In a breakthrough for artificial intelligence, an algorithm developed by researchers at Google’s life sciences unit Verily can successfully detect the risk of heart attack or stroke by a simple retina scan.
A popular proverb dating back to the biblical days states that ‘The eyes are the mirror to the soul’. But strikingly, eyes can even caution you of dozens of diseases as they offer a clear glimpse into your health. Google has combined this reality with deep learning to fasten the duration of screening and diagnosis of some life threatening diseases.
The newly developed algorithm can accurately accumulate other information that includes the patient’s age, blood pressure and smoking history. This latest deep learning tech outdid the current medical approaches. The research titled ‘Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning’ was published in the Journal Nature.
Algorithm for better health
This is not the first time that researchers at Google have stunned us with their attempts to tackle significant health problems. They have delved into a wide range of projects that include creating wristbands that help spot indicators of cancer and heart diseases, developing glucose-monitoring contact lenses, exploring how genetics influence cardiovascular diseases, or ways to reverse coronary heart diseases.
Researchers at Verily formulated the algorithm by using machine learning to analyze the dataset of around 300,000 patients from the US and UK.
Verily earlier worked on controlling the spread of Zika virus. Their plan of action was to release male mosquitoes infected with a particular bacterium that makes females sterile on mating. Verily’s four-year study ‘Project Baseline’ predicted why people shift from being healthy to sick. These projects garnered a lot attention for the company.
Researchers at Verily formulated the algorithm by using machine learning to analyze the dataset of around 300,000 patients from the US and UK. The images scanned from the retinas were fed into the intricate pattern-recognizing algorithm-neural networks.
Every year about 735,000 Americans suffer from a heart attack and over 795,000 get a stroke, as per data from Centers for Disease Control and Prevention. Scientists are hopeful of using these technologies to cut down the alarming increase in heart diseases.