Low accessibility to eye care is a huge problem in India. The reason: there are just about 15,000 ophthalmologists catering to the growing needs of the entire population. Worse, only about 30 per cent of them practice in the rural areas. Even the number of well trained optometrists is low.

It is to address this problem that the Chennai-based Healthcare Technology Innovation Centre (HTIC) has developed Eye-PAC computing technology that can be used even by people with minimal training. HTIC is an R&D centre of IIT Madras and is supported by the Department of Biotechnology (DBT).

The first product that has been launched in collaboration with the Bangalore-based Forus is the 3nethra. According to Dr. Shyam Vasudevarao, President and Chief Technology Officer of Forus, 110 3nethra devices have been sold in a year’s time both within and outside India.

3nethra can locate abnormalities indicative of diseases by studying the anterior and posterior parts of the eye. It can also study refractive errors. According to Dr. Vasudevarao, a few hours of training is all that is required for a person to operate the device. “It’s easy to use,” he said. HTIC is now looking at more complicated problems. “We are developing Eye-PAC computing technology for screening diabetic retinopathy and glaucoma,” said Dr. Niranjan Joshi, Researcher at HTIC. “The technology to screen diabetic retinopathy is at an advanced stage of development; glaucoma is at an intermediate stage.”

Eye-PAC computing technology is based on the principle of capturing the image of the eye, digitising it before transferring the information to a computing system. The technology teases out specific information from the images and provides a computer-assisted screening decision.

The technology can process the image obtained without dilating the eye, thus saving time both for the operator and patient.

Three-stage process

The technology goes through a three-stage process before arriving at a screening decision. In the first stage the image is processed —the image is first assessed for quality and then enhanced. Enhancement in the form of increasing the contrast and illumination is particularly essential to compensate for the less information collected while inspecting a non-dilated eye.

Once the image is processed, the details available on the image are computed. At this second stage, clinically significant conditions are assigned a value.

The value assigned depends on the severity of the condition. This becomes possible as the feature is compared with a large set of reference images with clinical annotations.

“All the three stages generate multitude of images depending on the end use. The parent image is also available,” said Dr. Mohanasankar S, who heads HTIC. This allows the end user to choose the one that he wishes to examine in detail.

Analytics, the final stage is where the decision-support capability of the technology comes into play. Here all the extracted values are integrated by Eye-PAC to provide a decision to either refer or not to refer the patient for a detailed examination.

The computer-assisted decision making is particularly useful during screening programmes.

“When used as a diagnostic solution, the computer-assisted decision saves the ophthalmologists’ time,” Dr. Joshi explained.

“Eye-PAC technology can be used to develop a range of applications such as disease screening systems, tools for clinical research and analytics,” explained Dr. Mohanasankar.

More In: Health | Sci-Tech | Medicine