How hospitals in India are helping Alphabet's AI project for diabetic retinopathy

Hospitals in India are helping Alphabet build an artificial intelligence algorithm that screens patients for diabetic retinopathy, The Guardian reports. Diabetic retinopathy, a diabetes-related eye disease, is a common cause of vision loss.

Four things to know:

1. Aravind Eye Hospital has been involved in a pilot project to test the algorithm, developed by Alphabet subsidiaries Google and Verily, since 2016. Two other hospitals, Narayana Nethralaya and Sankara Nethralaya, are also helping to build the AI algorithm.

"By partnering with well-known institutions like Aravind Eye Hospital and Sankara Nethralaya, we can continue our research and pilot studies in implementing AI-powered screening technology, and then extend it to clinical practice," Google product manager Lily Peng, PhD, told The Guardian.

2. The algorithm, which was trained on thousands of retina images, detects signs of diabetic retinopathy. To help build the AI, participating clinicians and researchers examined images of the interior surface of the eyeball, known as the fundus, and marked each image for abnormalities associated with the disease.

3. Now, Aravind Eye Hospital is assessing how the algorithm stacks up against trained ophthalmologists. The algorithm is 97.5 percent accurate, according to staff at the hospital. If there's a discrepancy between the AI's diagnosis and a physician's, the hospital brings in a senior retina specialist to make a final decision and fine-tune the system.

4. The AI project could be particularly helpful in countries like India, which suffers from a physician shortage. India has just over 1 million physicians to treat its population of 1.3 billion people, according to a National Health Profile report from 2017.

"It's been very exciting to watch this [project] take shape," Dr. Renu Rajan, a retinal surgeon at Aravind Eye Hospital, told The Guardian. "When a machine is trained to be capable of identifying abnormal patterns in this way, it saves a doctor so much time in diagnosis; time that could be better spent helping the patient manage the condition and in aftercare."

More articles on artificial intelligence:
Northwell Health adds AI for readmissions to its EMR
New algorithm from MIT might be able to 'de-bias' AI: 3 notes
Dr. Eric Topol: 10 potential AI applications for clinicians, hospitals

© Copyright ASC COMMUNICATIONS 2020. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.


Featured Webinars

Featured Whitepapers