Blog

The latest from Google Research

Improving the Effectiveness of Diabetic Retinopathy Models

Thursday, December 13, 2018

Posted by Rory Sayres PhD and Jonathan Krause PhD, Google AI, Healthcareour inaugural work in training deep learning modelsapply our technology to improve health outcomes in the worldannouncing a new partnerImproving Model Performance with High-quality Labelsa paperOphthalmologymicroaneurysmsMaking our Models More Transparentour latest studyOphthalmology

On the left is a fundus image graded as having proliferative (vision-threatening) DR by an adjudication panel of ophthalmologists (ground truth). On the top right is an illustration of our deep learning model’s predicted scores (“P” = proliferative, the most severe form of DR). On the bottom right is the set of grades given by physicians without assistance (“Unassisted”) and those who saw the model’s predictions (“Grades Only”).

On the left is a retinal fundus image graded as having moderate DR (“Mo”) by an adjudication panel of ophthalmologists (ground truth). On the top right is an illustration of the predicted scores (“N” = no DR, “Mi” = Mild DR, “Mo” = Moderate DR) from the model. On the bottom right is the set of scores given by physicians without assistance (“Unassisted”) and those who saw the model’s predictions (“Grades Only”).