Google Discovers Hidden Signs Of Anemia In Eye

Google shared its success in using deep learning to detect anemia using eye photographs.

According to Google, the human eye can reveal signs of underlying disease, which doctors can now uncover during a simple, non-invasive screening.
 
A deep learning model of Google can quantify hemoglobin using de-identified photographs of the back of the eye and common metadata. Lower level of the amount of hemoglobin indicates the anemia.
 
This research has uncovered new findings about the effects of anemia on the eye. The analysis has unveiled that much of the information comes from the optic disc and surrounding blood vessels. The optic disc is where nerves and blood vessels enter and exit the eye, and normally appears much more luminous than the surrounding areas on a photograph of the back of the eye.
 
Source: Google 
 
"this work is another example of using deep learning with explainable insights to discover new biomedical knowledge, extending our previous work on cardiovascular risk factors, refractive error, and progression of macular degeneration." wrote the company.