AI, Deep Learning Allow More Accurate Eye Testing with OCT

Facebook X LinkedIn Email
Researchers at Queensland University of Technology (QUT) in Australia are now able to analyze OCT images of the eye closer and more accurately than ever before.

OCT is commonly used by optometrists and ophthalmologists to obtain cross-sectional images of the eye that show different tissue layers. The researchers have developed a more detailed and accurate way to evaluate such images, particularly the back of the eye, using artificial intelligence (AI) deep learning techniques. “In our study, we looked for a new method of analyzing the images and extracting two main tissue layers at the back of the eye: the retina and choroid,” said David Alonso-Caneiro, senior research fellow and a professor in the QUT Health School of Optometry and Vision Science. The choroid, the pigmented vascular layer of the eyeball between the retina and the sclera, contains the major blood vessels that provide nutrients and oxygen to the eye.

“The standard imaging processing techniques used with OCT define and analyze the retinal tissue layers well,” he said, “but very few clinical OCT instruments have software that analyzes the choroidal tissue.”

Dr David Alonso-Caneiro performing an OCT scan at the QUT School of Optometry and Vision Science.
Dr. David Alonso-Caneiro performing an OCT scan at the School of Optometry and Vision Science. Courtesy of QUT.

In their study, the researchers gathered OCT chorio-retinal eye scans from an 18-month study of 101 children who had good vision and healthy eyes. After exploring a range of advanced deep learning techniques, they used the images to train the AI deep learning program to detect patterns and define the choroid boundaries. This method has already boosted the team’s understanding of changes in eye tissue that are the result of typical development and aging, refractive errors, or disease.

Next, the team will test the method on images from older populations and people with already diagnosed eye diseases such as glaucoma and age-related macular degeneration.

“Having more reliable information from these images of the choroid, which our program provides, is important clinically and also for advancing our understanding of the eye through research,” Alonso-Caneiro said. “We feel our methods could provide a way to better map and monitor changes in choroid tissue, and potentially diagnose eye diseases earlier.”

The study was published in Nature Scientific Reports (

Published: October 2019
Ophthalmology is a branch of medicine that focuses on the anatomy, physiology, and diseases of the eyes and visual system. Ophthalmologists are medical doctors who specialize in the diagnosis, treatment, and prevention of eye disorders and diseases. They are trained to provide comprehensive eye care, including medical, surgical, and optical interventions. Key areas within ophthalmology include: General eye care: Ophthalmologists perform routine eye examinations to assess visual acuity,...
optical coherence tomography
Optical coherence tomography (OCT) is a non-invasive imaging technique used in medical and scientific fields to capture high-resolution, cross-sectional images of biological tissues. It provides detailed, real-time, and three-dimensional visualization of tissue structures at the micrometer scale. OCT is particularly valuable in ophthalmology, cardiology, dermatology, and various other medical specialties. Here are the key features and components of optical coherence tomography: Principle of...
artificial intelligence
The ability of a machine to perform certain complex functions normally associated with human intelligence, such as judgment, pattern recognition, understanding, learning, planning, and problem solving.
deep learning
Deep learning is a subset of machine learning that involves the use of artificial neural networks to model and solve complex problems. The term "deep" in deep learning refers to the use of deep neural networks, which are neural networks with multiple layers (deep architectures). These networks, often called deep neural networks or deep neural architectures, have the ability to automatically learn hierarchical representations of data. Key concepts and components of deep learning include: ...
Research & TechnologyeducationAsia-PacificQUTophthalmologyoptical coherence tomographyOCTeye diseasemedicalartificial intelligencedeep learningBioScan

We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.