Microsoft is using machine learning to help fight blindness

Though robots and artificial intelligence may not replace our doctors entirely in the foreseeable future, they are already starting to make a difference. 

Microsoft is now using machine learning and artificial intelligence to help doctors in India to diagnoze and treat eye diseases.

SEE ALSO: How the scary potential of AI brought tech competitors together

Earlier this year, Microsoft began working with the not-for-profit LV Prasad Eye Institute (LVPEI) in India to have its Azure machine learning and Power BI services analyze patterns among cases and predict the surgical outcome of eye surgery patients. 

The collaboration saw Microsoft going through a trove of data — anonymized records of 1.1 million people — and provide doctors with insights into how the blindness spreads in the country, Anil Bhansali, Managing Director of Microsoft India (R&D), explained to Mashable India in a conversation.

Microsoft says it utilized Azure machine learning service to crunch the numbers and Power BI service to visualize those numbers to make sense out of them. These numbers helped the doctors, Microsoft says, to ascertain how much time a patient has before their eye issues extrapolate. 

Microsoft says after going through such voluminous data, its AI-platform was also able to predict how likely it is that doctors will be able to successfully perform eye surgeries. Though the company didn't share how accurate its projections have been, it said it is ready to expand the initiatives with other academic and research organizations. 

Today the company announced it has partnered with Bascom Palmer - University of Miami, Flaum Eye Institute - University of Rochester (USA), Federal University of Sao Paulo (Brazil), and Brien Holden Vision Institute (Australia). 

As part of the partnership, Microsoft and institutes will closely work on the diverse datasets of patients across geographies to find predictive models for vision impairment and eye disease. 

Some of the issues that doctors and Microsoft would be working on include finding the rate of change of myopia in children and conditions that impact children's eyesight. They will also try to predict outcomes of refractive surgery and determine optimal surgery parameters for higher probability of success.

India accounts for nearly 55 million of the 285 million eye disease cases worldwide. Bhansali says it was one of the reasons why the company piloted the initiative in India. 

The company maintains several other AI-driven initiatives in India, such as digital agricultural applications, and its work with the Andhra Pradesh sate government to help identify students who are likely to drop out.

Microsoft isn't the only Silicon Valley giant trying to make use of its artificial intelligence platforms to better serve Indians, though. In July, for instance, Manipal Hospitals partnered with IBM Watson at six locations to provide information and insights to physicians to help them identify personalized, evidence-based cancer care options across India. 

IBM's Watson has gotten impressively accurate over the years. At the University of North Carolina School of Medicine, Watson was able to recommend the same treatment as human experts for cancer diagnosis 99 percent of the time. More interestingly, in 30 percent of the cases, Watson found treatment options that the doctors had missed.