DrAid for Radiology v1

K221241 · Vinbrain Joint Stock Company · QFM · Sep 1, 2022 · Radiology

Device Facts

• Record ID: K221241
• Device Name: DrAid for Radiology v1
• Applicant: Vinbrain Joint Stock Company
• Product Code: QFM · Radiology
• Decision Date: Sep 1, 2022
• Decision: SESE
• Submission Type: Traditional
• Regulation: 21 CFR 892.2080
• Device Class: Class 2
• Attributes: AI/ML, Software as a Medical Device

Intended Use

The DrAid™ for Radiology v1 is a radiological computer-assisted triage & notification software to aid the clinical assessment of adult Chest X-Ray cases with features suggestive of pneumothorax in medical care environment. DrAid™ analyzes cases using an artificial intelligence algorithm to features suggestive of suspected findings. It makes case-level output available to a PACS for worklist prioritization or triage. As a passive notification for prioritization-only software tool with standard of care workflow, DrAid™ does not send a proactive alert directly to appropriately trained medical specialists. DrAid™ is not intended to direct attention to specific portions or anomalies of an image. Its results are not intended to be used on a stand-alone basis for clinical decisionmaking nor is it intended to rule out pneumothorax or otherwise preclude clinical assessment of X-Ray cases.

Device Story

DrAid for Radiology v1 is a radiological computer-assisted triage and notification software. It receives frontal chest X-ray images in DICOM format from PACS or imaging equipment. The device uses an AI algorithm to analyze images for features suggestive of pneumothorax. It outputs a case-level notification (flag or blank) to the PACS worklist to assist radiologists in prioritizing review of critical cases. It operates in parallel to standard clinical workflows; it does not remove cases from the queue or provide diagnostic information. Used in medical care environments by radiologists. The device aims to improve workload prioritization; it does not replace clinical assessment or rule out pneumothorax.

Clinical Evidence

Bench testing only. Two pivotal studies evaluated performance using 850 total chest X-ray cases (565 NIH dataset, 285 Vietnamese hospital dataset). Primary endpoints: sensitivity, specificity, and AUC. Aggregate results: Sensitivity 94.61% (95% CI: 92.16-96.76%), Specificity 97.58% (95% CI: 96.36-98.65%), AUC 96.10% (95% CI: 94.73-97.30%). Results demonstrate substantial equivalence to predicate performance.

Technological Characteristics

Software-based radiological computer-aided triage and notification system. Inputs: DICOM frontal chest X-rays. Processing: AI-based algorithm for pneumothorax detection. Outputs: Case-level notification to PACS. Deployment: On-premise or cloud-based. Connectivity: Integrates with PACS/workstations via API. Software: AI algorithm for automated analysis.

Indications for Use

Indicated for adult patients undergoing Chest X-Ray imaging to aid clinical assessment of pneumothorax. Not for stand-alone diagnostic use; not intended to rule out pneumothorax or direct attention to specific image anomalies.

Regulatory Classification

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.

Special Controls

Radiological computer aided triage and notification software must comply with the following special controls: 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. Standalone 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. 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.

*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.

Predicate Devices

• HealthPNX (K190362)

Submission Summary (Full Text)

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

Image /page/0/Picture/0 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 acronym with the full name of the agency on the right. The FDA part of the logo is in blue, with the acronym in a square and the full name written out to the right of the square.

VinBrain Joint Stock Company % Nguyen Linh Product Manager No 7 Bang Lang 1 Street, Vinhomes Riverside Ecological Urban Area Viet Hung Ward, Long Bien District, Ha Noi VIETNAM

# Re: K221241

September 1, 2022

Trade/Device Name: DrAid for Radiology v1 Regulation Number: 21 CFR 892.2080 Regulation Name: Radiological computer aided triage and notification software Regulatory Class: Class II Product Code: QFM Dated: July 22, 2022 Received: July 25, 2022

# Dear Nguyen Linh:

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 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 ([e-mail]) or phone [phone] or 301-796-7100).

