It came as a feather in the cap for an institution in the forefront of embracing medical and non medical technology and constantly improvising and innovating its ophthalmic and non ophthalmic processes to find solutions to emerging ophthalmic challenges and enhance quality, reach, speed and convenience in dispensing eye care, when Google the global technology and communication giant engaged in the pursuit of solutions to social and humanitarian issues chose the institution as its preferred partner in developing an Algorithm to detect diabetic retinopathy a major ophthalmic issue threatening to deprive the vision of millions around the world and macular oedema caused due to myriad contributory factors, a condition which could lead to permanent vision loss if not detected and treated early. The collaboration involved conducting real time validation of the deep learning algorithms and implementation of the same.
Senior Ophthalmologists from the institution provided the vital ophthalmic/biological inputs in terms of markers/indicators to the Google team who leveraged artificial intelligence, exhaustive neural networks and complex mathematical systems to develop a deep learning algorithm which can detect diabetic retinopathy and diabetic macular oedema with amazing accuracy, speed and perfection. The algorithm was put to real time test at the Sankara Nethralaya’s main campus at Chennai and at Thailand and its findings were evaluated by a team of senior ophthalmologists. It was a moment of great jubilation for the Google-SN team involved in the exercise when the specialists adjudicated the accuracy of the findings as being on par or even superior to findings by a qualified ophthalmologist. The institution credited with landmark technical innovations like developing and implementing one of the earliest Electronic Medical Records (EMR) systems, the DNA macro chip in ophthalmology, internet aided tele-ophthalmology camps and the one of its kind in Asia Mobile Eye Surgical Unit (MESU) will now be leveraging this most advanced screening algorithm combining exhaustive neural networks and complex mathematical systems to detect diabetic retinopathy.
Encouraged by the success of the DR detecting algorithm the Google-Sankara Nethralaya team is now working on developing a similar algorithm to detect the progression of diabetic retinopathy and macular oedema.