BriefCase is a radiological computer aided triage and notification software indicated for use in the analysis of head CTA images. The device is intended to assist hospital networks and trained radiologists in workflow triage by flagging and communication of suspected positive findings of Large Vessel Occlusion (LVO) pathologies.

BriefCase uses an artificial intelligence algorithm to analyze images and highlight cases with detected findings on a standalone desktop application in parallel to the ongoing standard of care image interpretation. The user is presented with notifications for cases with suspected findings. Notifications include compressed preview images that are meant for informational purposes only and not intended for diagnostic use beyond notification. The device does not alter the original medical image and is not intended to be used as a diagnostic device.

The results of BriefCase are intended to be used in conjunction with other patient information and based on their professional judgment, to assist with triage/prioritization of medical images. Notified clinicians are responsible for viewing full images per the standard of care.

Device Description

BriefCase is a radiological computer-assisted triage and notification software device. The software system is based on an algorithm programmed component and is comprised of a standard off-the-shelf operating system, the Microsoft Windows server 2012 64bit, and additional applications, which include PostgreSQL. DICOM module and the BriefCase Image Processing Application. The device consists of the following three modules: (1) Aidoc Hospital Server (AHS) for image acquisition; (2) Aidoc Cloud Server (ACS) for image processing; and (3) Aidoc Worklist Application for workflow integration, installed on the radiologist' desktop and provides the user interface in which notifications from the BriefCase software are received.

DICOM images are received, saved, filtered and de-identified before processing. Series are processed chronologically by running an algorithm on each series to detect suspected findings and then notifications on flagged series are sent to the Worklist desktop application, thereby prompting preemptive triage and prioritization. The user may opt to filter out notifications by pathology, e.g. a chest radiologist may choose to filter out notifications on Large Vessel Occlusion (LVO) cases, and a neuro-radiologist would opt to divert Pulmonary Embolism (PE) notifications. In addition, where several medical centers are linked to a shared PACS, a user may read cases for a certain center but not for another, and thus may opt to filter out notification by center. Activating the filter does not impact the order in which notifications are presented in the Aidoc worklist application.

The Worklist Application displays the pop-up text notifications of new studies with suspected findings when they come in. Notifications are in the form of a small pop-up containing patient name, accession number and the relevant pathology (e.g., LVO). A list of all incoming cases with suspected findings is also displayed. Hovering over a notification or a case in the worklist pops up a compressed, small black and white, unmarked image that is captioned "not for diagnostic use" and is displayed as a preview function. This compressed preview is meant for informational purposes only, does not contain any marking of the findings, and is not intended for primary diagnosis beyond notification.

Presenting the radiologist with notification facilitates earlier triage by prompting the user to assess the relevant original images in the PACS. Thus, the suspect case receives attention earlier than would have been the case in the standard of care practice alone.

AI/ML Overview

Based on the provided text, here's a detailed description of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

Metric	Acceptance Criteria (Performance Goal)	Reported Device Performance
Sensitivity	> 80%	88.8% (95% CI: 81.9%, 93.8%)
Specificity	> 80%	87.2% (95% CI: 82.5%, 91.1%)
Time-to-notification	Comparable to predicate device	3.8 min (95% CI: 3.6-4.0)
Positive Likelihood Ratio (PLR)	Not explicitly stated as an AC, but reported	6.9 (95% CI: 5.0-9.6)
Negative Likelihood Ratio (NLR)	Not explicitly stated as an AC, but reported	0.13 (95% CI: 0.1-0.2)

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

Sample Size for Test Set: 383 cases
Data Provenance: The data was collected from 3 US-based clinical sites and was retrospective.

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

The document does not explicitly state the number of experts or their specific qualifications (e.g., years of experience) used to establish the ground truth for the test set. It only mentions "reviewers" identified "True Positive cases."

4. Adjudication Method for the Test Set

The document does not explicitly state the adjudication method used for the test set.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size of Human Improvement

A Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not conducted. The study evaluated the standalone performance of the AI algorithm and compared its notification time to a predicate device. It did not assess human reader performance with and without AI assistance.

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

Yes, a standalone study was done. The pivotal study evaluated the BriefCase software's performance in identifying LVOs, reporting its sensitivity and specificity, which implies an algorithm-only evaluation. The statement "The BriefCase time-to-notification includes the time to get the DICOM exam, de-identify it, upload it to the cloud, analyze and send a notification on a positive suspect case back to the worklist application" further supports an algorithm-only performance measurement before any human interaction.

