Revolutionary AI Model Detects Eye Disease and Predicts Parkinson's Risk from Retinal Images

Scientists have developed an AI tool called RETFound that can diagnose and predict the risk of various health conditions, including ocular diseases, heart failure, and Parkinson's disease, based on retinal images. The tool was developed using self-supervised learning, similar to large-language models like ChatGPT. By training on unlabelled retinal images, RETFound can learn to predict missing portions of images and classify them for specific conditions. The model has shown promising results in detecting ocular diseases and outperforms other AI models in predicting the risk of systemic diseases. The researchers have made the model publicly available for adaptation and training in different medical settings, but caution is needed to ensure ethical and safe usage.
Reading Insights
0
1
4 min
vs 5 min read
87%
880 → 113 words
Want the full story? Read the original article
Read on Nature.com