Machine learning can predict eyes at risk for diabetic retinopathy progression

Key Takeaways

  • Automated machine learning models may help identify eyes at risk for diabetic retinopathy progression.
  • The use of machine learning algorithms may refine the risk of disease progression and identify those at highest short-term risk, thus reducing costs and improving vision-related outcomes.

Machine learning is a branch of artificial intelligence based on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention. In a recent study, Paolo S. Silva, M.D., from Harvard University in Boston, and colleagues assessed whether automated machine learning models using ultra-widefield retinal images, which represent over 80% of the retinal surface area, could predict diabetic retinopathy (DR) progression. Their analysis included 1,179 deidentified ultra-widefield retinal images with mild or moderate nonproliferative DR (NPDR), the early phase of the disease, with three years of longitudinal follow-up.

The researchers found that for eyes with mild NPDR, sensitivity (ability to designate an individual with NPDR as having NPDR) was 0.72, specificity (ability to designate an individual without NPDR as negative) was 0.63, and accuracy (ability to correctly differentiate between individuals who do or do not have NPDR) was 64.3 percent, while for eyes with moderate NPDR, performance was 0.80, 0.72, and 73.8 percent, respectively. The validation set identified six of nine eyes (75 percent) with mild NPDR and 35 of 41 eyes (85 percent) with moderate NPDR that progressed two steps or more on the Diabetic Retinopathy Severity Scale, a standardized scale that classifies diabetic retinopathy severity. The model identified all four eyes with mild NPDR that progressed within six months and one year, as well as eight of nine (89 percent) with moderate NPDR that progressed within six months and 17 of 20 (85 percent) that progressed within one year. The results were published in JAMA Ophthalmology.

“Potentially, the use of machine learning algorithms may refine the risk of disease progression and identify those at highest short-term risk, thus reducing costs and improving vision-related outcomes,” the authors write.

Source: Lori Solomon, Medical Xpress, February 13, 2024; see source article