November 23, 2024

IISc researchers develop a smartphone-based screening application for glaucoma

Software demo on an FOP device by the IISc researchers

Reading Time: 3 minutes

By Spoorthy Raman, Source: Research Matters; NHI March 19, 2019

This week, from the 10th to the 16th of March, is observed as the World Glaucoma Week globally, to spread awareness of glaucoma—a group of eye conditions that damage the optic nerve and lead to total blindness if left untreated. Glaucoma is the second leading cause of blindness in the world accounting for upto 8% of total blindness. In India, it is the leading cause of irreversible blindness affecting at least 12 million people with nearly a tenth of them going blind from the disease. About 90% of the cases remain undiagnosed, calling for better and early detection.

Now, researchers at the Indian Institute of Science, Bengaluru, have developed a smartphone-based application for the early detection of glaucoma. They have showcased it at the All India Ophthalmological Society Conference 2018 in Coimbatore. The innovation has also received the Digital Health Prize at the National Bio-Entrepreneurship Competition conducted by BIRAC and C-CAMP.

The software is based on the Fundus-on-Phone (FOP) technique, where a smartphone is used to capture the images of the fundus of the eye. The fundus is the posterior surface of the eye opposite the lens, which includes the retina, optic disc, macula, fovea, and the posterior pole. The captured images are then processed to detect a range of eye-related abnormalities, including glaucoma.

“There is a lot of emphasis on developing FOP imaging devices as smartphones today have become very powerful and have superior computing capabilities,” shares Prof Chandra Sekhar Seelamantula from the Department of Electrical Engineering, Indian Institute of Science (IISc), who has been leading the effort for the past four years.

The researchers have developed a software, which can easily be installed on any smartphone. Using optical devices that can be connected to a smartphone, the images of the fundus are captured on the phone. These images are then automatically processed to assess the severity of glaucoma. In contrast to the conventional ‘cup-to-disc ratio’ (CDR)—the ratio of the diameters of the cup-like structure formed on the optic disc due to intraocular pressure inside the eye to the diameter of the optic disc—used by ophthalmologists to detect glaucoma progression, their software looks for anomalies in the blood vessels, the optic nerve head and other areas that are more indicative of the progression of glaucoma.

“Traditional methods rely on a trained ophthalmologist to look at the images, manually compute the parameters, and make an appropriate decision – this is time-consuming and causes a lot of fatigue,” points out Prof Seelamantula.

The detection process with the new software takes only a few seconds per image and the processing happens on the phone without the need for a high-speed internet connection or a high-end desktop computer. In the blink of an eye, a detailed report is readied, which can be sent to the patient and ophthalmologist by e-mail or WhatsApp.

In today’s world of handheld devices, the Fundus-on-Phone technology is rapidly growing with many innovative applications. “Most fundus-on-phone devices do not come with a dedicated image processing software and our contribution precisely fills this gap”, says Prof Seelamantula adding that their software can be used by a semi-skilled person to capture an image and make a preliminary assessment, which is ideal while conducting screening camps in rural areas. The abnormal cases can then be referred to an expert ophthalmologist or eye surgeon.

The researchers hope to redefine access to eye care for marginalized communities that are remote and cannot afford expensive hospital consultation. “We have attempted to use our knowledge of image processing to create a societal impact,” they say. The team is now rapidly integrating Artificial Intelligence (AI) into their apparatus, keeping pace with emerging machine learning paradigms.