AI Screening for Diabetic Eye Disease May Save Time and Money

MedicalResearch.com Interview with:

Prof. Yogesan Kanagasingam, PhD Australian of the Year 2015 (WA Finalist) Research Director, Australian e-Health Research Centre Visiting Scholar,  Harvard University Adjunct Professor, School of Medicine University of Notre Dame

Prof. Kanagasingam

Prof. Yogesan Kanagasingam, PhD
Australian of the Year 2015 (WA Finalist)
Research Director, Australian e-Health Research Centre
Visiting Scholar,  Harvard University
Adjunct Professor, School of Medicine
University of Notre Dame

MedicalResearch.com: What is the background for this study? What are the main findings?

Response: We wanted to evaluate how an artificial intelligence (AI)–based grading system for diabetic retinopathy will perform in a real-world clinical setting, at a primary care clinic. 

MedicalResearch.com: What should readers take away from your report?

Response: Sensitivity and specificity of the AI system compared with the gold standard of ophthalmologist evaluation is provided.

The results demonstrate both the potential and the challenges of using AI systems to identify diabetic retinopathy in clinical practice. Key challenges include the low incidence rate of disease and the related high false-positive rate as well as poor image quality.

MedicalResearch.com: What recommendations do you have for future research as a result of this work?

Response: Low incidence rate of disease is an issue. May be a controlled environment, e.g. endocrinology clinic, may overcome this low incidence rate of diseases and high number of patients with diabetes.

Another research direction is how to improve image quality when capturing retinal images from a fundus camera.

How to overcome the issues related to sheen reflection is another research direction.  

MedicalResearch.com: Is there anything else you would like to add?

Response: At present, ophthalmologists or optometrists read all images.

If AI is introduced for image reading then, based the results from this study, ophthalmologists have to check only 8% of the images. This is a huge cost savings to the health system and save lot of time.

The accuracy rate (sensitivity and specificity) from this study is better than human graders.

Citation: 

Kanagasingam Y, Xiao D, Vignarajan J, Preetham A, Tay-Kearney M, Mehrotra A. Evaluation of Artificial Intelligence–Based Grading of Diabetic Retinopathy in Primary Care. JAMA Network Open. 2018;1(5):e182665. doi:10.1001/jamanetworkopen.2018.2665

Oct 6, 2018 @ 12:17 pm

The information on MedicalResearch.com is provided for educational purposes only, and is in no way intended to diagnose, cure, or treat any medical or other condition. Always seek the advice of your physician or other qualified health and ask your doctor any questions you may have regarding a medical condition. In addition to all other limitations and disclaimers in this agreement, service provider and its third party providers disclaim any liability or loss in connection with the content provided on this website.

 

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.