MD.ai Viewer is a software-based viewer intended to be used with off-the-shelf hardware for the display of DICOM and non-DICOM medical images and other healthcare data to aid in diagnosis for healthcare professionals. It performs operations relating to the transfer, storage, display, and measurement of image data.

MD.ai Viewer allows users to perform image manipulations, including window/level, rotation, measurement and markup. MD.ai Viewer provides 2D display, Multi-Planar Reformatting and 3D visualization of medical image data.

Mobile usage is for reference and referral only.

MD.ai Viewer is not intended for primary mammography interpretation.

Device Description

MD.ai Viewer is a software-based medical image viewer used with off-the-shelf workstation and web browsers for the 2D & 3D visualization of DICOM and non-DICOM medical images. MD.ai Viewer is intended for storage, display, manipulation, measurement and processing of radiological data, including images, reports and other clinical information. It has the following primary features and functions

. Zero-footprint HTML5 medical image upload, transfer and display of medical images between facilities
Easy access to images for all participants in the healthcare process, including radiologists, technologists, physicians, nurses and other patient care practitioners
Serve as information and data management system for for DICOM and non-DICOM medical images
. Tools for image manipulation, annotation and measurement.
. Metadata information and orientation labels display
. Advanced image manipulation functions like view synchronization across series, 3D visualization like MIP and MPR
. Advanced image processing filters like histogram equalization (CLAHE) filter to aid in visualization of pathological features in the images
. Encrypted transmission of medical images through secured networks
. Encrypted storage of medical images
. HIPAA-compliant data management, including centralized storage of user activities via audit trails.
. Management of users, roles, and permissions

MD.ai Viewer consists of configurable software-only modules that display and process digital medical images, and associated medical information to aid in the day-to-day operations and workflow of clinicians and healthcare practitioners. The web browser based medical image viewer serves as the frontend module which users interact with in viewing the imaging data. The backend module handles the connection and processing of data from a variety of sources within the health system, in view of preparing visualizations to be rendered by the viewer.

MD.ai Viewer can connect and access the medical images across different sources in a health system: an existing PACS or VNA, cloud storage or local server-based storage. Users can also upload images securely into MD.ai Viewer which can be shared and enables collaboration with other users. The data connection and imaging data processing is handled by MD.ai Viewer backend module which supports the standardized transmission protocol as defined in the DICOM standard. In situation where secure network link is not available between health system and MD.ai cloud instance, the MD.ai Viewer proxy server can provide a secure and encrypted transfer of imaging data.

Users interact with MD.ai Viewer through a standard web browser, thus providing access to full quality images from anywhere and supporting a greater efficiency for care. MD.ai Viewer utilizes authorization and authentication mechanisms that enforces authorized users to access the imaging data. The system extends beyond the hospital and its internal network. With proper authorization, MD.ai Viewer can be accessed by clinical users outside of the hospital network. This way referring physicians can easily call up the imaging data of their patients or external expert accessing the imaging data for additional opinion.

MD.ai Viewer provides end-users with the ability for industry standard features such as Window/ Level, Image Flip and Rotate, Invert, Hanging Protocol, Image Measurements, and Keyboard/Mouse shortcuts. Images are initially displayed in the 2D view mode, but with the ability to toggle into advanced viewing mode of 3D/MPR for relevant exam type. It supports processing and displaying Multiplanar Reconstruction (MPR) and different intensity rendering modes based on user-defined slab thickness. It also provides image processing filters like histogram equalization (CLAHE) filter to better visualize pathological features when displaying low contrast images from some modality devices.

MD.ai Viewer provides an image rendering mechanism that preloads lower resolution images during image scrolling to improve interactivity and performance for users operating in lower network bandwidth while the full quality image is loaded in the background.

The use of a secure data transmission protocol and data encryption ensure high data security for data management via the Internet. MD.ai Viewer tracks user activity via audit trails and stores the audit data on the centralized server

AI/ML Overview

This document does not contain the detailed information necessary to complete all parts of your requestregarding acceptance criteria and a specific study proving the device meets them. The provided text is a 510(k) summary for the MD.ai Viewer, which focuses on demonstrating substantial equivalence to a predicate device rather than presenting a detailed performance study with acceptance criteria.

Here's what can be extracted and what information is missing:

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

The document does not provide a formal table of acceptance criteria with corresponding performance statistics. Instead, it describes general functionalities and features, indicating that "the design requirements were successfully met" and "Intended use and user needs were successfully validated." It also states that "The measurement features of MD.ai Viewer were validated using Digital Reference Objects and comparison with the reference device - K202335." However, specific numerical acceptance criteria (e.g., accuracy within X% for measurements) and the measured performance are not detailed.