7. The Type of Ground Truth Used

The ground truth was established by "reviewers." The document implies it's an expert consensus or expert interpretation of the images, as "True Positive cases (i.e., identified as positive both by the reviewers as well as the BriefCase device)" are mentioned. It does not mention pathology or outcomes data as ground truth.

8. The Sample Size for the Training Set

The document does not specify the sample size for the training set. It only states that the device uses an "artificial intelligence algorithm trained on medical images" and an "artificial intelligence algorithm with database of images."

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

The document does not explicitly state how the ground truth for the training set was established. It only mentions that the algorithm was "trained on medical images."

Summary

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

Image /page/0/Picture/0 description: The image contains the logos of the Department of Health & Human Services and the Food and Drug Administration (FDA). The Department of Health & Human Services logo is on the left, and the FDA logo is on the right. The FDA logo includes the letters "FDA" in a blue square, followed by the words "U.S. Food & Drug Administration" in blue text.

Aidoc Medical, Ltd. % John Smith Regulatory Counsel Hogan Lovells US LLP 555 Thirteenth Street NW Washington, District of Columbia 20004 December 20, 2019

Re: K192383

Trade/Device Name: BriefCase Regulation Number: 21 CFR 892.2080 Regulation Name: Radiological Computer Aided Triage And Notification Software Regulatory Class: Class II Product Code: QAS Dated: November 22, 2019 Received: November 22, 2019

Dear John Smith:

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,

For

Thalia T. Mills, Ph.D. Director 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) K192383

Device Name

BriefCase

Indications for Use (Describe)

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

മ Prescription Use (Part 21 CFR 801 Subpart D) Ll 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) Summarv Aidoc Medical, Ltd.'s BriefCase (K192383)

Submitter:

Aidoc Medical, Ltd.3 Aminadav St.Tel-Aviv, IsraelPhone:	+972-73-7946870
Contact Person:	N. Epstein, Ph.D.
Date Prepared:	December 16, 2019
Name of Device:	BriefCase
Classification Name:	Radiological computer-assisted triage and notification softwaredevice
Regulatory Class:	Class II
Product Code:	QAS (21 C.F.R. 892.2080)
Primary Predicate Device:	BriefCase (K180647, for ICH triage)
Secondary Predicate Device:	ContaCT (DEN170073)

Device Description

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

diagnostic use" and is displayed as a preview function. This compressed preview is meant for informational purposes only, does not contain any marking of the findings, and is not intended for primary diagnosis beyond notification.

Intended Use / Indications for Use

Comparison of Technological Characteristics

The subject BriefCase for LVO triage and primary predicate BriefCase for ICH triage (K180647) are identical in all aspects and differ only with respect to the algorithm on LVO and ICH images, respectively. The addition of a notification filter feature to the software platform design is considered a minor change which does not alter the safety and efficacy profile of the device.

Both devices are radiological computer-aided triage and notification software programs. Both devices are artificial intelligence algorithms incorporated software packages for use with CT scanners, PACS, and radiology workstations. Both devices are intended to aid in triage and prioritization of radiological images and utilize the same design of deep learning algorithm trained on medical images. They differ only with regard to the training on image type, specifically, the predicate device processes head CTs and is indicated for intracranial hemorrhage triage, while the subject device processes head CTA images and is indicated for Large Vessel Occlusion (LVO) triage. Both devices are intended to provide radiologists with notifications and unannotated low-quality preview images of suspect studies for the purpose of preemptive triage.

Both software devices notify the attending radiologist of the availability of time sensitive radiological images for review based on computer aided image analysis. Both devices send notifications within the range of several minutes from the time a case comes in for processing as well as compressed previews to the radiology workstations' desktop. Notifications are meant to prompt the radiologist to start preemptive triage of a flagged case, upon which he may decide after observing the unannotated, low quality preview on his desktop (captioned 'not for diagnostic use''), to turn to the local PACS to perform evaluation of the original series earlier than would have been the case without BriefCase.

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

Thus, the subject and primary predicate BriefCase raise the same types of safety and effectiveness questions, namely, accurate detection of findings within the processed study. It is important to note that, like the predicate, the subject device does not remove cases from the standard of care reading queue. Both devices operate in parallel with the standard of care, which remains the default option for all cases.

In addition, it is also important to note that the proposed indications for use are substantially equivalent to a secondary predicate, the ContaCT device (DEN170073), which is cleared for LVO triage. Both devices send triage notifications as well as compressed preview images within the range of several minutes from the time a case comes in for processing. A table comparing the key features of the subject and the predicate devices is provided below.

