Australian researchers create AI-based method to detect eye cancer

25 March 2019 (Last Updated December 23rd, 2019 10:23)

The Australian Research Council’s Centre for Nanoscale BioPhotonics (CNBP) has developed an automated system to detect ocular surface squamous neoplasia (OSSN).

Australian researchers create AI-based method to detect eye cancer
Automated non-invasive technique can be used to diagnose eye cancer. Credit: Amanda Dalbjörn on Unsplash.

The Australian Research Council’s Centre for Nanoscale BioPhotonics (CNBP) has developed an automated system to detect ocular surface squamous neoplasia (OSSN).

This non-invasive technique combines advanced imaging microscopy with computing and artificial intelligence (AI) to differentiate diseased and non-diseased eye tissue in real-time.

It scans the natural light reflected by certain eye cells following stimulation with safe levels of artificial light.

According to the researchers, the light-wave signature of diseased cells is unique and can be identified by a computational algorithm.

“We will be able to confirm the disease straight away through a simple eye scan with no biopsy required.”

CNBP researcher Habibalahi said: “Clinical symptoms of OSSN are known to be variable and in early stages can be extremely hard to detect so patients may experience delays in treatment or be inaccurately diagnosed.

“The early detection of OSSN is critical as it supports simple and more curative treatments such as topical therapies whereas advanced lesions may require eye surgery or even the removal of the eye, and also has the risk of mortality.”

When tested using tissue samples from 18 OSSN patients, the automated system was able to detect the diseased cells in all cases.

The CNBP research team said that the system eliminates the need for biopsy, which could be invasive, expensive and time-consuming.

Habibalahi noted: “We will be able to confirm the disease straight away through a simple eye scan with no biopsy required and appropriate action can be quickly progressed by the specialist.”

The AI-based system can precisely map the location of abnormal tissue margins, enabling more accurate and effective treatment. The researchers believe the technology can be applied in an outpatient setting by integrating into a standard retinal camera.