PreOp is intended for use by a trained/qualified EEG technologist or physician on both adult and pediativ subjects at least 3 years of age for the visualization of human brain function by fusing a variety of EEG information with rendered images of an individualized head model and an individualized MRI image.

Device Description

PreOp is medical device software that combines EEG data and MR images to visualize recorded EEG activity in 3D in the brain. PreOp can be subdivided in 3 main modules: 3D Electrical Source Imaging (i.e. 3D ESI), Report generation and Viewer generation. The device's input is the MRI and EEG data that are uploaded by the user to the PreOp cloud environment. The output of the device is a report containing the results of the visualization and the ability to evaluate the results in 3D using the 3D viewer. The user can access the output through the PreOp cloud environment.

AI/ML Overview

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

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state quantitative acceptance criteria in a dedicated section. Instead, the acceptance is based on demonstrating "substantial equivalence" to a predicate device, particularly in source localization performance.

Acceptance Criterion (Implicit)	Reported Device Performance (PreOp vs. Predicate)
Source Localization Equivalence (Study 1): The PreOp algorithms should be substantially equivalent to the predicate device algorithm in terms of source localization accuracy for epileptic spikes.	"The results demonstrated that the proposed PreOp algorithms were substantially equivalent to the predicate device algorithm" based on concordance ratings by three experienced epileptologists on a sublobular level. This comparison was between sLORETA with FDM using individualized anatomical MRI (PreOp) and sLORETA with FDM using idealized anatomical MRI (predicate).
Source Localization Consistency (Study 2): Performance of spike source localization should be consistent between HD-EEG and LD-EEG recordings within PreOp.	In 13 epileptic spikes across 8 patients, both algorithms (HD-EEG vs. LD-EEG within PreOp) provided identical source locations. In only 3 spikes, the localization was not 100% equivalent but "very close to each other."
Clinical Usability: The device should meet usability requirements.	"Usability validation is part of the Clinical Performance data and PreOp was tested and meets the requirements of following standard: AAMI/ANSI/IEC 62366:2007, Medical devices - Application of usability engineering to medical devices."
Software Verification and Validation: The software should be fit for clinical use and meet relevant standards.	"Validation testing involved algorithm testing which validated the accuracy of PreOp. The product was deemed fit for clinical use." "PreOp was designed and developed as recommended by FDA’s Guidance, 'Guidance for the Content of Premarket Submissions for Software Contained in Medical Device'." "According to AAMI/ANSI/IEC 62304 Standard, PreOp safety classification has been set to Class B."

2. Sample Size Used for the Test Set and Data Provenance

Study 1 (Source Localization Equivalence):
- Sample Size: 18 epilepsy subjects.
- Data Provenance: Retrospective data analysis. Country of origin is not explicitly stated, but given Epilog is based in Belgium, it's likely European or a mix.
Study 2 (Spike Source Localization Consistency):
- Sample Size: Data from 8 patients, evaluating 16 epileptic spikes (13 identical + 3 very close).
- Data Provenance: Not explicitly stated as retrospective or prospective, but likely retrospective. Country of origin is not specified.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

Study 1:
- Number of Experts: Three experienced epileptologists.
- Qualifications: "Experienced epileptologists." Specific years of experience are not provided.
Study 2: Not applicable in the same way, as this study was comparing internal algorithm performance rather than having experts establish a new ground truth based on device output.

4. Adjudication Method for the Test Set

Study 1: The document states that the three experienced epileptologists "were asked to rate whether each of the algorithm solutions (sLORETA with the finite difference model [FDM] using an idealized or individualized anatomical MRI) were concordant on a sublobular level." The specific method for consolidating these ratings (e.g., 2 out of 3 agreement) is not detailed, but it implies a consensus-based approach for determining "substantial equivalence."