The ContaCT differs from the subject device regarding the party receiving the notifications. While the ContaCT sends notifications on time-sensitive cases with a critical finding to a neurovascular specialist, the subject device sends them to the radiologist who then contacts a specialist (neurologist, neuro-interventionalist or any other clinician capable of treating stroke patients). Yet, there is clinical benefit in the subject workflow similar to alerting a radiologist of an ICH (or an LVO in this case). Thus, it is relevant to report time-to-notification to alert a radiologist, as well as estimate the time-to-open, based on the alert and worst-case assumption for responding. In addition, the subject device features a notification filter that enables the user to refrain from receiving certain notifications, per pathology or per center.

Lastly, again, all three devices do not remove cases from the standard reading queue and are operating in parallel to the standard of care, which remains the default option for all cases.

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

	Table 1. Key feature comparison
	Subject Device	Primary Predicate Device	Secondary Predicate Device
	Aidoc Briefcase for LVO triage	Aidoc Briefcase for ICH triage	Viz.Al ContaCT Software
		(K180647)	(DEN170073)
Intended Use /Indications forUse	BriefCase is a radiological computeraided triage and notification softwareindicated for use in the analysis ofhead CTA images. The device isintended to assist hospital networksand trained radiologists in workflowtriage by flagging andcommunication of suspected positivefindings of Large Vessel Occlusion(LVO) pathologies.BriefCase uses an artificialintelligence algorithm to analyzeimages and highlight cases withdetected findings on a standalonedesktop application in parallel to theongoing standard of care imageinterpretation. The user is presentedwith notifications for cases withsuspected findings. Notificationsinclude compressed preview imagesthat are meant for informationalpurposes only and not intended fordiagnostic use beyond notification.The device does not alter the originalmedical image and is not intended tobe used as a diagnostic device.The results of BriefCase areintended to be used in conjunctionwith other patient information andbased on their professionaljudgment to assist with	BriefCase is a radiological computeraided triage and notification softwareindicated for use in the analysis ofnon-enhanced head CT images.The device is intended toassist hospital networks and trainedradiologists in workflow triageby flagging and communication ofsuspected positive findings ofpathologies in head CT images,namely Intracranial Hemorrhage(ICH).BriefCase uses an artificial intelligencealgorithm to analyze images andhighlight cases with detected ICH on astandalone desktop application inparallel to the ongoing standard ofcare image interpretation. The user ispresented with notifications for caseswith suspected ICH findings.Notifications include compressedpreview images that aremeant for informational purposes onlyand not intended for diagnostic usebeyond notification. The device doesnot alter the original medical imageand is not intended to be used as adiagnostic device.The results of BriefCase are intendedto be used in conjunction with otherpatient information and based on	ContaCT is a notification-only, parallelworkflow tool for use by hospitalnetworks and trained clinicians toidentify and communicate images ofspecific patients to a specialist,independent of standard of careworkflow.ContaCT uses an artificial intelligencealgorithm to analyze images forfindings suggestive of a pre-specifiedclinical condition and to notify anappropriate medical specialist of thesefindings in parallel to standard of careimage interpretation. Identification ofsuspected findings is not for diagnosticuse beyond notification. Specifically,the device analyzes CT angiogramimages of the brain acquired in theacute setting and sends notificationsto a neurovascular specialist that asuspected large vessel occlusion hasbeen identified and recommendsreview of those images. Images canbe previewed through a mobileapplication.Images that are previewed through themobile application are compressedand are for informational purposesonly and not intended for diagnosticuse beyond notification. Notifiedclinicians are responsible for viewing
	triage/prioritization of medicalimages. Notified clinicians areresponsible for viewing full imagesper the standard of care.	professional judgment, to assist withtriage/prioritization of medical images.Notified clinicians are responsible forviewing full images per the standard of non-compressed images on adiagnostic viewer and engaging inappropriate patient evaluation andrelevant discussion with a treatingphysician before making care-relateddecisions or requests. ContaCT islimited to analysis of imaging data andshould not be used in-lieu of fullpatient evaluation or relied upon tomake or confirm diagnosis.
User population	Radiologist	Radiologist	Clinician (e.g., neurovascularspecialist)
Anatomicalregion ofinterest	Head	Head	Head
Inclusion/Exclusioncriteria	Inclusion criteria Head CTA protocol with a 64-slice scanner or higher; Scans performed onadults/transitional adults ≥ 18years of age; Slice thickness 0.5 mm – 1.0 mm. Exclusion Criteria All scans that are technicallyinadequate, including motionartifacts, severe metal artifacts,sub-optimal bolus timing or aninadequate field of view.	Inclusion Criteria Head non-enhanced CT (NECT)with a 64-slice scanner or higher; Scans performed onadults/transitional adults ≥ 18years of age; Slice thickness 0.625 mm to 5.1mm. Exclusion Criteria All scans that are technicallyinadequate; including motionartifacts, severe metal artifacts,inadequate field of view, etc.	Inclusion Criteria: The patient is older than 22 yearsof age when presenting to thehealthcare facility; The images were from patientswho underwent a stroke protocolassessment; andHead and neck CTA. Exclusion Criteria:Series used to identify potential LVOswere axial thin slice CTAs. CTA seriesmay have been excluded because ofinsufficient technical quality. Exclusioncriteria included: Series containing metal artifacts inthe soft matter of the brain; Series that are non-axial; Series containing missing slices; Series displaying no visible
			contrast due to bad bolus timingduring the series acquisitionprocess;• Series containing inconsistent pixelspacing;• Series containing slices thickerthan 0.625mm;• Series containing improperlyordered slices (e.g. as a result ofmanual correction by an Imagingtechnician); and• Series containing an incomplete skull.
View DICOMdata	DICOM Information about thepatient, study and current image	DICOM Information about the patient,study and current image	DICOM Information about the patient,study and current image
Segmentationof region ofinterest	No; device does not mark, annotate,or direct users' attention to a specificlocation in the original image.	No; device does not mark, annotate,or direct users' attention to a specificlocation in the original image.	No; device does not mark, annotate,or direct users' attention to a specificlocation in the original image.
Algorithm	Artificial intelligence algorithm withdatabase of images.	Artificial intelligence algorithm withdatabase of images.	Artificial intelligence algorithm.
Notification/Prioritization	Yes	Yes	Yes
Preview images	Presentation of a small, compressed,black and white preview image thatis labeled "Not for diagnostic use";The device operates in parallel withthe standard of care, which remainsthe default option for all cases.	Presentation of a small, compressed,black and white preview image that islabeled "Not for diagnostic use";The device operates in parallel withthe standard of care, which remainsthe default option for all cases.	Presentation of a small, compressed,black and white preview image that islabeled "Not for diagnostic use".
Alteration oforiginal image	No	No	No
Removal ofcases fromworklist queue	No	No	No
Structure	AHS module (image acquisition);	AHS module (image acquisition);	Image Forwarding Software
- ACS module (image processing);- Aidoc Worklist application forworkflow integration (worklist andImage Viewer).Addition of minor changes in thesoftware platform, e.g. notificationfilter, which neither impacts theorder in which notification come in,nor affects the safety and efficacyprofile of the device.	- ACS module (image processing);- Aidoc Worklist application forworkflow integration (worklist andImage Viewer).	- Image Processing and AnalysisSoftware- Non-diagnostic DICOM viewingmobile application

