Google developed algorithms for AI to analyze the heart disease. From the retina.

Researchers from Google have published a paper in Nature Biomedical Engineering on the success of developing artificial intelligence algorithms. Analyzes and predicts high blood pressure and heart disease. Only scan the retina.

The method is to enter the retinal image from patients in the United States and the United Kingdom with over 280,000 images to find patterns of indications that lead to heart disease or high blood pressure.



The diagnosis of the doctor is similar. The same is to see the retina. Together with the analysis of age, sex, weight and smoking factors. While the algorithms are not trained on those supplementary factors But it's learned from the information that was entered and created its own pattern.

However, the precision of artificial intelligence does not outweigh the current medical diagnostic process. The researchers say that this method can not replace the old method, and has to be repeated repeatedly until it is widely accepted. Yet, it is a good sign that technology is available to reduce the time it takes to reduce dependency on physicians.

💬 AI Content Assistant

Ask me anything about this article. No data is stored for your question.

Researchers from Google have published a paper in Nature Biomedical Engineering on the success of developing artificial intelligence algorithms. Analyzes and predicts high blood pressure and heart disease. Only scan the retina.

The method is to enter the retinal image from patients in the United States and the United Kingdom with over 280,000 images to find patterns of indications that lead to heart disease or high blood pressure.



The diagnosis of the doctor is similar. The same is to see the retina. Together with the analysis of age, sex, weight and smoking factors. While the algorithms are not trained on those supplementary factors But it's learned from the information that was entered and created its own pattern.

However, the precision of artificial intelligence does not outweigh the current medical diagnostic process. The researchers say that this method can not replace the old method, and has to be repeated repeatedly until it is widely accepted. Yet, it is a good sign that technology is available to reduce the time it takes to reduce dependency on physicians.