The Neuro.Al Algorithm is a standalone image processing software device that can be deployed as a Microsoft Windows® executable on off-the-shelf hardware or as a containerized application (e.g., a Docker container) that runs on off-the-shelf hardware or on a cloud platform. Data and images are acquired via DICOM compliant imaging devices. DICOM results may be exported, combined with or utilized by other DICOM-compliant systems and results.

The Neuro.AI algorithm provides analysis capabilities for static, functional, dynamic and derived imaging datasets acquired with CT or MRI. It can be used for the analysis of dynamic brain image data, showing properties of changes in contrast over time. This functionality includes calculation of parameters related to brain tissue perfusion, vasular assessment and tissue blood volume and other parametric maps with or without the ventricles included in the calculation. The algorithm also includes volume reformat in various orientions, rotational MIP 3D batch while removing the skull. This "tumble view" allows qualitative review of vascular structure in direct correlation to the perfusion maps for comprehensive review.

The results of the Neuro.AI Algorithm can be delivered to the end-user through image viewers such as TeraRecon's Aquarius iNtuition system, TeraRecon's Northstar AI Results Explorer, or other image viewing systems like PACS that can support DICOM results generated by Neuro.AI.

The Neuro.AI Algorithm results are designed for use by trained healthcare professionals and are intended to assist the physician in diagnosis, who is responsible for making all final patient management decisions.

Device Description

The Neuro.Al Algorithm is a modification of the predicate device, iNtuition-TDA, TVA, Parametric Mapping which was cleared under K131447. The predicate device is an optional module/workflow for the iNtuition system (K121916). The Neuro.Al Algorithm is a standalone image processing software device that can be deployed as a Microsoft® Windows executable on off-the-shelf hardware or as a containerized application (e.g., Docker container) that runs on off-the-shelf hardware or on a cloud platform. The device has limited network connectivity or external medical support.

The Neuro.Al Algorithm allows motion correction and processes, calculates and outputs brain perfusion analysis results for static, functional, dynamic and derived imaging datasets acquired with CT or MRI. Neuro.Al results are used for visualization and analysis of dynamic brain perfusion image data, showing properties of changes in contrast over time. This functionality includes calculation of parameters related to brain tissue perfusion, vascular assessment displayed in rotational Maximum Intensity Projection (MIP) called the tumble view, and tissue blood volume and other parametric maps with or without brain ventricles included in the calculation.

Outputs include text and parametric map displays of measurements including time to peak (TTP), take off time (TOT), recirculation time (RT), mean transit time (MTT), blood volume (BV/CBV), blood flow (BF/CBF), classification maps, reformatted images and rotational MIPs for 2D and 3D visualization of brain tissues and blood vessels, and for correlation to the perfusion maps.

The results of the Neuro.Al Algorithm can be delivered to the end-user through image viewers such as TeraRecon's iNtuition system, TeraRecon's Northstar Al Results Explorer ("Northstar"), or other third-party image viewing systems like PACS that can display the DICOM results generated by Neuro.Al output does not depend on the viewing system's capabilities as the results are self-contained and the only interface is through DICOM.

When the Neuro.Al Algorithm results are used on iNtuition, all the standard features offered by iNtuition are employed such as image manipulation tools like drawing the region of interest, manual or automatic segmentation of structures, tools that support creation of a report, transmitting and storing this report in digital form, and tracking historical information about the studies analyzed by the software.

The Neuro.Al algorithm can be used by physicians to aid in the diagnosis. The software is not intended to replace the skill and judgment of a qualified medical practitioner and should only be used by individuals that have been trained in the software's function, capabilities and limitations. The device is intended to provide supporting analytical tools to a physician, to speed decision-making and to improve communication, but the physician's judgment is paramount, and it is normal practice for physicians to validate theories and treatment decisions multiple ways before proceeding with a risky course of patient management.

AI/ML Overview

The provided document describes the Neuro.AI Algorithm and its substantial equivalence to a predicate device, iNtuition-TDA, TVA, Parametric Mapping (K131447). However, it does not contain a detailed performance study with specific acceptance criteria and reported device performance in the format requested. The document focuses on regulatory compliance, outlining the device's indications for use, technological characteristics, and a general statement about software verification and validation.