Sincerely,

Jessica Lamb, Ph.D. Assistant Director Imaging Software Team DHT 8B: Division of Radiological Imaging Devices and Electronic Products OHT 8: 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) K221241

Device Name DrAid™ for Radiology v1

#### Indications for Use (Describe)

The DrAid™ for Radiology v1 is a radiological computer-assisted triage & notification software to aid the clinical assessment of adult Chest X-Ray cases with features suggestive of pneumothorax in medical care environment. DrAid™ analyzes cases using an artificial intelligence algorithm to features suggestive of suspected findings. It makes case-level output available to a PACS for worklist prioritization or triage.

As a passive notification for prioritization-only software tool with standard of care workflow, DrAid™ does not send a proactive alert directly to appropriately trained medical specialists. DrAid™ is not intended to direct attention to specific portions or anomalies of an image. Its results are not intended to be used on a stand-alone basis for clinical decisionmaking nor is it intended to rule out pneumothorax or otherwise preclude clinical assessment of X-Ray cases.

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

X Prescription Use (Part 21 CFR 801 Subpart D)

Over-The-Counter Use (21 CFR 801 Subpart C)

# CONTINUE ON A SEPARATE PAGE IF NEEDED.

This section applies only to requirements of the Paperwork Reduction Act of 1995.

# *DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.*

The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:

> Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff(@fda.hhs.gov

"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."

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

Traditional 510(k) DrAid™ for Radiology v1 Appendix F: 510(k) Summary

K221241

Image /page/3/Picture/2 description: The image shows the logo for VinBrain. The logo features the word "VINBRAIN" in a bold, sans-serif font, with "VIN" in red and "BRAIN" in a darker red. Above and to the right of the text is a stylized graphic of a brain, represented by a network of interconnected dots and lines, resembling a neural network or a complex system.

# 510(k) Summary DrAid™ for Radiology v1

| Name and Address of Applicant: | VinBrain Joint Stock Company<br>No 7 Bang Lang 1 Street,<br>Vinhomes Riverside Ecological Urban Area,<br>Viet Hung Ward, Long Bien District,<br>Ha Noi, Vietnam |
|----------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------|
| Date of Submission: | April 29, 2022 |
| Device Name: | DrAid™ for Radiology v1 |
| Product Code: | QFM |
| Classification Name:<br>software | Radiological computer aided triage and notification |
| Regulation Number: | 892.2080 |
| Classification: | Class II |
| Classification Panel: | Radiology |

# Indications for Use:

The DrAid™ for Radiology v1 is a radiological computer-assisted triage & notification software product designed to aid the clinical assessment of adult Chest X-Ray cases with features suggestive of pneumothorax in medical care environment. DrAid™ analyzes cases using an artificial intelligence algorithm to features suggestive of suspected findings. It makes case-level output available to a PACS for worklist prioritization or triage.

As a passive notification for prioritization-only software tool with standard of care workflow, DrAid™ does not send a proactive alert directly to appropriately trained medical specialists. DrAid™ is not intended to direct attention to specific portions or anomalies of an image. Its results are not intended to be used on a stand-alone basis for clinical decision-making nor is it intended to rule out pneumothorax or otherwise preclude clinical assessment of X-Ray cases.

#### Device Description:

DrAid™ for Radiology v1 (hereafter called DrAid™ or DrAid) is a a radiological computer-assisted triage & notification software product that automatically identifies suspected pneumothorax on frontal chest x-ray images and notifies PACS of the presence of pneumothorax in the scan. This notification enables prioritized review by the appropriately trained medical specialists who are qualified to interpret chest radiographs. The software does not alter the order or remove cases from the reading queue. The device's aim is to aid in the prioritization and triage of radiological medical images only.

