Sleepware G3 is a software application used for analysis (automatic and manual scoring), display, retrieval, summarization, report generation, and networking of data received from monitoring devices used to categorize sleep related events that help aid in the diagnosis of sleep-related disorders. It is indicated for use with adults (18 and older) and infant patients (one year old or less) in a clinical environment by or on the order of a physician.

The optional Somnolyzer scoring algorithms are for use with adults (18 and older) to generate an output that is ready for review and interpretation by a physician. Cardio-Respiratory Sleep Staging (CReSS) is an additionality of Somnolyzer which uses standard Home Sleep Apnea Test HSAT signals (in the absence of EEG signals) to infer sleep stage.

Device Description

Sleepware G3 software is a polysomnography scoring application, used by trained clinical professionals, for managing data from sleep diagnostic devices using a personal computer. Sleepware G3 is able to configure sleep diagnostic device parameters, transfer data stored in sleep diagnostic device memory to the personal host computer, process and auto-score data to display graphical and statistical analyses, provide aid to clinical professionals for evaluating the physiological data waveforms relevant to sleep monitoring, and create unique patient reports.

Sleepware G3 includes an optional Somnolyzer plug-in. The auto-scoring algorithms of the Somnolyzer Inside software can be used in addition to, or in the place of, the auto-scoring algorithms that are included in Sleepware G3.

Sleepware G3, remains unchanged in function and fundamental scientific technology from Sleepware G3 which was cleared under K142988.

AI/ML Overview

Here's a breakdown of the acceptance criteria and the study proving the device meets those criteria, based on the provided text:

Acceptance Criteria and Reported Device Performance

The acceptance criteria for the Somnolyzer scoring algorithms are based on demonstrating non-inferiority to manual expert scoring. The reported device performance indicates that all primary and secondary endpoints were met.

Acceptance Criterion (Non-Inferiority to Manual Expert Scoring)	Reported Device Performance
Full PSG Acquisition:
Sleep stages according to AASM criteria	Non-inferior (all primary and secondary endpoints met)
Arousals during sleep according to AASM criteria	Non-inferior (all primary and secondary endpoints met)
Apneas and hypopneas during sleep according to AASM criteria	Non-inferior (all primary and secondary endpoints met)
Periodic limb movements during sleep according to AASM criteria	Non-inferior (all primary and secondary endpoints met)
HST Acquisition:
Apneas and hypopneas according to AASM criteria	Non-inferior (all primary and secondary endpoints met)
Cardio-Respiratory Sleep Staging (CReSS):
REI based on cardio-respiratory feature-based sleep time is superior to REI based on monitoring time (for HST acquisition)	Evidence provided that REI calculated using CReSS is a more accurate estimate of AHI than REI calculated using total recording time. Accuracy further improved with additional signals: mean difference between REI and AHI reduced from -6.6 events/hour (95% CI -7.51 to -5.71) to -1.76 events/hour (95% CI -2.27 to -1.24).

Detailed Study Information:

Sample sizes used for the test set and the data provenance:
- Test Set Sample Size: A total of 1,204 polysomnography (PSG) and home sleep apnea test (HSAT) files were used in the five clinical studies.
- Data Provenance: The document does not explicitly state the country of origin. The studies are described as using a "large, diverse sample... collected via a number of different platforms," suggesting diverse sources but not specifying geographical location. The studies were likely retrospective, as they involved validating algorithms against existing manual scoring, but this is not explicitly stated.
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Number of Experts: Not explicitly stated how many individual experts were used across all studies. However, the non-inferiority margin for comparisons was "set at the lower-margin of the agreement observed across expert technologists." This implies multiple experts were involved in defining the range of agreement for ground truth.
- Qualifications of Experts: The experts are referred to as Registered Polysomnographic Technologists (RPSGT). This indicates their professional qualification in sleep study scoring.
Adjudication method for the test set:
- The document implies a form of consensus or agreement among experts was utilized to set the non-inferiority margin, but it does not explicitly describe a specific adjudication method like 2+1 or 3+1 for individual cases within the test set. The focus is on comparing the algorithm's performance against the established range of agreement among experts.
If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:
- No, an MRMC comparative effectiveness study was not the primary focus described. The study design primarily involved a standalone evaluation of the AI algorithm (Somnolyzer) against human expert scoring, demonstrating its non-inferiority.
- The document states that Somnolyzer's output is "ready for review and interpretation by a physician," implying it assists human readers by providing a pre-scored output. However, it does not quantify the improvement in human reader performance with AI assistance versus without AI assistance.
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, a standalone evaluation was performed. The clinical studies "validated the Somnolyzer and CReSS algorithms against manual scoring." The non-inferiority claims ("Somnolyzer scoring... is non-inferior to manual expert scoring") directly refer to the algorithm's performance without human intervention in the scoring process.
The type of ground truth used:
- The ground truth was expert consensus scoring. The document states that the algorithms were validated "against manual scoring" by "expert technologists" (RPSGTs). The non-inferiority margin was based on "the agreement observed across expert technologists."
The sample size for the training set:
- The document does not provide information on the training set sample size. The provided text focuses solely on the clinical performance testing for validation.
How the ground truth for the training set was established:
- As the training set information is not provided, the method for establishing its ground truth is also not described in the document.

