AI may help predict risk of disease progression in patients with diabetic retinopathy
USA: For people with diabetic eye disease, one of the most difficult and important tasks is estimating the risk of diabetic retinopathy (DR) progression. A recent study published in the journal Diabetes has shed light on dentifying the risk of the progression of diabetic retinopathy using machine learning on ultrawide field retinal images.
The study findings, presented at the 83rd American Diabetes Association Scientific Sessions (ADA 2023), demonstrated the feasibilty and accuracy of using machine learning models for identification of DR progression using ultrawide field (UWF) retinal images. The artificial intelligence prediction for about 91% of the images were either correct labels; or the labels showed greater progression than the original labels.
"The use of machine learning algorithms may further refine the risk of disease progression and personalize screening intervals that may improve vision-related outcomes and reduce costs," the researchers wrote in their study.
Current scales of diabetic retinopathy severity have informed clinicians on the progression risk and provided recommendations for treatment and follow-up. Using AI (artificial intelligence) algorithms may improve in this process. In the study, Amber Nigam, Boston, MA, Cambridge, MA, and colleagues developed and validated machine learning models for DR progression from UWF retinal images. They were labeled for baseline DR severity and progression on the basis of clinician review of the images and 3-year longitudinal follow-up using the Early Treatment Diabetic Retinopathy Study (ETDRS) severity scale.
The dataset consisted of eight classes: proliferative DR (16.55%), severe NPDR progression (11.27%) /nonprogression (10.68%), Moderate NPDR progression (10.1%) /nonprogression (15.85%), Mild nonproliferative DR (NPDR) progression (10.16%) /nonprogression (10.73%), and no DR nonprogression (14.62%). Based on 60-20-20 proportions, 9970 unique images were split into the train, the validation, and the test datasets.
Study’s key findings include:
- The ResNet model fine-tuned on this dataset has a classification test accuracy of 81% and an AUC of 0.967 on the test dataset.
- The objective of the model is to reduce false negatives, which refers to predicting a class that is less progressive than the true label.
- The predicted labels for 91% of the images were either correct labels or were the labels with greater progression than the original labels.
The researchers suggests that the use of machine learning algorithms may further refine the risk of disease progression and personalize screening intervals that may reduce costs and improve vision-related outcomes.
Reference:
AMBER NIGAM, JIE SUN, VARSHINI SUBHASH, PAOLO S. SILVA; 26-LB: Identifying the Risk of Diabetic Retinopathy Progression Using Machine Learning on Ultrawide Field Retinal Images. Diabetes 20 June 2023; 72 (Supplement_1): 26–LB. https://doi.org/10.2337/db23-26-LB
Disclaimer: This website is primarily for healthcare professionals. The content here does not replace medical advice and should not be used as medical, diagnostic, endorsement, treatment, or prescription advice. Medical science evolves rapidly, and we strive to keep our information current. If you find any discrepancies, please contact us at corrections@medicaldialogues.in. Read our Correction Policy here. Nothing here should be used as a substitute for medical advice, diagnosis, or treatment. We do not endorse any healthcare advice that contradicts a physician's guidance. Use of this site is subject to our Terms of Use, Privacy Policy, and Advertisement Policy. For more details, read our Full Disclaimer here.
NOTE: Join us in combating medical misinformation. If you encounter a questionable health, medical, or medical education claim, email us at factcheck@medicaldialogues.in for evaluation.