COMPUTER engineers have joined hands with eye specialists to detect and prevent blindness, as well as understand its genetic roots.  They are developing specialised image processing software to analyse photographs taken of the retina - the nerve layer at the back of the eye - to screen people and detect glaucoma early.

Glaucoma is an eye disease where fluid pressure within the eye becomes too high, damaging the delicate fibres of the optic nerve which carries visual impulses from the eye to the brain. This damage is irreversible and can cause blindness.  'Most patients are not aware they even have glaucoma in its early stages, but through our imaging technologies we can detect it even before any vision is lost,' says Dr Jimmy Liu, a research scientist at the Institute of Infocomm Research who is leading the effort.  This means patients can start treatment early to slow its progression.

According to the World Health Organisation, glaucoma is the second leading cause of blindness globally, after cataract.  The disease affects about 6 per cent of Singaporeans and accounts for 40 per cent of vision loss here.  Although it cannot be cured, it can be controlled successfully in most cases. Treatment can be in the form of medication, laser therapy or surgery, depending on the type of glaucoma.

The team's automated glaucoma diagnostic system combines images taken by various devices and studies different parts of the eye such as the blood vessels.  Its success lies in innovative medical image processing and computations that can then pinpoint the disease.  It has produced results better than any high-tech equipment in the market, says Dr Liu. In fact, its report card is similar to when glaucoma specialists scan the images to look for signs of the illness, he adds.

The work is a collaboration between the institute and the Singapore Eye Research Institute (Seri). The project is called Aglaia - Automatic Glaucoma Diagnosis and Its Genetic Association Study through Medical Image Informatics.

Commenting on the effort, Seri director Wong Tien Yin says it is a model of how various groups could work together, ranging from doctors and epidemiologists to statisticians and computer scientists.  The team is already talking to commercial partners to take the project into the clinic.

Professor Wong, who is co-principal investigator of the project, says: 'I think in the future, we could be... the first country where there's automated early detection of glaucoma, leading the way for possibly implementation of this system throughout Asia and the rest of the world.'  The team has developed prototypes to detect a host of other eye diseases, and is also working with 6,000 patients here to study the genetic risk factors associated with glaucoma.






Dr Liu is leading an effort to develop image processing software for analysing photos of the retina (above) to help detect glaucoma early.