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done, What was the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance

No, a traditional MRMC comparative effectiveness study evaluating human reader improvement with AI assistance was not performed. The study focused on comparing the performance of the device's algorithms against a predicate device's algorithms (Study 1) or comparing different internal algorithm configurations (Study 2), with human experts acting as adjudicators for the output in Study 1.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was Done

Yes, the studies primarily assess the standalone performance of the algorithms.

Study 1 directly compares "the source localization accuracy of the PreOp software algorithms" to "the predicate algorithm." Human experts then rated the concordance of these algorithm outputs with post-operative reports. This is a standalone comparison.
Study 2 compares "the performance of spike source localization using HD-EEG recordings... and Low Density LD-EEG recordings" within the PreOp algorithms themselves. This is also a standalone assessment.

7. The Type of Ground Truth Used

Study 1 (Source Localization Equivalence):
- Ground Truth: Clinical outcomes data combined with expert consensus. Specifically, the resected zone (operative data) for subjects who were Engel I postoperatively (favorable outcome) was used as the reference. The "summaries of the postoperative reports" were provided to the epileptologists, who then rated the concordance of the algorithm solutions with this clinical ground truth.
Study 2 (Spike Source Localization Consistency):
- Ground Truth: The "ground truth" here is internal consistency of the PreOp algorithm itself under different input conditions (HD-EEG vs. LD-EEG). There isn't an external ground truth like pathology for this specific study; instead, it verifies the consistency of the algorithm's output.

8. The Sample Size for the Training Set

The document does not provide information on the sample size for the training set used to develop the PreOp algorithms. The studies described are validation studies (test sets) for the already developed software.

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

Since the document does not mention the training set size, it also does not detail how the ground truth for any training set was established.

Summary

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

Image /page/0/Picture/0 description: The image shows 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 consists of the letters "FDA" in a blue square, followed by the words "U.S. FOOD & DRUG" in blue, and then the word "ADMINISTRATION" in a smaller font size below that.

January 8, 2018

Epilog % Patsy Trisler Consultant Qserve Group US Inc. 5600 Wisconsin Avenue Chevy Chase, Maryland 20815

Re: K172858

Trade/Device Name: PreOp Regulation Number: 21 CFR 882.1400 Regulation Name: Electroencephalograph Regulatory Class: Class II Product Code: OLX Dated: October 4, 2017 Received: October 10, 2017

Dear Ms. Trisler:

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. 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 801); medical device reporting of medical device-related adverse events) (21 CFR 803); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820);

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

K172858

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 http://www.fda.gov/MedicalDevices/Safety/ReportaProblem/default.htm for the CDRH's Office of Surveillance and Biometrics/Division of Postmarket Surveillance.

For comprehensive regulatory information about mediation-emitting products, including information about labeling regulations, please see Device Advice (https://www.fda.gov/MedicalDevices/DeviceRegulationandGuidance/) and CDRH Learn (http://www.fda.gov/Training/CDRHLearn). 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 (http://www.fda.gov/DICE) for more information or contact DICE by email (DICE@fda.hhs.gov) or phone (1-800-638-2041 or 301-796-7100).

Sincerely,

Michael J. Hoffmann -S

Carlos L. Peña. PhD. MS for Director Division of Neurological and Physical Medicine Devices Office of Device Evaluation Center for Devices and Radiological Health

Enclosure

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

Indications for Use

510(k) Number (if known) K172858

Device Name PreOp

Indications for Use (Describe)

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."

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

510(k) Summary

1. SUBMITTER

Submitter Name:	Epilog
Submitter Address:	Vlasgaardstraat 529000 Gent, Belgium
Phone Number:	+32484777651
Contact Person:	Gregor Strobbe
Date Prepared:	18 September 2017; updated 18 December 2017
2. DEVICE
Device Trade Name:	PreOp
Common Name:	Electroencephalograph Software
Classification Name,Number &Product Code:	Electroencephalograph21 CFR 882.1400OLX
Class:	II
Classification Panel:	Neurology

3. PREDICATE DEVICES

Primary Predicate Device:	K092844, GeoSource
Intended use:	GeoSource is intended for use by a trained/qualifiedEEG technologist or physician on both adult andpediatric subjects at least 3 years of age for thevisualization of human brain function by fusing a varietyof EEG information with rendered images of anidealized head model and an idealized MRI image.