Summary

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

October 29, 2020

Image /page/0/Picture/1 description: The image contains the logo of the U.S. Food and Drug Administration (FDA). The logo consists of two parts: the Department of Health & Human Services logo on the left and the FDA logo on the right. The FDA logo features the letters 'FDA' in a blue square, followed by the words 'U.S. FOOD & DRUG ADMINISTRATION' in blue text.

Respironics, Inc. Jonquil Mau Regulatory Engineer 1001 Murry Ridge Lane Murrysville, Pennsylvania 15668

Re: K202142

Trade/Device Name: Sleepware G3 Regulation Number: 21 CFR 882.1400 Regulation Name: Electroencephalograph Regulatory Class: Class II Product Code: OLZ Dated: July 30, 2020 Received: July 31, 2020

Dear Jonquil Mau:

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 (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 located 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.

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

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

801); medical device reporting of medical device-related adverse events) (21 CFR 803) for devices or postmarketing safety reporting (21 CFR 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 4. Subpart A) for combination products; and, if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.

Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR Part 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 medical devices and radiation-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).

Sincerely,

Patrick Antkowiak -S

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 Ouality Center for Devices and Radiological Health

Enclosure

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

Indications for Use

510(k) Number (if known) K202142

Device Name Sleepware G3

Indications for Use (Describe)

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

X 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."

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

TRADITIONAL 510(K) SUMMARY K202142

I. Submitter

Contact Person	Jonquil MauRegulatory Engineer
	jonquil.mau@philips.com
Address & Phone	1740 Golden Mile HighwayMonroeville, PA 15146Phone: (678) 768-2773
Manufacturing Facility	Respironics Inc.1001 Murry Ridge LnMurrysville, Pennsylvania 15668

October 29, 2020 Date of Preparation

II. Subject Device

Name of Device:	Sleepware G3
Common/Usual Name:	Polysomnography (PSG) Software
Classification Name:	Electroencephalograph (21 CFR 882.1400)
Regulatory Class:	II
Product Code:	OLZ Automatic Event Detection Software for Polysomnograph withElectroencephalograph

Legally Marketed Predicate Device III.

K142988 Sleepware G3, Respironics Inc. .

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

Device Description IV.

Sleepware G3, remains unchanged in function and fundamental scientific technology from Sleepware G3 which was cleared under K142988.

Intended Use V.

The optional Somnolyzer software application is intended to mark sleep study signals in order to aid in the identification of events and annotation of traces; automatically calculate measures obtained from recorded signals (e.g. magnitude, time, frequency and statistical measures of marked events); and infer sleep staging in the absence of EEG. All output subject to verification by a qualified user.

Comparison of Technological Characteristics with the Predicate Device

The predicate Sleepware G3 software (cleared in K142988), and the subject device software Sleepware G3 K202142 are designed to detect events for polysomnography with electroencephalography. While the Indications for Use statement for Sleepware G3 is not identical to the predicate device, the differences do not alter the intended diagnostic use of the device relative to the predicate. The Indications for Use have been updated to now include that Somnolyzer's scoring results are ready to review by a physician, which is one of the modifications being proposed in this subject and predicate device is based on the following technological elements:

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

. Same intended use as the Sleepware G3 (K142988).
. Same control mechanism and real-time operating principle as Sleepware G3 (K142988).
. Same technology and user interface as Sleepware G3 (K142988).

The following differences exist between the subject and predicate devices:

Modification #1: Added additional claim to Sleepware G3. .

Somnolyzer's auto-scoring events are non-inferior to manual expert scoring and do not require review by a lab technician.

For a full PSG acquisition, Somnolyzer scoring for sleep stages according to AASM criteria is o non-inferior to manual expert scoring from a Registered Polysomnographic Technologist (RPSGT).
For a full PSG acquisition, Somnolyzer scoring for arousals during sleep according to AASM o criteria is non-inferior to manual expert scoring from an RPSGT.
For a full PSG acquisition, Somnolyzer scoring for apneas and hypopneas during sleep о according to AASM criteria is non-inferior to manual expert scoring from an RPSGT.
For a full PSG acquisition, Somnolyzer scoring for periodic limb movements during sleep о according to AASM criteria is non-inferior to manual expert scoring from an RPSGT.
For an HST acquisition, Somnolyzer scoring for apneas and hypopneas according to AASM o criteria is non-inferior to manual expert scoring from an RPSGT.

. Modification #2: Cardio-Respiratory Sleep Staging (CReSS)

For an HST acquisition, Somnolyzer scoring provides an approximate sleep staging based on cardiorespiratory signals (CReSS) such that (given Somnolyzer apnea and hypopnea scoring) the AHI based on the cardio-respiratory feature-based sleep time is superior to the REI based on monitoring time.

