Accepted for/Published in: Journal of Medical Internet Research
Date Submitted: Apr 14, 2023
Open Peer Review Period: Apr 12, 2023 - Jun 7, 2023
Date Accepted: Nov 5, 2023
(closed for review but you can still tweet)
Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Developing and evaluating an artificial intelligence-based computer-aided diagnosis system for retinal disease: A diagnostic study
ABSTRACT
Background:
Recent advancements in computer vision and deep learning techniques, deep neural networks have helped achieve expert-level performance in clinical diagnoses.
Objective:
This diagnostic study aimed to determine the usefulness of a proposed artificial intelligence (AI)-based computer-aided diagnosis (AI-CAD) system in assisting ophthalmologists with the diagnosis of retinal diseases using optical coherence tomography (OCT) images.
Methods:
For the training and evaluation of the proposed deep learning model, 1,693 OCT images were collected and annotated. The dataset included 929 and 764 cases of acute and chronic central serous chorioretinopathy, respectively. Sixty-six ophthalmologists (two groups: 36 retina and 30 non-retina specialists) participated in the observer performance test. To evaluate the deep learning algorithm used in the proposed AI-CAD system, the training, validation, and test sets were split in an 8:1:1 ratio. One hundred randomly sampled OCT images from the test set were used for the observer performance test, and the participants were instructed to select a central serous chorioretinopathy subtype for each of these images. Each image was provided under different conditions: (i) without AI support (No AI), (ii) with an AI probability score (AI prob), and (iii) with an AI probability score and visual evidence (AI Prob+Evid). The sensitivity, specificity, and area under the receiver operating characteristic curve were used to measure the diagnostic performance of the model and ophthalmologists.
Results:
The proposed system achieved a high detection performance (99.0% of the area under the curve), outperforming the 66 ophthalmologists who participated in the observer performance test. In both groups, ophthalmologists with the support of AI Prob+Evid achieved the highest mean diagnostic performance compared with that of those subjected to other conditions (No AI or AI Prob). Non-retina specialists achieved expert-level diagnostic performance with the support of the proposed AI-CAD system.
Conclusions:
Our proposed AI-CAD system improved the diagnosis of retinal disease by ophthalmologists, which may support decision-making regarding retinal disease detection and alleviate the workload of ophthalmologists.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.