Chest radiographs are automatically received from the user's image storage system (e.g. Picture Archiving and Communication System (PACS)) or other radiological imaging equipment (e.g. Xray systems) and processed by DrAid™ for analysis. Following receipt of chest radiographs, the software device de-identifies a copy of each chest radiographs in DICOM format (.dcm) and automatically analyzes each image to identify features suggestive of pneumothorax. Based on the analysis result, the software notifies PACS/workstation for the presence of Pneumothorax as

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

Traditional 510(k) DrAid™ for Radiology v1 Appendix F: 510(k) Summary

Image /page/4/Picture/1 description: The image shows the logo for VinBrain. The logo consists of the word "VINBRAIN" in red, with the "VIN" portion being larger and bolder than the "BRAIN" portion. Above and to the right of the wordmark is a stylized brain graphic made up of interconnected lines and dots, resembling a neural network or a connected system. The overall design is modern and suggests a focus on technology and artificial intelligence.

indicating either "flag" or "(blank)". This would allow the appropriately trained medical specialists to group suspicious exams together that may potentially benefit for their prioritization. Chest radiographs without an identified anomaly are placed in the worklist for routine review, which is the current standard of care.

The DrAid™ device works in parallel to and in conjunction with the standard care of workflow. After a chest x-ray has been performed, a copy of the study is automatically retrieved and processed by the DrAid™ device, therefore, the analysis result can also be provided in the form of DICOM files containing information on the presence of suspicious Pneumothorax. In parallel, the algorithms produce an on-device notification indicating which cases were prioritized by DrAid™ in PACS. The on-device notification does not provide any diagnostic information and it is not intended to inform any clinical decision, prioritization, or action to who are qualified to interpret chest radiographs. It is meant as a tool to assist in improving workload prioritization of critical cases. The final diagnosis is provided by the radiologist after reviewing the scan itself.

The following modules compose the DrAid™:

Data input and validation: Following retrieval of a study, the validation feature assessed the input data (e.g. age, modality, view) to ensure compatibility for processing by the algorithm.

AI algorithm: Once a study has been validated, the AI algorithm analyzes the frontal chest x-ray for detection of suspected pneumothorax.

API Cognitive service: The study analysis and the results of a successful study analysis are provided through an API service, to then be sent to the PACS for triaging & notification.

Error codes feature: In the case of a study failure during data validation or the analysis by the algorithm, an error is provided to the system.

# Predicate Device:

DrAid™ for Radiology v1 is substantially equivalent to the HealthPNX (K190362) for Pneumothorax.

# Substantial Equivalence Comparison:

A comparison of the subject and predicate device is provided in the table below.