Somnolyzer's Cardio-Respiratory Sleep Staging (CReSS) applies clinically-validated algorithms to provide sleep staging for HST studies.

. Modification #3: Indications for Use update

The Indications for Use statement now includes language that Somnolyzer's scoring results are ready to review by a physician. There are no other changes to the Indications for Use from what was previous cleared in K142988. While the Indications for Use statement for Sleepware G3 is not identical to the

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

predicate device, the differences do not alter the intended diagnostic use of the device to the predicate. Please refer to the table below for a side by side comparison of the predicate and subject device Indications for Use statements.

Table 1: Indications for Use Comparison
Predicate Device/ Premarket Notification	Subject Device
Sleepware G3 /K142988	Sleepware G3
Sleepware G3 is a software application used foranalysis (automatic and manual scoring), display,retrieval, summarization, report generation andnetworking of data received from monitoringdevices used to categorize sleep related eventsthat help aid in the diagnosis of sleep relateddisorders. It is indicated for use with infant oradult patients in a clinical environment by or onthe order of a physician.	Sleepware G3 is a software application used foranalysis (automatic and manual scoring), display,retrieval, summarization, report generation, andnetworking of data received from monitoringdevices used to categorize sleep related eventsthat help aid in the diagnosis of sleep-relateddisorders. It is indicated for use with adults (18and older) and infant patients (one year old orless) in a clinical environment by or on the orderof a physician.
The optional Somnolyzer Inside scoring packagehas the same intended use as Sleepware G3, butis indicated for use with adult patients only.	The optional Somnolyzer scoring algorithms arefor use with adults (18 and older) to generate anoutput that is ready for review and interpretationby a physician.
	Cardio-Respiratory Sleep Staging (CReSS) is anadditional functionality of Somnolyzer which usesstandard Home Sleep Apnea Test (HSAT) signals(in the absence of EEG signals) to infer sleepstage.

Modification #4: Notable features and modifications to Sleepware G3 since K142988 was ● cleared:

Improvements to Standard Analysis Desaturation Detection O
Portable Device Manager This feature provides a workflow for assigning, managing and tracking o HST devices assigned to patients.
Sleepware Support Surface Pro Supporting running Sleepware G3 software on a Surface Pro O tablet for offline study review.
Added Sleepware Bedside Assistant ("SBA")– Application for the Microsoft Surface Pro platform. O This application provides the ability for sleep techs to perform bedside bio-calibration and impedance tests.
Added feature allowing the customer to view hypopneas that are associated with 4% desaturation O in addition to the currently-available hypopneas with 3% desaturation per AASM 2.4.
Updated the encryption algorithm. O

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

Performance Data Provided to Demonstrate Substantial VI. Equivalence:

Software Verification and Validation Testing

Software verification and validation testing were conducted and documentation was provided as recommended by FDA's Guidance for Industry and FDA Staff, "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices." The software for this device was considered to have a "moderate" level of concern, since a failure or latent flaw in the software could result in minor harm to the patient.

Non-Clinical Performance Testing

Software verification and validation included software code reviews, automated testing, bench verification testing and labeling review.

Testing confirmed that the device under review performs equivalently to the device predicate Sleepware G3 (K142988). All tests had passing results.

Clinical Performance Testing

Five clinical studies were provided in this submission. These clinical studies included a total of 1.204 polysomnography (PSG) and home sleep apnea test (HSAT) files to validate the Somnolyzer and CReSS algorithms against manual scoring. Four studies assessed the Somnolyzer algorithm. In these studies, The non-inferiority margin for comparisons of Somnolyzer against human scoring were set at the lower-margin of the agreement observed across expert technologists. All of the primary and secondary endpoints were met across all studies, and thus it was concluded that Somnolyzer is non-inferior to human sleep staging and scoring of the apnea hypopnea index (AHI), arousal index (Arl), periodic limb movement of sleep index (PLMS)), and respiratory event index (REI). By demonstrating non-inferiority against human scoring in a large, diverse sample of PSGs and HSATs collected via a number of different platforms, the data demonstrate that Somnolyzer produces output that is as reliable as the output of sleep technologist scoring and is therefore ready for review and interpretation by a physician. A fifth study assessed the CardioRespiratory Sleep Staging (CReSS) algorithm. In that study, we provide evidence that the REI calculated using total sleep time determined by the CReSS algorithm applied to heart rate and airflow signals is a more accurate estimate of the AHI than the REI calculated using total recording time. Accuracy is further improved when CReSS is applied to the heart rate, airflow, and respiratory effort signals (the mean difference between the REI and AHI was reduced from -6.6 events/hour (95% confidence interval -7.51 to -5.71) to -1.76 events/hour (95% confidence interval -2.27 to -1.24) when CReSS was applied to three signals).

VII. Conclusion

The modified Sleepware G3 includes new technological characteristics that do not raise different questions of safety and effectiveness. The proposed scientific evidence and performance testing to

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

evaluate these new and different characteristics are acceptable and demonstrate substantial equivalence to the legally marketed predicate K142988 Sleepware G3.

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).