Google, just like its Cupertino-based counterpart, Apple; has been closely working with AI and machine-based deep learning firms to improve disease prevention and healthcare delivery around the world. Scientists at Google and its healthcare research subsidiary, Verily have documented a new artificial intelligence based preventive diagnosis process for assessing the risk of cardiovascular diseases in patients.
A joint study between the two firms has given rise to a new process of the diagnosing risk of heart diseases among individuals. Unlike prevalent preventive tests involving various blood reports, the new method uses a simple eye-scan to determine the patient’s underlying heart conditions. The eye-scan is based on an AI algorithm that has been developed by Verily by feeding the system with medical data from nearly 3,00,000 patients. This included not only eye-scan data of such patients but also their general medical data.
The idea of analyzing cardiovascular risk based on mere eye scans may sound rather absurd, but it indeed has a strong medical basis. Often than not, our eye cells especially retina is affected when we suffer from health conditions like heart diseases or diabetes. Moreover, the rear of our eyes houses a ton of blood vessels, the changes in which are indicative of a variety of parameters like blood pressure, patient’s age, smoking habits etc.
Researchers at Verily reveal that their AI based eye-scanning has a 70 percent accuracy in predicting cardiovascular risk. This is marginally lower (2 percent) than the current Systematic Coronary Risk Evaluation (SCORE) method used for diagnosing similar risks. On the flipside, the new AI-based eye-scanning developed by Google and Verily doesn’t involve drawing of patient’s blood and saves a lot of time. The results are almost instant, and unlike SCORE test, we don’t need to wait for the blood results before getting the final report.
The artificial intelligence-based software developed by Verily for these eye-scans is still in its initial stages. Most doctors and medical researchers believe that the study needs to be conducted on a larger scale for a longer period of time before establishing itself as a medically approved method of diagnosing the risk of coronary diseases. Previously Apple had launched a medical research study based on its Apple Watch’s heart rate sensor reading to find a way of detecting possible signs of atrial fibrillation.