| | Subject | Predicate | Comparison |
|----------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------|
| 510(k) Number | Subject of submission | K190362 | Subject Device Under Review |
| Device Name | DrAidTM for Radiology v1 | HealthPNX | Subject Device Under Review |
| Manufacturer | VinBrain Joint Stock Company | Zebra Medical Vision Ltd. | Subject Device Under Review |
| Regulation Number | 892.2080, Radiological computer aided triage and notification software | 892.2080, Radiological computer aided triage and notification software | Identical |
| | Subject | Predicate | Comparison |
| Product Code | QFM, Radiological<br>Computer-Assisted<br>Prioritization Software<br>For Lesions | QFM, Radiological<br>Computer-Assisted<br>Prioritization Software For<br>Lesions | Identical |
| Target Anatomy | Chest/Lung | Chest/Lung | Identical |
| Image Modality | Frontal Chest X-ray | Frontal Chest X-ray | Identical |
| Targeted Clinical<br>Condition | Pneumothorax | Pneumothorax | Identical |
| Indications for<br>Use | The DrAidTM for<br>Radiology v1 is a<br>radiological computer-<br>assisted triage &<br>notification software<br>product designed to aid<br>the clinical assessment of<br>adult Chest X-Ray cases<br>with features suggestive<br>of pneumothorax in<br>medical care<br>environment. DrAidTM<br>analyzes cases using an<br>artificial intelligence<br>algorithm to features<br>suggestive of suspected<br>findings. It makes case-<br>level output available to a<br>PACS for worklist<br>prioritization or triage.<br><br>As a passive notification<br>for prioritization-only<br>software tool with<br>standard of care<br>workflow, DrAidTM does<br>not send a proactive alert<br>directly to appropriately<br>trained medical<br>specialists. DrAidTM is<br>not intended to direct<br>attention to specific<br>portions or anomalies of<br>an image. Its results are<br>not intended to be used<br>on a stand-alone basis for<br>clinical decision-making<br>nor is it intended to rule | The Zebra Pneumothorax<br>device is a software<br>workflow tool designed to<br>aid the clinical assessment<br>of adult Chest X-Ray cases<br>with features suggestive of<br>Pneumothorax in the<br>medical care environment.<br>HealthPNX analyzes cases<br>using an artificial<br>intelligence algorithm to<br>identify suspected findings.<br>It makes case-level output<br>available to a<br>PACS/workstation for<br>worklist prioritization or<br>triage. HealthPNX is not<br>intended to direct attention<br>to specific portions or<br>anomalies of an image. Its<br>results are not intended to<br>be used on a stand-alone<br>basis for clinical decision-<br>making nor is it intended to<br>rule out Pneumothorax or<br>otherwise preclude clinical<br>assessment of X-Ray<br>cases. | Similar; different only<br>in semantics but not<br>substance. |
| | Subject | Predicate | Comparison |
| | out pneumothorax or<br>otherwise preclude<br>clinical assessment of X-<br>Ray cases. | | |
| Notification-only,<br>parallel workflow<br>tool | Yes | Yes | Identical |
| User | Radiologist | Radiologist | Identical |
| Radiological<br>images format | DICOM | DICOM | Identical |
| Computational<br>Platform | DrAid is designed as a<br>software module that can<br>be deployed on several<br>computing and X-ray<br>imaging platforms such<br>as radiological imaging<br>equipment, PACS, On<br>Premise or On Cloud | HealthPNX is designed as<br>a software module that can<br>be deployed on PACS and<br>Standalone desktop<br>application, Zebra<br>Worklist. | Similar |
| Alert to finding | Passive notification<br>flagged for review | Passive notification<br>flagged for review | Identical |
| Independent of<br>standard of care<br>workflow | Yes; No cases are<br>removed from worklist | Yes; No cases are removed<br>from worklist | Identical |
| Artificial<br>Intelligence<br>algorithm | Yes | Yes | Identical |
| Limited to<br>analysis of<br>imaging data | Yes | Yes | Identical |
| Aids prompt<br>identification of<br>cases with<br>indicated findings | Yes | Yes | Identical |
| Where results are<br>received | PACS / Workstation | PACS / Workstation | Identical |
| Performance level<br>- Timing of<br>notification | Passive notification is<br>visible upon transfer to<br>the PACS with a delay of<br>about 3.83 minutes for<br>image transfer to the<br>cloud, computation, and<br>results transfer. | Passive notification is<br>visible upon transfer to the<br>PACS with a delay of<br>about 22.1 seconds for<br>image transfer to the cloud,<br>computation, and<br>results transfer. | Similar |
| Total Validation<br>Data | Total: 850 chest X-ray<br>cases | Total: 588 chest X-ray<br>cases | Similar |
| | Subject | Predicate | Comparison |
| | Positive Pneumothorax:<br>354 cases | Positive Pneumothorax:<br>146 cases | |
| | Negative Pneumothorax:<br>496 cases | Negative Pneumothorax:<br>442 cases | |
| Performance | AUC: 96.10% (95% CI:<br>[94.73, 97.30]) | AUC: 98.3% (95% CI:<br>[97.40, 99.02]) | Similar |
| | Sensitivity: 94.61% (95%<br>CI: [92.16, 96.76]) | Sensitivity: 93.15% (95%<br>CI: [87.76%, 96.67%]) | |
| | Specificity:97.58% (95%<br>CI: [96.36, 98.65]) | Specificity: 92.99% (95%<br>CI: [90.19%, 95.19%]) | |

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

Image /page/5/Picture/1 description: The image contains the word "VINBRAIN" in red, with the "VIN" portion being a darker shade of red than the "BRAIN" portion. To the right of the word is a stylized image of a brain, made up of interconnected red dots and lines. The brain graphic is positioned above and to the right of the word "VINBRAIN".

