Published 19:25 IST, September 25th 2019
AI applied to make complex eye scans easier: Study
Researchers have used artificial intelligence (AI) to develop a more accurate and detailed method for analysing images of the back of the eye. Here's how
Advertisement
Researchers have used artificial intelligence (AI) to develop a more accurate and detailed method for analysing ims of back of eye, an vance that can help opthalmologists better detect and track eye diseases like glaucoma, and -related macular degeneration. In study, published in journal Scientific Reports, researchers looked for a new method of analysing ims from a state-of--art instrument called Optical Coherence Tomography (OCT). researchers, including those from Queensland University of Techlogy (QUT) in Australia, explored a range of machine learning techniques to analyse OCT ims.
Optical Coherence Tomography (OCT)
y tried extracting ims from two main tissue layers at back of eye from retina and choroid. OCT, a commonly used by optometrists and ophthalmologists, takes cross-sectional high-resolution ims of eye, showing different tissue layers. se ims, study ted, are of tissues about four microns thick. To put that in perspective, human hair is about 100 microns thick, researchers said. OCT can be used to map and monitor thickness of tissue layers in eye, helping clinicians to detect eye diseases, said David Alonso-Caneiro, le author of study from QUT.
Advertisement
" choroid is area between retina and sclera, and it contains major blood vessels that provide nutrients and oxygen to eye," Alonso-Caneiro said.
Advertisement
standard imaging processing techniques used with OCT, he ded, defined and analysed retinal tissue layers well, but very few clinical OCT instruments h software that analysed choroidal tissue.
"So we trained a deep learning network to learn key features of ims and to accurately and automatically define boundaries of choroid and retina," he said.
Advertisement
researchers collected OCT chorio-retinal eye scans from an 18-month study of 101 children with good vision and healthy eyes. Using se ims, y trained software to detect patterns and define choroid boundaries. y compared results with what y developed with standard im analysis methods and found that machine learning programme was reliable and more accurate.
Advertisement
"Being able to analyse OCT ims has improved our understanding of eye tissue changes associated with rmal eye development, ing, refractive errors and eye disease," Alonso-Caneiro said.
He ded that having more reliable information from se ims of choroid was clinically important and for understanding more about eye through research. According to Alonso-Caneiro, new method could provide a way to better map and monitor changes in choroid tissue, and potentially diagse eye diseases earlier. He ded that new programme was shared with eye researchers in Australia and overseas, and was hopeful that commercial OCT instruments may incorporate it.
Advertisement
19:14 IST, September 25th 2019