Therefore, many of the requested items cannot be extracted directly from this document.

Here's a breakdown of what can and cannot be answered based on the provided text:

1. A table of acceptance criteria and the reported device performance

Not provided in the document. The text states: "During software testing, all predefined acceptance criteria for the Neuro.Al Algorithm were met and all software test cases passed." However, it does not specify what those acceptance criteria were or provide a table of performance metrics.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

Not provided in the document. The document mentions "software testing and performance evaluation" but does not detail the test set's sample size or data provenance.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

Not provided in the document. The document describes the device's intended use by "trained professionals, including but not limited to physicians, surgeons and medical clinicians" but doesn't specify how ground truth was established for testing.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

Not provided in the document.

5. 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

Not provided in the document. The document does not describe a comparative effectiveness study involving human readers with and without AI assistance. The focus is on the device's substantial equivalence to a predicate device.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, implicitly. The document describes the "Neuro.AI Algorithm as a standalone image processing software device." The testing mentioned ("software testing and performance evaluation") would inherently be evaluating the algorithm's standalone performance against its predefined acceptance criteria, even if those criteria aren't explicitly detailed. The statement "The Neuro.AI Algorithm is as safe and effective as the predicate device" implies standalone testing for functional equivalence.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

Not explicitly provided in the document. While the device assists in diagnosis, the method for establishing ground truth for testing is not described.

8. The sample size for the training set

Not provided in the document. The document describes a "510(k) summary," which focuses on demonstrating substantial equivalence to a predicate device rather than detailing AI model development specifics like training set size.

9. How the ground truth for the training set was established

Not provided in the document. Similar to the training set size, the method for establishing ground truth for training is not included in this regulatory summary.

Summary

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

November 6, 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.

TeraRecon, Inc. % Mr. Patrick Willhite Director, Quality Assurance and Regulatory Affairs 4309 Emperor Blvd., Suite 310 DURHAM NC 27703

Re: K200750

Trade/Device Name: Neuro.AI Algorithm Regulation Number: 21 CFR 892.2050 Regulation Name: Picture archiving and communications system Regulatory Class: Class II Product Code: LLZ Dated: October 26, 2020 Received: October 28, 2020

Dear Mr. Willhite:

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/cfpmp/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 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

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

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,

For

Thalia T. Mills, Ph.D. Diretor Division of Radiological Health OHT7: Office of In Vitro Diagnostics and Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

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

Indications for Use

510(k) Number (if known) K200750

Device Name Neuro.AI Algorithm

Indications for Use (Describe)

The Neuro.AI Algorithm is an algorithm for use by trained professionals, including but not limited to physicians, surgeons and medical clinicians.

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

[In accordance with 21CFR 807.92]

1. Submitter

510(k) Sponsor:	TereRecon, Inc.
Address:	4309 Emperor Blvd., Suite 310Durham, NC 27703, USA
Contact Person:	Patrick WillhiteDirector of Quality Assurance and Regulatory Affairs
Contact Information:	Email: pwillhite@terarecon.comPhone: 919.670.1539Facsimile: 650.372.1101
Date Summary Prepared:	10/30/2020

2. Device

Proprietary (Trade) Name:	Neuro.Al Algorithm (“Neuro.Al”)
Common Name:	Medical Imaging System
Classification:	§ 892.2050, Picture Archiving and CommunicationSystem.
Product Codes:	LLZ – System, Image Processing, Radiological

3. Predicate Device

Predicate Device	iNtuition-TDA, TVA and Parametric Mapping (K131447)
Reference Device	iNtuition system (K121916)

4. DEVICE DESCRIPTION

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

5. INDICATIONS FOR USE

The Neuro.Al Algorithm is an algorithm for use by trained professionals, including but not limited to physicians, surgeons and medical clinicians.

The Neuro.Al Algorithm provides analysis capabilities for statio, functional, dynamic and derived imaging datasets acquired with CT or MRI. It can be used for the analysis of dynamic brain perfusion image data, showing properties of changes in contrast over time. This functionality includes calculation of parameters related to brain tissue perfusion, vascular assessment and tissue blood volume and other parametric maps with or without the ventricles in the calculation. The algorithm also includes volume reformat in various orientations, rotational MIP 3D batch while removing the skull. This "tumble view" allows qualitative review of vascular structure in direct correlation to the perfusion maps for comprehensive review.