Here's a table based on the functional comparisons and statements, but it lacks specific quantitative acceptance criteria and detailed reported performance metrics:

Acceptance Criteria (Inferred Functionality)	Reported Device Performance
Display of DICOM and non-DICOM medical images	Supported
Image transfer, storage, display, measurement	Supported
Image manipulations (window/level, rotation, measurement, markup)	Supported
2D display, MPR, 3D visualization	Supported
Mobile usage for reference and referral only	Supported
Not for primary mammography interpretation	Supported
Zero-footprint HTML5 browser-based viewer	Supported
DICOM communication protocol	Supported
Support for key modalities	CR, CT, DX, IVOCT, MR, MG, NM, OCT, OT, PT, RF, SC, US, XA supported
Standard image manipulation tools (window/level, rotate/pan/zoom, etc.)	Supported
Multi-study viewing, Image Export, Image Sharing	Supported
Metadata Display/Hide, Orientation Labels, Keyboard Shortcuts	Supported
Measurements, Annotations	Supported
Full Screen Mode, Multi-monitor, Layouts	Supported
Linking Series, Image Scrolling, Linked Scrolling, Reference Lines	Supported
Multiplanar reformat (MPR)	Supported
Maximum Intensity Projection (MIP)	Supported
Sharpen, blur, emboss, edge filters	Supported
Histogram Equalization filter	Supported
Data Encryption (HTTPS)	Supported
Data Security (stored on server)	Supported
Built-in access control	Supported
Measurement features validation using Digital Reference Objects and comparison with reference device	Successfully validated (specific metrics not provided)
Non-clinical performance testing met design requirements	Successfully met

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

This information is not provided in the document. The filing mentions "non-clinical performance testing" and "Digital Reference Objects" for measurement validation, but no details on sample size, data origin, or whether the study was retrospective or prospective.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

This information is not provided in the document. No clinical study involving expert interpretation for ground truth is described.

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

This information is not provided in the document, as no expert-adjudicated test set is described.

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

A MRMC comparative effectiveness study was not done. The document explicitly states: "No clinical performance data were performed for this submission." The MD.ai Viewer is a viewer, not an AI diagnostic aid in the context of this 510(k), thus an MRMC study comparing human readers with and without AI assistance would not be directly applicable to its stated indications for use as a viewer.

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

A standalone performance study for an AI algorithm was not done. The MD.ai Viewer is a display and processing system for medical images, not an AI algorithm intended for standalone diagnostic performance.

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

For the "measurement features" validation, the ground truth was established using Digital Reference Objects. The nature of the ground truth for other "design requirements" is not specified but would generally relate to technical specifications and functional correctness rather than clinical ground truth like pathology.

8. The sample size for the training set

This is not applicable as the MD.ai Viewer, as described in this 510(k), is a medical image viewer and processing system, not an AI model that requires a training set. If it incorporates AI filters, the training of those specific filters is not detailed here.

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

This is not applicable for the same reason as point 8.

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, with the letters "FDA" in a blue square. To the right of the blue square is the text "U.S. FOOD & DRUG ADMINISTRATION" in blue.

MDAI Inc. % Leon Chen CEO 110 Wall Street NEW YORK NY 10005

Re: K223425

February 10, 2023

Trade/Device Name: MD.ai Viewer Regulation Number: 21 CFR 892.2050 Regulation Name: Medical image management and processing system Regulatory Class: Class II Product Code: LLZ Dated: November 9, 2022 Received: November 14, 2022

Dear Leon Chen:

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 801); medical device reporting of medical device-related adverse events) (21 CFR 803) for

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

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

Sincerely.

Jessica Lamb

Jessica Lamb, Ph.D. Assistant Director Imaging Software Team DHT8B: Division of Radiological Imaging Devices and Electronic Products OHT8: Office of Radiological Health Office of Product Evaluation and Quality Center for Devices and Radiological Health

Enclosure

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

Indications for Use

510(k) Number (if known) K223425

Device Name MD.ai Viewer

Indications for Use (Describe)

Mobile usage is for reference and referral only.

MD.ai Viewer is not intended for primary mammography interpretation.

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}------------------------------------------------

Image /page/3/Picture/0 description: The image shows the logo for MD.ai. The letters "M" and "D" are in a light blue color, while the ".ai" is in black. The logo is simple and modern, and the use of color makes it visually appealing.