The primary predicate device has not been subject to a design-related recall.

Secondary Predicate Device:	K001781, Curry Multimodal Imaging Software
Intended use:	No 510(k) Summary Posted; not publicly available

The secondary predicate device has not been subject to a design-related recall.

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

4. DEVICE DESCRIPTION

PreOp is medical device software that combines EEG data and MR images to visualize recorded EEG activity in 3D in the brain. In figure 5.1 we present a general overview of the PreOp device. PreOp can be subdivided in 3 main modules: 3D Electrical Source Imaging (i.e. 3D ESI), Report generation and Viewer generation. The device's input is the MRI and EEG data that are uploaded by the user to the PreOp cloud environment. The output of the device is a report containing the results of the visualization and the ability to evaluate the results in 3D using the 3D viewer. The user can access the output through the PreOp cloud environment.

Image /page/4/Figure/4 description: The image shows a diagram of a process with inputs and outputs. The inputs are MRI and EEG, which are uploaded to a cloud. The data is then processed by PreOp 3D ESI, which generates a report and a 3D viewer. The report and 3D viewer are then downloaded from the cloud as outputs.

Figure 5.1: General overview of PreOp

5. INDICATIONS FOR USE

PreOp is intended for use by a trained/qualified EEG technologist or physician on both adult and pediatric subjects at least 3 years of age for the visualization of human brain function by fusing a variety of EEG information with rendered images of an individualized head model and an individualized MRI image.

6. COMPARISON OF TECHNOLOGICAL CHARACTERISTICS WITH PREDICATE DEVICE

	New Device	Primary PredicateDevice	SecondaryPredicate Device
Device name	PreOp	GeoSource	Curry multimodalneuroimagingsoftware
510(k) number		K092844	K001781
Manufacturer	Epilog	EGI	Neurosoft Inc.
Regulation	882.1400	882.140	882.1400
DeviceClassificationName	Class II	Class II	Class II
Software onlyproduct	Yes	Yes	Yes
Computer OS	MS-windows 7	Mac OS	MS-windows 7

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

MRI visualization	Individualized MRI	Idealized MRI(average)	Idealized MRI(average) andindividualized MRI
Sourceestimationmethods:Dipole fit Linear inverse methods	NosLORETA	YessLORETA,LORETA, LAURA	YesLORETA
Forward headmodeling	Finite DifferenceModel (FDM)	Sphere, FiniteDifference Model(FDM)	Sphere, BoundaryElement Model(BEM), FiniteElement Model(FEM)

Table 5.1: Comparison of new device to predicate device

The intended use are the same, and the technological characteristics Summary of Technological are essentially the same, as those of the predicate, K092844, Characteristics GeoSource.

The PreOp device has two different modeling features that the predicate does not have.

First, the predicate device only works with idealized anatomical MR images. The usage of idealized anatomical MR images is however a special case of the more generic modeling framework of PreOp. If an idealized MRI is given to PreOp, an equivalent forward model will be constructed as in GeoSource (see section 20.1 for a comparison). The use of individualized anatomical MR images is implemented as described in the Secondary Predicate Device, K001781. There is no essential difference in technological characteristics between PreOp and the K001781 device since the Finite Element Model is equivalent to the Finite Difference Model from an individualized forward modeling point of view.

Second, a segmentation of CSF, white matter tissue and air has been added. The use of additional tissues does not raise different questions of safety and effectiveness. See Neuroimage: Clinical 5(2014) 77-83, which concludes that the PreOp modeling of CSF, white matter and air in addition to 3-compartment, scalp, skull and gray matter of the predicate device is substantially equivalent for clinically accurate ESI.

