Search Results

The AEYE-DS is indicated for use by health care providers to automatically detect more than mild diabetic retinopathy (mtmDR) in adults diagnosed with diabetes who have not been previously diagnosed with diabetic retinopathy. The AEYE-DS is indicated for use with the Topcon NW400 camera and the Optomed Aurora camera.

AEYE-DS is a retinal diagnostic software device that incorporates an algorithm to evaluate retinal images for diagnostic screening to identify retinal diseases or conditions. Specifically, the AEYE-DS is designed to perform diagnostic screening for the condition of more-than-mild diabetic retinopathy (mtmDR).

The AEYE-DS is comprised of 5 software components: (1) Client; (2) Service; (3) Analytics; (4) Reporting and Archiving; and (5) System Security.

The AEYE-DS device is based on the principle of operation, whereby a fundus camera is used to obtain retinal images. The fundus camera is attached to a computer, where the Client module/software is installed. The Client module/software guides the user to acquire the images and enables the user to interact with the server-based analysis software over a secure internet connection. Using the Client module/software, users identify the fundus images per eye to be dispatched to the Service module/software. The Service module/software is installed on a server hosted at a secure datacenter, receives the fundus images and transfers them to the Analytics module/software. The Analytics module/software, which runs alongside the Service module/software, processes the fundus images and returns information on the image quality and the presence or absence of mtmDR to the Service module/software. The Service module/software then returns the results to the Client module/software.

Here's a breakdown of the acceptance criteria and the study that proves the AEYE-DS device meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The document primarily focuses on establishing substantial equivalence to a predicate device (AEYE-DS K221183), rather than explicitly listing pre-defined, quantitative acceptance criteria for each metric in the same way one might find in a clinical trial protocol. However, we can infer the implicitly accepted performance by comparing the subject device's results to the predicate's and demonstrating robust performance across two studies. The table below presents the key performance metrics reported for the subject device (AEYE-DS K240058 with Optomed Aurora camera) and the predicate device (AEYE-DS K221183 with Topcon NW400 camera).

Note: While PPV in Study 2 (53%) for the subject device is below the predicate's performance (60%), the document attributes this to the actual prevalence of mtmDR+ patients in the study's diabetic population (i.e., 12%), stating that the robustness of the studies is demonstrated by the similar PPV and NPV results across both studies despite this. The overall conclusion is substantial equivalence.

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

• Study 1 Sample Size: 317 subjects
• Study 2 Sample Size: 362 subjects
• Data Provenance: Both studies were prospective, multi-center, single-arm, blinded studies conducted at study sites in the United States.

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

The ground truth was established by an independent reading center. While the exact number of experts (readers) is not specified, their role in determining the severity of retinopathy and clinically significant diabetic macular edema (DME) according to the Early Treatment for Diabetic Retinopathy Study severity (ETDRS) scale implies a high level of expertise, typical of ophthalmic specialists or certified graders.

4. Adjudication Method for the Test Set

The document states that the "Reading Center diagnostic results formed the reference standard (ground truth) for the study." It does not explicitly describe an adjudication method (e.g., 2+1, 3+1) among multiple readers within the reading center. It implies a single, definitive determination by the reading center.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No multi-reader multi-case (MRMC) comparative effectiveness study was done. The studies were designed to evaluate the standalone performance of the AEYE-DS device, not to compare its performance in assisting human readers. The device is intended to "automatically detect" mtmDR.

6. Standalone (Algorithm Only) Performance

Yes, a standalone (algorithm only) performance evaluation was done. The reported sensitivity, specificity, PPV, and NPV values are for the AEYE-DS device's automated detection of mtmDR.

7. Type of Ground Truth Used

The ground truth used was expert consensus / standardized clinical assessment based on:

• Dilation four widefield color fundus images
• Lens photography for media opacity assessment
• Macular optical coherence tomography (OCT) imaging
• Severity determination according to the Early Treatment for Diabetic Retinopathy Study (ETDRS) scale by an independent reading center.

8. Sample Size for the Training Set

The document does not explicitly state the sample size for the training set. The clinical studies (Study 1 and Study 2) are described as the basis for the performance evaluation of the device (i.e., the test set performance). The training of the AI model would have occurred prior to these validation studies.

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

The document does not explicitly describe how the ground truth for the training set was established. However, it is standard practice for AI models in medical imaging to be trained on large datasets where ground truth is established by experienced clinical experts (e.g., ophthalmologists, retina specialists) thoroughly reviewing and annotating images, often with consensus protocols, similar to the method described for the test set's ground truth (ETDRS grading by a reading center). Given the device's predicate status and the detailed description of the ground truth for the test sets, it is highly probable that a rigorous, expert-based process was applied to the training data as well.

Ask a specific question about this device