K223425

Administrative Information

Submitter:	MDAI Inc
Submission Type:	Traditional 510(k), New device
Address:	110 Wall Street Suite 6-028, NY, NY 10005
Phone Number:	917-725-1883
Contact Person/Company Representative	Leon Chen CEO, MDAI Inc
Email:	fda@md.ai

DEVICE INFORMATION

Trade Name	MD.ai Viewer
Common Name	Medical image viewing and analysis software
Product Code	LLZ
Regulation Number	892.2050
Regulatory Class	Class 2
Regulatory Name	System, Image Processing, Radiological
Review Panel	Radiology

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

Predicate Device Information

Predicate Device Name	eUnity
Predicate Device K Number	K172490
Product Code	LLZ
Regulation Number	892.2050
Regulatory Class	Class 2
Regulatory Name	System, Image Processing, Radiological
Review Panel	Radiology

To our knowledge the predicate device has not been subject to a design related recall.

REFERENCE DEVICE

Reference Device Name	Ambra PACS including Ambra ProViewer
Predicate Device K Number	K202335
Product Code	LLZ
Regulation Number	892.2050
Regulatory Class	Class 2
Regulatory Name	System, Image Processing, Radiological
Review Panel	Radiology

Device Description

. Zero-footprint HTML5 medical image upload, transfer and display of medical images between facilities

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

MD.ai

510(k) Summary MD.ai Viewer (21 CFR 807.92)

Easy access to images for all participants in the healthcare process, including radiologists, technologists, physicians, nurses and other patient care practitioners
Serve as information and data management system for for DICOM and non-DICOM medical images
. Tools for image manipulation, annotation and measurement.
. Metadata information and orientation labels display
. Advanced image manipulation functions like view synchronization across series, 3D visualization like MIP and MPR
. Advanced image processing filters like histogram equalization (CLAHE) filter to aid in visualization of pathological features in the images
. Encrypted transmission of medical images through secured networks
. Encrypted storage of medical images
. HIPAA-compliant data management, including centralized storage of user activities via audit trails.
. Management of users, roles, and permissions

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

510(k) Summary MD.ai Viewer (21 CFR 807.92)

Image /page/6/Figure/2 description: The image shows a diagram of a system architecture for a medical imaging platform. On the left side, under the label "Users", are icons representing physicians, radiologists, project admins, and image upload, all connected to a cloud labeled "MD.ai Viewer" via HTTPS. On the right side, under the label "Health System", a proxy server is connected to PACS and modality via DICOM, and cloud storage and identity provider via HTTPS, all connected to the cloud labeled "MD.ai Viewer" via HTTPS.

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

imaging data. The system extends beyond the hospital and its internal network. With proper authorization, MD.ai Viewer can be accessed by clinical users outside of the hospital network. This way referring physicians can easily call up the imaging data of their patients or external expert accessing the imaging data for additional opinion.

SUBJECT DEVICE INDICATIONS FOR USE

Mobile usage is for reference and referral only.

MD.ai Viewer is not intended for primary mammography interpretation.

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

SUBJECT DEVICE CONTRAINDICATIONS FOR USE

MD.ai Viewer is not indicated for interpreting mammography.

Substantial Equivalence Discussion

Predicate Device Indications for Use

eUnity is a software application that displays medical image data and associated clinical reports to aid in diagnosis for healthcare professionals. It performs operations relating to the transfer, storage, display, and measurement of image data.

eUnity allows users to perform image manipulations, including window/level, rotation, measurement and markup. eUnity provides 2D display, Multi-Planar Reformatting and 3D visualization of medical image data, and mobile access to images.

eUnity displays both lossless and lossy compressed images. For lossy images, the medical professional user must determine if the level of loss is acceptable for their purposes. Display monitors or mobile devices used for reading medical images for diagnostic purposes must comply with applicable regulatory approvals and with quality control requirements for their use and maintenance. For mobile diagnostic usage when a full workstation is not available.

Mobile usage for mammography is for reference and referral only.

1.1. Subject Device Indications for Use

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

Mobile usage is for reference and referral only. MD.ai Viewer is not intended for primary mammography interpretation.

Indications for Use Equivalence Discussion 1.2.

The Predicate Device (eUnity) and Subject Device (MD.ai Viewer) are both software-based viewer designed to receive, display, and measure image data and indicated for use to aid in diagnosis for trained healthcare professionals.

