection component

Similarities

Differences

7.3.1.4 Calculation of quantitative measures

Similarities

Differences

7.3.1.5 Calculation of aEEG

Similarities

Differences

7.3.1.6 Notifications on an on-screen display

Similarities

Differences

7.3.1.7 Artifact Reduction

Similarities

Differences

7.3.1.8 Diagnostic output

Similarities

Differences

7.3.1.9 Component for Detection of Seizures and Electrographic Status Epilepticus

Similarities

Differences

7.3.2 Technological Characteristics Comparison

7.3.2.1 User

7.3.2.2 Use enivronment

7.3.2.3 Input Data

7.3.2.4 Patient registration method

7.3.2.5 Output

7.3.2.6 Mode of operation

8 Non-Clinical performance Data

9 Clinical Performance Data

9.1 Seizure detection performance testing

Study population

Reference Standard

Detection Performance

Results

9.2 Detection of seizure and status epilepticus for acute care performance testing

Study population

Reference Standard