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

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

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

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

Performance Data

Pivotal Study Summary

Aidoc conducted a retrospective, blinded, multicenter, study with the BriefCase software to evaluate the software's performance in identifying head CTA images containing Large Vessel Occlusion (LVO) in 383 cases from 3 US-based clinical sites.

Primary Endpoint

Sensitivity and specificity exceeded the 80% performance goal. Sensitivity was 88.8% (95% C1: 81.9%, 93.8 %) and specificity was 87.2% (95% CI: 82.5%, 91.1%).

Secondary Endpoint

The positive likelihood ratio (PLR) and negative likelihood ratio (NLR) are 6.9 (95% Cl: 5.0-9.6) and 0.13 (95% Cl: 0.1-0.2), respectively. The table below presents the positive value (PPV) and negative predictive value (NPV) with varying prevalence:

Prevalence	PPV	NPV
10%	43.5%	98.6%
15%	55.0%	97.8%
20%	63.4%	96.9%
25%	69.8%	95.9%
30%	74.8%	94.8%
35%	78.9%	93.5%
40%	82.2%	92.1%
45%	85.0%	90.5%
50%	87.4%	88.6%

In addition, the time-to-notification metric observed for the BriefCase software in the three medical centers was compared to the equivalent metric of the predicate devices.

The BriefCase time-to-notification includes the time to get the DICOM exam, de-identify it, upload it to the cloud, analyze and send a notification on a positive suspect case back to the worklist application.
The BriefCase time-to-notification was measured for all True Positive cases (i.e., identified as positive both by the reviewers as well as the BriefCase device) and is given in Table 2 below. The Table also displays the same metric reported for the BriefCase ICH.

The time-to-notification results obtained for the subject BriefCase device show comparability with the primary predicate with regard to time savings to the standard of care review. The BriefCase mean time-to-notification for LVO was 3.8 min (95% Cl: 3.6-4.0). The time-to-notification for the BriefCase for ICH triage was 4.46 minutes (95% CL: 4.10-4.83).