Both the The PreOp and the predicate have the same intended use. Substantial Equivalence Both devices enable visualization of human brain function by fusing a Comparison variety of EEG information with MRI image.

From the standpoint of both functionality and workflow The PreOp device is substantially equivalent to the identified predicate as follows:

Within The PreOp and its predicate, the user can provide EEG ● input data.
. The PreOp and its predicate are designed to segment EEG

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

activity and visualize EEG activity in 3D in the brain using an MR image.

The PreOp and its predicate use externally acquired medical data and user input to achieve the result.
The intended patient population is > 3 years in both devices.
Both devices use the standard 10/20 positioning system of the . electrodes and work with a distributed dipole source space.
. Both devices use the finite difference method for forward modeling and sLORETA as linear inverse source estimation method.

Verification tests were written and executed to ensure that the system is working as designed. The PreOp passed testing and was determined safe and effective for its intended use.

Performance testing data for PreOp is available in the relevant sections of the 510(k) document to support the Substantial Equivalence determination.

7. PERFORMANCE DATA

Non-Clinical testing Validation and Verification Testing carried out on the PreOp indicates that it meets its predefined product's requirements and requirements from the following product standard: AAMI/ANSI/IEC 62304:2006. Medical Device . Software - Software Life Cycle Processes Clinical Performance In order to compare the source localization performance of Testing - study 1 PreOp as compared to that of the predicate device, a retrospective data analysis of 18 epilepsy subjects aged 3 to 55 who had previously undergone resection surgery was provided. The analysis compared the source localization accuracy of the PreOp software algorithms (i.e., sLORETA with the finite difference model [FDM] using an individualized anatomical MRI) to that of the predicate algorithm (i.e., sLORETA with the finite difference model [FDM] using a idealized anatomical MRI). All subjects had previously undergone a long-term EEG registration prior to resection surgery, had operative data available that described the resected zone, and were determined to be Engel I postoperatively. The study included subjects with temporal and extratemporal resected zones. Each subject's EEG was automatically processed using the FDA approved spike detection algorithm of Persyst and spikes were then grouped according to topographic distribution and then averaged relative to the peak of the spike to increase the signal-tonoise ratio. The average of the most dominant group was used in the source estimate. The time point used in the source estimate was the peak of the spike. The data were

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

	then run through the PreOp software algorithms and thepredicate algorithm. Three experienced epileptologists wereprovided the source localization results along with summariesof the postoperative reports and asked to rate whether eachof the algorithm solutions (sLORETA with the finite differencemodel [FDM] using an idealized or individualized anatomicalMRI) were concordant on a sublobular level. The resultsdemonstrated that the proposed PreOp algorithms weresubstantially equivalent to the predicate device algorithm.
Clinical PerformanceTesting - study 2	In this study the performance was compared of spike sourcelocalization using HD-EEG recordings (128 or 256electrodes) and Low Density LD-EEG recordings (25electrodes). By evaluating the source localization results, wefound that both algorithms provided identical source locationsin 13 epileptic spikes using LD-EEG and HD-EEG recordingsin 8 patients. Only in 3 spikes the spike localization was not100% equivalent but very close to each other.
Software Verificationand Validation Testing	Validation testing involved algorithm testing which validatedthe accuracy of PreOp. The product was deemed fit forclinical use. Usability validation is part of the ClinicalPerformance data and PreOp was tested and meets therequirements of following standard:AAMI/ANSI/IEC 62366:2007, Medical devices -●Application of usability engineering to medicaldevices.
	PreOp was designed and developed as recommended byFDA's Guidance, "Guidance for the Content of PremarketSubmissions for Software Contained in Medical Device".PreOp was considered to represent "moderate" level ofconcern as it is not intended to provide recommendations fortreatment nor to provide decisive information. According toAAMI/ANSI/IEC 62304 Standard, PreOp safety classificationhas been set to Class B.

8. CONCLUSION

The information discussed above and provided in the 510(k) submission demonstrate that the PreOp device is substantially equivalent to the predicate.

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).