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

Image /page/6/Picture/1 description: The image contains the word "VINBRAIN" in a bold, red font. To the right of the word is a graphic of a brain made up of blue lines and red dots. The lines connect the dots to form a network-like structure, resembling the connections within a brain.

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

Image /page/7/Picture/1 description: The image shows the logo for VinBrain. The logo features the word "VinBrain" in red, with the "Vin" portion being larger and bolder than the "Brain" portion. Above and to the right of the text is a stylized graphic of a brain, constructed from a network of blue lines and red dots, resembling a neural network or interconnected nodes.

The Indications for Use statement between the subject and predicate devices are equivalent. In addition, there are no differences that affect the safety and effectiveness of the subject device relative to the predicate; therefore, they can be considered substantially equivalent.

# Software Verification and Validation:

Software verification and validation has been performed in accordance with software specifications and applicable performance standards through Software Development and Validation & Verification Process to ensure performance according to specifications, User Requirements and Federal Regulations and Guidance documents, and FDA Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices.

# Training Data Set and Validation Data Set Separation

The data was intentionally managed to prevent overlap between training and validation data sets. For algorithm training, data from a hospital system in Vietnam and the publicly available CheXpert data set were utilized. For algorithm validation, the NIH Public data set and an additional Vietnamese data set were utilized. The two US data sets, CheXpert and the NIH Public data set are separate. Training and validation data from Vietnam come from separate hospitals and patient identification information was checked to confirm no patient overlap between the data sets. The data sets utilized to train and validate the algorithm are completely separate. See below for further breakdown of the data sets.

# Performance Testing - Stand-Alone:

The performance of the DrAid™ for Radiology v1 device has been validated in two separate pivotal studies. The studies were conducted with chest x-ray data from the National Institute of

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

Image /page/8/Picture/1 description: The image shows the logo for VinBrain. The logo features the word "VINBRAIN" in red, with "VIN" in a bold, sans-serif font and "BRAIN" in a slightly smaller, less bold font. Above and to the right of the text is a stylized graphic of a brain, represented by a network of interconnected lines and dots, resembling a neural network or a complex system.

Health (NIH) and another from four Vietnamese hospitals.

The NIH data set was used to demonstrate the generalizability of the device to the demographics of the US population. The data set consisted of 565 radiographs with 386 negative and 179 positive pneumothorax cases. This data set was truthed by a panel of 3 US board certified radiologists. A table of the results is provided below:

| Metrics | Mean | Standard<br>Deviation | Upper 95% CI<br>bound | Lower 95% CI<br>bound |
|-------------|--------|-----------------------|-----------------------|-----------------------|
| Sensitivity | 0.9387 | 0.0180 | 0.9721 | 0.8994 |
| Specificity | 0.9947 | 0.0036 | 1.0000 | 0.9845 |
| AUC | 0.9667 | 0.0091 | 0.9834 | 0.9473 |

A summary of the NIH data set characteristics are provided in the table below:

| Characteristics | Quantity/Type |
|--------------------|-------------------------------------|
| Number of Images | 565 |
| Number of Patients | 565 |
| Male | 326 |
| Female | 239 |
| Age (22 - 35) | 102 |
| Age (35-60) | 295 |
| Age (> 60) | 168 |
| Ethnicity | Representative of the US Population |
| View Position (AP) | 380 |
| View Position (PA) | 185 |
| Scanner Type | Unknown |

Due to lack of scanner information from the NIH data set, a secondary data set from four Vietnamese hospitals (University Medical Center Hospital, Nam Dinh Lung Hospital, Hai Phong Lung Hospital, and Vinmec Hospital) was used to demonstrate the generalizability to different scanner types. This data set consisted of 285 radiographs with 110 negative and 175 positive pneumothorax cases. This data set was truthed by a panel of 3 US board certified radiologists. A table of the results is provided below:

| Metrics | Mean | Standard<br>Deviation | Upper 95% CI<br>bound | Lower 95% CI<br>bound |
|-------------|--------|-----------------------|-----------------------|-----------------------|
| Sensitivity | 0.9535 | 0.0160 | 0.9826 | 0.9186 |
| Specificity | 0.9464 | 0.0126…