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

Parameter	N	Meanestimate	LowerConfidenceLimit	UpperConfidenceLimit
Time-to-notificationof BriefCase LVO	111	3.8	3.6	4.0
Time-to-notificationof BriefCase ICH	59	4.46	4.10	4.83

Table 2. Time-to- notification comparison for BriefCase devices

Thus, the reported similar time-to-notification data demonstrates that when using the subject BriefCase for LVO triage the radiologists may have the same benefit in time saving as with the BriefCase for ICH triage.

In summary, performance validation data, combined with a comparison of time-to-notification metric with the predicate device establish the achievement by the subject BriefCase of preemptive triage in the range of several minutes.

Conclusions

The subject BriefCase for LVO triage, the primary predicate BriefCase for ICH triage and the secondary predicate ContaCT for LVO triage are all intended to aid in prioritization and triage of radiological images for the indications of Large Vessel Occlusion and Intracranial Hemorrhage respectively. The labeling of both the subject and the predicate devices clearly states that the devices are not for diagnostic use. Both devices are software packages with similar technological characteristics and principles of operation (with minor changes in the software platform of the subject device, such as a notification filter), both incorporating AI algorithms that process images, and software to send notifications and display unannotated compressed low-quality preview images. In all three devices, the labeling instructs the user to further evaluate and diagnose based only on the original images in the local PACS.

All three devices operate in parallel to the standard of care workflow in the sense that they do not change the original image, do not provide any marking on the output preview, and do not remove images from the standard of care FIFO queue, thus not disturbing standard interpretation of the images by the attending radiologists. The subject and the primary predicate achieve performance of time-to-notification metric in the range of several minutes, and thus contribute similarly to triage and early involvement of the radiologist in evaluating suspected images of LVO and ICH respectively.

The BriefCase device for LVO triage is thus substantially equivalent to the primary predicate BriefCase for ICH triage on time-to-notification metric.

Regulation Number and Section

§ 892.2080 Radiological computer aided triage and notification software.

(a)
Identification. Radiological computer aided triage and notification software is an image processing prescription device intended to aid in prioritization and triage of radiological medical images. The device notifies a designated list of clinicians of the availability of time sensitive radiological medical images for review based on computer aided image analysis of those images performed by the device. The device does not mark, highlight, or direct users' attention to a specific location in the original image. The device does not remove cases from a reading queue. The device operates in parallel with the standard of care, which remains the default option for all cases.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Design verification and validation must include:
(i) A detailed description of the notification and triage algorithms and all underlying image analysis algorithms including, but not limited to, a detailed description of the algorithm inputs and outputs, each major component or block, how the algorithm affects or relates to clinical practice or patient care, and any algorithm limitations.
(ii) A detailed description of pre-specified performance testing protocols and dataset(s) used to assess whether the device will provide effective triage (
e.g., improved time to review of prioritized images for pre-specified clinicians).(iii) Results from performance testing that demonstrate that the device will provide effective triage. The performance assessment must be based on an appropriate measure to estimate the clinical effectiveness. The test dataset must contain sufficient numbers of cases from important cohorts (
e.g., subsets defined by clinically relevant confounders, effect modifiers, associated diseases, and subsets defined by image acquisition characteristics) such that the performance estimates and confidence intervals for these individual subsets can be characterized with the device for the intended use population and imaging equipment.(iv) Stand-alone performance testing protocols and results of the device.
(v) Appropriate software documentation (
e.g., device hazard analysis; software requirements specification document; software design specification document; traceability analysis; description of verification and validation activities including system level test protocol, pass/fail criteria, and results).(2) Labeling must include the following:
(i) A detailed description of the patient population for which the device is indicated for use;
(ii) A detailed description of the intended user and user training that addresses appropriate use protocols for the device;
(iii) Discussion of warnings, precautions, and limitations must include situations in which the device may fail or may not operate at its expected performance level (
e.g., poor image quality for certain subpopulations), as applicable;(iv) A detailed description of compatible imaging hardware, imaging protocols, and requirements for input images;
(v) Device operating instructions; and
(vi) A detailed summary of the performance testing, including: test methods, dataset characteristics, triage effectiveness (
e.g., improved time to review of prioritized images for pre-specified clinicians), diagnostic accuracy of algorithms informing triage decision, and results with associated statistical uncertainty (e.g., confidence intervals), including a summary of subanalyses on case distributions stratified by relevant confounders, such as lesion and organ characteristics, disease stages, and imaging equipment.