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

The results of the Neuro.Al Algorithm can be delivered to the end-user through image viewers such as TeraRecon's Aquarius iNtuition system. TeraRecon's Northstar Al Results Explorer, or other image viewing systems like PACS that can support DICOM results generated by Neuro.Al.

The Neuro.Al Algorithm results are designed for use by trained healthcare professionals and are intended to assist the physician in diagnosis, who is responsible for making all final patient management decisions.

6. SUMMARY OF TECHNOLOGICAL CHARACTERISTICS

The Neuro.Al Algorithm is substantially equivalent to the predicate device, iNtuition-TDA, TVA, Parametric Mapping (K131447). It has the same intended use and the same basic technological characteristics as the predicate device. The main difference between the subject and predicate device is the standalone nature of the subject device.

Both the subject and predicate device allow motion correction and processes, calculates and outputs brain perfusion analysis results for static, functional, dynamic and derived imaging datasets acquired with CT or MRI. The results are used for visualization and analysis of dynamic brain perfusion image data, showing properties of changes in contrast over time. This functionality includes calculation of parameters related to brain tissue perfusion, vascular assessment displayed in rotational Maximum Intensity Projection (MIP) called the tumble view, and tissue blood volume and other parametric maps. The subject device can also display maps with or without brain ventricles included like the reference device, iNtuition system (K121916). The reference device includes segmentation functionality where the segmentation can be displayed or hidden for any part of the body, including brain ventricles.

Both the subject and predicate devices are interoperable with CT and MR scanners, thirdparty hospital systems such as PACS, and the iNtuition platform. The subject device is a standalone software device, the results of which can also be consumed by TeraRecon's Northstar Al Results Explorer via EnvoyAl as the algorithm hosting platform or by other third-party image viewing systems that can display the DICOM results generated by the Neuro.Al Algorithm.

The differences in technological characteristics do not raise any new or different questions of safety or effectiveness. Software verification and validation testing and performance testing validate that the Neuro.Al Algorithm is as safe and effective as the predicate device to support a determination of substantial equivalence.

See the table below for a description of the technological similarities and differences among the subject, predicate, and reference devices.

