Screening Diabetic Retinopathy and Age-Related Macular Degeneration with Artificial Intelligence : New Innovations for Eye Care
Ripatti, Anna-Riitta (2025)
2025
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202503043664
https://urn.fi/URN:NBN:fi:amk-202503043664
Tiivistelmä
Background: The incidence of diabetic retinopathy (DR) and age-related macular degeneration (AMD) is expected to increase worldwide significantly during the next decades. The prevention of these proceeding pathologies is essential to avoid social and economic burden. All possible means should be implemented in the field of eye care to find the first signs of DR and AMD. One choice is to use artificial intelligence (AI) systems for screening ocular pathologies.
Objective: The aim of this study was to clarify eye care specialists’ experiences and views of the potential of artificial intelligence in screening AMD and DR. The purpose was to raise awareness of AI solutions in screening AMD and DR in fundus photography.
Methods: This study was implemented as a qualitative descriptive study. The participants (n=7) were a small group of more advanced professionals of eye care or AI, and the other group consisted of optometrists who had an opportunity to test an AI system in their eye examinations for one month during the summer of 2024. The data was collected with theme interviews. Inductive content analysis was performed to the qualitative data. The reporting of the study was conducted with Consolidated Criteria for Reporting Qualitative Data (COREQ) checklist.
Results: The title of the main category formulated in this study was “Continuously developing artificial intelligence helps eye care professionals to interpretate fundus images by facilitating the workflow”. The data demonstrated that AI systems could be used as a part of eye examination for screening diabetic retinopathy and/or age-related macular degeneration. AI solutions can facilitate workflow and human resources management, patients’ referrals and follow-ups. It was seen that more of education and information is required before implementing AI solutions more broadly. All participants agreed that AI solutions will become more common in eye care. It was also revealed that there are some concerns regarding treatment paths and financial questions.
Conclusions: The usage of AI systems as a supportive mean for screening fundus images is a promising way to advance retinal analyses. The advanced professionals had the same kind of experiences and views as the optometrists testing in real-world settings. It was revealed that it is crucial to educate eye care professionals with more extensive AI training in order to utilize the potential of the entire AI systems.
Keywords: diabetic retinopathy, DR, age-related macular degeneration, AMD, artificial intelligence, AI, machine learning, deep learning, imaging, screening, scanning, fundus photography and fundus images
Objective: The aim of this study was to clarify eye care specialists’ experiences and views of the potential of artificial intelligence in screening AMD and DR. The purpose was to raise awareness of AI solutions in screening AMD and DR in fundus photography.
Methods: This study was implemented as a qualitative descriptive study. The participants (n=7) were a small group of more advanced professionals of eye care or AI, and the other group consisted of optometrists who had an opportunity to test an AI system in their eye examinations for one month during the summer of 2024. The data was collected with theme interviews. Inductive content analysis was performed to the qualitative data. The reporting of the study was conducted with Consolidated Criteria for Reporting Qualitative Data (COREQ) checklist.
Results: The title of the main category formulated in this study was “Continuously developing artificial intelligence helps eye care professionals to interpretate fundus images by facilitating the workflow”. The data demonstrated that AI systems could be used as a part of eye examination for screening diabetic retinopathy and/or age-related macular degeneration. AI solutions can facilitate workflow and human resources management, patients’ referrals and follow-ups. It was seen that more of education and information is required before implementing AI solutions more broadly. All participants agreed that AI solutions will become more common in eye care. It was also revealed that there are some concerns regarding treatment paths and financial questions.
Conclusions: The usage of AI systems as a supportive mean for screening fundus images is a promising way to advance retinal analyses. The advanced professionals had the same kind of experiences and views as the optometrists testing in real-world settings. It was revealed that it is crucial to educate eye care professionals with more extensive AI training in order to utilize the potential of the entire AI systems.
Keywords: diabetic retinopathy, DR, age-related macular degeneration, AMD, artificial intelligence, AI, machine learning, deep learning, imaging, screening, scanning, fundus photography and fundus images