Both devices shared common technological characteristics. Both are intended as a zero-download, zero-footprint browser-based viewer. By leveraging the industry standard browser-based technology, both devices enable image data display with off-the-shelf hardware.

Both devices are capable of displaying image data in DICOM format from imaging modalities that are standard in the provision of care as well as non-DICOM image data captured in widely used JPEG or PNG format.

Further comparison between the two devices as performed in Device Comparison Table.

The indications for use of the Predicate Device and Subject Device are substantially equivalent.

Characteristic	SubjectDevice:MD.ai CloudPlatform	PrimaryPredicate:ClientOutlook eUnity	Comment
User Install Requirements	Thin Client - noinstall, runs withinbrowser	Thin Client - no install,runs within browser	Equivalent
Communication	DICOM, non-DICOM	DICOM, non-DICOM,IHE	The Subject Device supports the DICOMcommunication protocol used within theIHE framework. Similar to the PredicateDevice, it fulfills the role of Image Displaydescribed in the IHE Radiology IntegrationProfile.
Modalities	CR, CT, DX, IVOCT,MR, MG, NM, OCT,OT, PT, RF, SC, US,XA	CR, CT, DX, ECG, MR,MG, NM, OP, PR, PT, RF,SC, SR, US, XA, XL	Key modalities used in provision of careare supported.
Window Level,Rotate/Pan/Zoom, Reset,Presets, Invert	Yes	Yes	Equivalent
Multi-study viewing, ImageExport, Image Sharingcompliant	Yes	Yes	Equivalent
Metadata Display/Hide	Yes	Yes	Equivalent
Orientation Labels,Keyboard Shortcuts	Yes	Yes	Equivalent
Measurements,Annotations	Yes	Yes	Equivalent
Full Screen Mode,Multi-monitor, Layouts	Yes	Yes	Equivalent
Linking Series, ImageScrolling, Linked Scrolling,Reference Lines	Yes	Yes	Equivalent
GSPS, KIN	No	Yes	Subject Device manages the presentationstate as an internal record that allowsIntended users to interact with thepresentation state within the device. TheSubject Device currently does not supportthe export of presentation state in GSPSand KIN. The difference does not impactthe equivalence for the Predicate andSubject Device.
Multiplanar reformat (MPR)	Yes	Yes	Equivalent
Maximum IntensityProjection (MIP)	Yes	Yes	Equivalent
Oblique, VolumeRendering, Opacity Presets,Scalpel tool, bone removal	No	Yes	The Predicate Device provides additionalcapabilities in 3D advanced visualizationwhich the Subject Device does not have.The difference does not impact theequivalence for the Predicate and SubjectDevice.
Sharpen, blur, emboss,edge filters	Yes	No	These filters are part of the imagemanipulation tool included in the SubjectDevice which the users could use whendisplaying the image data. The differencedoes not impact the equivalence for thePredicate and Subject Device.
Histogram Equalizationfilter	Yes	No	These filters are part of the imagemanipulation tool included in the SubjectDevice which the intended users coulduse when displaying the image data. Thedifference does not impact theequivalence for the Predicate and SubjectDevice.
Data Encryption	HTTPS	HTTPS	Equivalent
Data Security	Stored on server	Stored on server	Equivalent
Built-in accesscontrol or parentapplication accesscontrol	Built-in access controlor parent applicationcontrol	Equivalent	Equivalent

Device Comparison Table

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

MD.ai

510(k) Summary MD.ai Viewer (21 CFR 807.92)

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

Non-Clinical Performance Testing

The MD.ai Viewer was subjected to non-clinical performance testing which verified that the design requirements were successfully met. Intended use and user needs were successfully validated. The measurement features of MD.ai Viewer were validated using Digital Reference Objects and comparison with the reference device - K202335.

As the intended use, functionality and performance of the subject MD.ai Viewer and predicate eUnity device are equivalent, the result of the non-clinical performance testing is evidence that the MD.ai Viewer performs in an equivalent manner to the eUnity device.

Clinical Performance Data Equivalence Discussion

No clinical performance data were performed for this submission.

Animal Studies Data

No animal studies were performed for this submission.

Conclusion

Based on the comparisons and analyses detailed above in this summary, we believe that the information and performance test reports of the MD.ai Viewer provided in this Traditional

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

Image /page/12/Picture/0 description: The image shows the logo for MD.ai. The letters "MD" are in a light blue color, while the ".ai" is in black. The letters are simple and modern.

510(k) submission are sufficient to demonstrate the safety and effectiveness as compared to the eUnity (K172490) predicate device.

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).