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

	Subject DeviceNeuro.AI Algorithm(TBD)	Predicate DeviceiNtuition-TDA, TVA,Parametric Mapping(K131447)	Reference DeviceiNtuition system(K121916)
Areas of Use	Same and other trainedclinical users	Radiology	Radiology
Modality Type	Same	Vendor-neutral - CT, MR andother volumetric imagingmodalities. Images areexposed over time.	Vendor-neutral - CT, MR,Nuc, PET, Angio, US/Echo,SPECT, CR/DR Review
DICOM®formats	Same and DICOM 3.x	Yes, supports DICOM 3.0	Yes, supports DICOM 3.0
OperatingSystem	Same andCentOS (Interoperability)	MicrosoftWindows® executable onoff-the-shelf hardware	MicrosoftWindows® executable onoff-the-shelf hardware
Body Part	Same	Head - entire brain or fromlower edge of the base ofnucleus to upper edge of theventricles.	Head and other regions andorgans within the body
KeyFunctionality/Features	2D, 3D and 4D viewing,multi-phase seriessupport, zoom, pan,window level, rotate, cineand display layouts andtemplates	2D, 3D and 4D viewing,multi-phase seriessupport, zoom, pan,window level, rotate, cineand display layouts andtemplates	2D, 3D and 4D viewing,multi-phase seriessupport, zoom, pan,window level, rotate, cineand display layouts andtemplates
	ROI Markers: Ability tocreate preset shapes orfreehand ROI formeasurements orsegmentations	ROI Markers: Ability tocreate preset shapes orfreehand ROI formeasurements orsegmentations	ROI Markers: Ability tocreate preset shapes orfreehand ROI formeasurements orsegmentations
	Arterial and venous inputfunction selection,automatic and manual	Arterial and venous inputfunction selection,automatic and manual	Arterial and venous inputfunction selection,automatic and manual
	Ventricle segmentation		Ventricle segmentation
VentricleSegmentation	Setting allows software todisplay maps with orwithout brain ventriclesincluded	This device is a module ofiNtuition. When used withiNtuition, the segmentationtools can be applied to anypart of the body, includingbrain ventricles.	Editing and segmentationtools are providedincluding freehand crop,cut, dynamic region grow,bone removal tools, ribcage removal, tableremoval tools, and tools toprovide an initial selectionof bone or air-filled vessels(e.g. lung or colon) forremoval or improvement.Any segmentation can bedisplayed or hidden andthis is applicable for any
	Subject DeviceNeuro.Al Algorithm(TBD)	Predicate DeviceiNtuition-TDA, TVA,Parametric Mapping(K131447)	Reference DeviceiNtuition system(K121916)
			part of the body, includingbrain ventricles.
Perfusionmeasurementsand color maps	Same	Time to Peak (TTP) Take off Time (TOT orMaximum Slope ofIncrease) Recirculation Time (RT) Mean Transit Time (MTT) Blood Volume (BV/CBV) Blood Flow (BF/CBF) Perfusion Maps	Time to Peak (TTP) Take off Time (TOT orMaximum Slope ofIncrease) Recirculation Time (RT) Mean Transit Time (MTT) Blood Volume (BV/CBV) Blood Flow (BF/CBF) Perfusion Maps
Graph Displays	Same	Artery and Vein Fitted andRaw curves - time/activity	Artery and Vein Fitted andRaw curves - time/activity
Export Format	Same	DICOM format	DICOM format plus JPEG,BMP, AVI, Word
Methods forMathematicalModeling	Same	SVD	SVD
Arterial andVenous InputFunctionSelection	Same	Automatic and manual	Automatic and manual
Interoperability/Compatibility	CT and MR Scanners Third-party hospitalsystems such as a PACSserver, EMR or other iNtuition AdvancedVisualization system Algorithm dockerizationusing Docker™ hosted inTeraRecon's EnvoyAl platf	CT and MR Scanners Third-party hospitalsystems such as PACSserver, EMR or other iNtuition advancedvisualization system	CT and MR Scannersplus other imagingmodalities Third-party hospitalsystems such as PACSserver, EMR or other
Subject DeviceNeuro.Al Algorithm(TBD)	Predicate DeviceiNtuition-TDA, TVA,Parametric Mapping(K131447)	Reference DeviceiNtuition system(K121916)
Northstar Al ResultsExplorer
• Other image viewingsystems that can supportDICOM results generatedby the Neuro.Al Algorithm
• Notification systems

Table 1: Technological Characteristics comparison

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

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

7. PERFORMANCE DATA

Safety and performance of the Neuro.Al Algorithm have been verified and validated through software testing and performance evaluation. Software development and testing were performed in accordance with IEC 62304:2006/A1:2015, Medical Device Software - Software life cycle processes, utilizing a risk-based methodology. Risk has been evaluated in accordance with testing lso 14971:2007, Medical Devices - Application of Risk Management to Medical Devices. During software testing, all predefined acceptance criteria for the Neuro.Al Algorithm were met and all software test cases passed. The same verification and validation methodology, risk assessment and acceptance criterion were used for predicate device.

The results of the software and performance testing validate that the Neuro.AI Algorithm meets its qualified requirements, performs as intended, and is as safe and effective as the predicate device. No new or different questions of safety or efficacy have been raised as a result of the verification and validation process.

8. CONCLUSION

The Neuro.Al Algorithm is as safe and effective as the predicate device, iNtuition-TDA, TVA, Parametric Mapping module. The Neuro.Al Algorithm has the same intended use and the indications for use fall within the scope of that for the predicate device. Many of the technological characteristics are the same for the subject and predicate devices. Any differences in technological characteristics between the subject and predicate devices have been addressed through software verification testing and performance testing and do not raise any new or different questions of safety or effectiveness. Additionally, the differences in technological characteristics have been compared to a reference device which is currently legally marketed in the United States.

All risks were analyzed and no new risks, changes to existing risks, or new risk controls were identified as a result of the Neuro.Al Algorithm. The testing results and analysis above support a determination of Substantial Equivalence of the Neuro.Al Algorithm to the predicate device in terms of safety, efficacy, and performance.

Regulation Number and Section

§ 892.2050 Medical image management and processing system.

(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).