With 4.7 million blind people, India is home to 20% of world’s visually impaired population. According to another startling revelation, there are only 20,000 ophthalmologists for a population of 1.3 billion people in India. So, the challenge of providing adequate diagnostic eyecare services is truly massive in the country. The good news, however, is that AI is being effectively used in India along with edge devices to bring eyecare to the last mile.

Many causes of vision loss that can be prevented or treated are cumulatively termed as avoidable blindness, including conditions such as cataract, refractive errors, xerophthalmia, trachoma, glaucoma and diabetic retinopathy. Due to its image recognition capabilities, AI has been growing in popularity where image analysis is essential to disease diagnosis and treatment. AI-based systems are trained with many pictures of the eye, following which the algorithms learn the difference between normal image and abnormal images. And early diagnosis and treatment can prevent or minimise vision loss or impairement.

Where optical care is hard to find, AI-based systems can help people who otherwise go without treatment, eventually resulting in their blindness. Here's how AI tools are raising India's sights for eradicating avoidable blindness.

• AI-based cataract screening app by TNeGA: Tamil Nadu e-Governance Agency (TNeGA) has developed an AI based mobile app to enable the screening of a large number of people for cataract. By simply clicking a picture, the application can be used for preliminary screening of the eye. The application has been designed to detect macular disintegration as well. Launched with help of Tamil Nadu State Blind Control Society (TNSBCS), the app is being tested in a few districts. By addressing the resource constraint in cataract detection, this aims to eradicate preventive blindness in the state. Read more...
• AI screening tool for diabetic retinopathy detection: Aravind Eye Hospital partnered with Google in 2013 to develop an algorithm that would help in the early detection of this disease.The complete screening tool was developed using Google’s deep machine learning algorithm, which has achieved an accuracy of 98.6% in detecting diabetic retinopathy, on a par with the performance of ophthalmologists and retinal specialists. 71 vision centers are using this tool across rural Tamil Nadu which have been established by Aravind Eye Hospital. These centers are supervised by trained technicians who take pictures of patients’ eyes with retinal cameras and send the digital reports to doctors at the hospital. Read more...
• AI is being used in eyecare screening for children: Telangana government was concerned about the high number of blindness and vision impairment cases occurred in children due to a lack of trained and skilled ophthalmologists. The screening tool focuses on diverse datasets of patients across geographies to come up with machine learning predictive models for vision impairment and eye diseases and can predict whether a LASIK, cataract or other surgery is going to be successful for a given patient or not. Read more...
• Democratising retinal imaging screening with AI: 3Nethra is a portable device by Forus Health that can screen common eye problems that can lead to blindness. This retinal imaging screening device is available in more than 20 countries and has already touched more than two million patients. This has been achieved by integrating AI-based retinal imaging APIs into Forus Health’s 3Nethra devices using Microsoft Azure IoT Suite and Azure IOT Edge, which delivers the cloud intelligence locally and closer to the eye. This enables operators of 3Nethra device to get AI-powered insights even when they are working at eye checkup camps in remote areas with no or intermittent connectivity to the cloud. Read more...

Image by Gerd Altmann from Pixabay