Google AI has reportedly made a huge breakthrough. It can now successfully predict cardiovascular problems such as strokes and heart attacks by simply looking into the images of the retina. The best part, there is no need to draw any blood or take any other necessary tests.
While announcing this breakthrough, Google AI officials opined that this was a big step forward scientifically as their AI was not imitating any existing diagnostic, it was actually using machine learning in order to uncover a new ways of predicting such problems. The new system even shows what particular parts and areas of the eye image lead can lead to successful predictions. This gives researchers new leads into the reasons that are behind various cardiovascular diseases.
The amazing results of the Google AI research have also been published in an interesting article titled “Prediction of Cardiovascular Risk Factors from Retinal Fundus Photographs via Deep Learning.” The article was published in Nature Biomedical Engineering.
Google AI Product Manager MD Lily Peng wrote in an article in Google AI official blog, “Using deep learning algorithms trained on data from 284,335 patients, we were able to predict CV risk factors from retinal images with surprisingly high accuracy for patients from two independent data sets of 12,026 and 999 patients.”
“For example, our algorithm could distinguish the retinal images of a smoker from that of a non-smoker 71 percent of the time, compared to a ~50 percent (i.e. random) accuracy by human experts,” she said.
Furthermore, where doctors can easily distinguish between the retinal images of their patients suffering with severe high blood pressure from normal patients, Google AI’s careful algorithm can even go further to predict the systolic blood pressure within 11 mmHg on average for patients all patients. This includes patients suffering and not suffering from high blood pressure.
