Deep learning is a buzzword that one hears everywhere today. Now Google has used a deep learning algorithm to analyse a photograph of the retina, and from this photograph, estimate the owner’s age, blood pressure, blood count and smoking status.
This helps get a handle on cardiovascular risk factors and thereby figure out whether the person runs a risk of a heart attack.
The facility of handling big data has made it possible for the researchers to train their models on 284,335 patients and thereby come up with the results.
Apart from age, gender and smoking status, the algorithm could guess the HbA1C and systolic blood pressure, all this by studying different parts of the retina – such as the optic disc or blood vessels.