May 2018 smartGOV_eMagazine_May2018 - Page 15

cloud technology to address this rising global concern. MINE, now part of Microsoft AI Network for Healthcare, is a global network of eyecare providers and academic institutions. “The utility of artificial intelligence a data set of 335,799 instances from 176,037 patients aged between 0 and 25 years to predict the progression of refractive errors for a period of two years. The specialty of this model is that it predicts the refractive error for a two-year period considering a person’s is that it can make these associations and integrate data within seconds or minutes, whereas it would take humans years to aggregate all that data and find patterns within it. While research papers have an N of 30 patients, AI-powered technology can help report on an N of a billion patients,” says Dr. Ranya Habash, Assistant Professor of Ophthalmology, Bascom Palmer Eye Institute and a MINE partner. Microsoft in partnership with LVPEI, has developed a machine learning approach for prediction and progression of refractive error in children and young adults. The objective is to combine anonymous medical records and therapy data from the above healthcare providers and train cutting-edge machine learning models to predict the progression of refractive errors over a two-year period. With AI and cloud computing, monitoring and predictive analysis can be delivered on a much larger scale. The Microsoft model is currently integrated and being validated in EMR systems at 174 centers of LVPEI in India. The model was trained on medical history, eye condition and gender. Accuracy measures showed that the model proved statistically accurate and provided optimal results when used to predict cases of myopia that could become more severe and cause physical changes to the eye’s sphere and cylinder. High myopia can cause blindness for which there exists no effective restorative treatment currently. In this context, the machine learning model that predicts the refractive error will be critical in identifying patients at risk of developing high myopia, when integrated into clinical practice. “Technological tools allowing predictive analytics and application of artificial intelligence to arrive at these insights for doctors will be of immense help in developing strategies to control the problem. L V Prasad Eye Institute’s partnership with Microsoft will help pave the path for this progress,” Dr. G.N. Rao believes. When expanded, this program should help eyecare providers, to closely monitor cases of myopia, find the patients who are at high risk of eventual blindness and implement anti-myopia strategies and remedies to prevent loss of vision. This should help bring down the rate of blindness across the world. The MINE work was presented by Raghu Gullapalli, Executive Director of Emerging T