Researchers at the Queensland University of Technology (QUT) are overcoming a gap in the eye health market by turning deep learning tools to eye scans in a bid to better identify diseases like glaucoma.
In a paper recently published in scientific journal Nature, the researchers outlined how a number of artificial intelligence (AI) techniques were used to analyse Optical Coherence Tomography (OCT) scans to give optometrists and ophthalmologists more information about eye health.
OCT scans are commonly used to measure and monitor the thickness of tissue layers in the eye, indicating changes to health over time.
The study’s lead author, Dr David Alonso-Caneiro, said that while standard OCT processing tools define and analyse retinal tissues layers well, “very few clinical OCT instruments have software that analyses choroidal tissue”.
“The choroid is the area between the retina and the sclera, and it contains the major blood vessels that provide nutrients and oxygen to the eye.
“So we trained a deep learning network to learn the key features of the images and to accurately and automatically define the boundaries of the choroid and the retina.”
Over an 18-month longitudinal study of 101 children with healthy eyes, the team used OCT images to train the AI to detect patterns and define the boundaries of the choroid.
The images were then compared with the results of standard imaging analysis methods, and found the AI alternative to be more reliable and accurate.
“Being able to analyse OCT images has improved our understanding of eye tissue changes associated with normal eye development, ageing, refractive errors and eye disease,” Alonso-Caneiro added.
The researchers added that the addition of AI techniques to commercial OCT instruments, which is still largely unexplored, could result in superior performance over standard anaylsis methods.
