Researchers at the Queensland University of Technology (QUT) have applied artificial intelligence (AI) to develop a more accurate and detailed method for analysing images of the back of the eye to help clinicians better detect and track eye diseases.
In the study, the group of researchers explored a range of deep learning techniques to analyse Optical Coherence Tomography (OCT) images, said David Alonso-Caneiro, QUT senior research fellow and study lead author.
OCT, which takes cross-sectional images of the eye to show different tissue layers, is a common instrument used by optometrists and ophthalmologists. These images are around four microns in size and can help clinicians detect eye diseases such as glaucoma and age-related macular degeneration.
The team collected OCT chorio-retinal eye scans from an 18-month longitudinal study of 101 children with good vision and healthy eyes, and used these images to train the AI program to detect patterns and define the choroid boundaries.
“In our study we looked for a new method of analysing the images and extracting two main tissue layers at the back of the eye, the retina and choroid, with special interest in the choroid. 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,” Alonso-Caneiro said.
“The standard imaging processing techniques used with OCT define and analyse the retinal tissue layers well, but very few clinical OCT instruments have software that analyses the choroidal tissue.
“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.”
The study compared the analysis performed by the AI program with standard image analysis methods, which QUT said produced findings that showed the AI was reliable and more accurate in analysing OCT data.
“Notably, all machine learning methods performed substantially better than the automatic baseline on the [retina tests] with respect to both accuracy and consistency with a relatively smaller improvement observed on the inner limiting membrane,” the study said.
Medicine has been touted as one of the early success stories where humans augmented by machines could literally save lives. As one example, in North Central Pennsylvania, integrated health network Geisinger has developed AI machines that can outperform its cardiologists in analysing electrocardiograms.
Similarly at the University of Sydney, researchers are developing a customised AI digital health program to prevent heart attacks. The digital health program aims to use patients’ digital footprints recorded in technologies and combine the data with AI to deliver tailored advice, nudges through text messages, and accurate risk assessments to patients who have been to the hospital with chest pains.
The research team at QUT will continue to perform further research to test the program on images from older populations and people with diagnosed diseases.