IBM researchers train Watson to identify eye abnormalities

Advancements in image analytics and deep learning technology could aid doctors in their fight against preventable blindness.
Written by Tas Bindi, Contributor

IBM researchers in Melbourne have trained a research version of Watson to recognise abnormalities in retina images that could assist doctors in the early detection of eye diseases such as glaucoma, nicknamed "the silent thief of sight" as many patients remain undiagnosed until irreversible vision loss occurs.

Commencing in 2015, the IBM researchers applied deep learning techniques and image analytics technology to 88,000 de-identified retina images accessed through EyePACS to analyse key anomalies of the eye and streamline some of the manual processes that doctors have to undertake when diagnosing eye diseases.

This includes distinguishing between left and right eye images, evaluating the quality of retina scans, and measuring the ratio of the optic cup to disc, which is one of the key indicators of glaucoma.

"It is estimated that at least 150,000 Australians have undiagnosed glaucoma, with numbers expected to rise due to our rapidly ageing population," said Dr Peter van Wijngaarden, principal investigator at Centre for Eye Research Australia, Department of Ophthalmology, University of Melbourne.

"The development of image analytics and deep learning technology will provide great promise in this area."

In the future, Watson will be capable of detecting features of other eye diseases such as diabetic retinopathy and age-related macular degeneration, IBM said in an announcement.

"Medical image analysis with cognitive technology has the capacity to fundamentally change the delivery of healthcare services," said Dr Joanna Batstone, vice president and lab director at IBM Research Australia.

"Medical images represent a rich source of data for clinicians to make early diagnosis and treatment of disease, from assessing the risk of melanomas to identifying eye diseases through the analysis of retinas. Cognitive technology holds immense promise for confirming the accuracy, reproducibility, and efficiency of clinicians' analyses during the diagnostic workflow."

IBM has also outlined plans to expand its cognitive computing footprint by connecting Watson to primary, acute, and behavioural data to obtain a complete view of patients. The Watson-primary care provider connection is being rolled out in Central New York in a six-county region with more than 2,000 providers.

In Australia, the Commonwealth Scientific and Industrial Research Organisation (CSIRO) ran a teleophthalmology trial in the Torres Strait Islands and Western Australian goldfields a few years ago, where nurses were trained to take retinal images on site using a mobile device. Those images were then forwarded for reading by ophthalmologists in Perth or Brisbane.

"We could actually provide ophthalmic screening to, in particular, diabetes patients in the community without the doctor having to travel to them or them having to travel to the doctor. That provided screening to a lot of people who wouldn't normally be screened," David Hansen, CEO of the Australian eHealth Research Centre, a joint venture between the CSIRO and Queensland Health, told ZDNet earlier in February.

"Through that study, we were able to diagnose people with diabetes who didn't even realise they had it. There were some people who had really bad diabetic retinopathy who needed to be urgently seen for sight-saving treatment."

CSIRO's teleophthalmology work, which has expanded into areas such as wound care, is currently run by the organisation's group in Perth.

Second Sight's Argus device is one of a handful of commercial retinal implants now being used to help patients with late-stage degenerative eye diseases regain some degree of vision.

Editorial standards