Reviewed by Danielle Ellis, B.Sc.Oct 3 2024
According to a study published in JAMA Ophthalmology, a team of researchers at Arizona State University's School of Computing and Augmented Intelligence is working on new diagnostic tools that will use artificial intelligence, or AI, to more efficiently screen for myopic maculopathy.
Professor Yalin Wang at work in his office. The researcher in the School of Computing and Augmented Intelligence and his team have published a series of papers in the peer-reviewed research journal JAMA Ophthalmology that outline their innovative. Image Credit: Nora Skrodenis/ASU
Myopia, or nearsightedness, is on the rise, particularly among children.
Experts predict that by 2050, myopia will affect roughly half of the world’s population. Researchers believe that an increase in “near work” — interactions with close objects such as phones and screens — is partly to blame.
For many people, the inability to see distant objects is easily remedied with glasses or contacts, but for others, it progresses to a far more serious condition called myopic maculopathy.
Myopic maculopathy occurs when the part of the eye that allows one to see straight ahead in sharp detail is stretched or damaged. The shape of the eye changes over time, becoming more elongated and less sphere-like. When this happens, vision becomes distorted.
The primary cause of severe vision loss or blindness is this dangerous condition. Ten million people had visual impairment due to myopic maculopathy in 2015. If nothing changes, it is estimated that by 2050, over 55 million people will have vision loss and that the disease will cause about 18 million blind people globally.
Since myopic maculopathy cannot be reversed, doctors advise early intervention. Early diagnosis can lead to better health outcomes, which is especially important for children. Ophthalmologists can prescribe specialized contact lenses or eye drops that slow the disease's progression.
According to Yalin Wang, a computer science and engineering professor at the Fulton Schools, technological advancements can offer significant remedies.
AI is ushering in a revolution that leverages global knowledge to improve diagnosis accuracy, especially in its earliest stage of the disease. These advancements will reduce medical costs and improve the quality of life for entire societies.
Yalin Wang, Professor, School of Computing and Augmented Intelligence, Arizona State University
A Challenge to See Things in a New Way
The Medical Image Computing and Computer Assisted Intervention, or MICCAI, Society issued a challenge in 2023 in response to this need. Experts were tasked by a professional association that promotes biomedical research innovation to enhance computer-aided screening systems for retinal images.
Currently, optical coherence tomography scans—which use reflected light to create images of the back of the eye—are used to diagnose myopic maculopathy. An ophthalmologist will then frequently manually review these scans, which can be a laborious procedure that calls for specific training.
The call was answered by Wang and his group in the Geometry Systems Laboratory. Among the MICCAI challenge winners were the researchers.
In the first section of the study, Wang and his team—which also included Dr. Oana Dumitrascu, an adjunct faculty member at Fulton Schools and a neurologist—addressed the classification of myopic maculopathy. Zhu is a doctoral student in computer engineering. There are five categories for the disease that indicate how severe it is. Identifying the appropriate level enables ophthalmologists to offer patients more individualized, efficient care.
The Fulton Schools researchers developed new AI algorithms known as NN-MobileNet. These sets of instructions that computer programs use to complete their tasks are intended to help software scan retinal images more effectively and predict the correct classification of myopic maculopathy.
The team then focused on scientific efforts to use deep neural networks, a type of artificial intelligence, to predict the spherical equivalent in retinal scans. The spherical equivalent is an estimate of the eye’s refractive error that doctors use to prescribe glasses or contacts. Deep neural networks allow researchers to assign computers the task of analyzing large amounts of data and applying AI-powered algorithms to reach useful conclusions.
Doctors can make more accurate treatment recommendations if they have a better understanding of the spherical equivalent. So, Wang and his team created new algorithms that prioritized data quality and relevance. Their new model of retinal image analysis produced excellent results while requiring minimal computing power.
Finally, Wang collaborated with other MICCAI challenge winners on a third research paper, which was published in JAMA Ophthalmology in September and presented their collected data. Researchers from universities around the world shared their challenging findings to spur further advancements and discoveries in the early and effective diagnosis of myopic maculopathy, as well as to improve healthcare outcomes for people all over the world.
A Better Vision for Global Health
Wang explains that one of the driving forces behind his work is to address health disparities.
Wang added, “People living in rural areas find it difficult to access sophisticated imaging devices and see physicians. Once AI-powered technology becomes available, it will significantly improve the quality of life in worldwide populations, including those who live in developing countries.”
According to Ross Maciejewski, director of the School of Computing and Augmented Intelligence, Wang’s project is a fine representation of faculty members’ work in the medical field.
With both myopia and myopic maculopathy increasing, solutions are needed to prevent vision loss and help health care professionals provide the best treatment options for their patients. Yalin Wang’s innovative research is a principled use of artificial intelligence to address this dire medical issue.
Ross Maciejewski, Director and Professor, School of Computing and Augmented Intelligence, Arizona State University
Journal Reference:
Qian, B., et. al. (2024) A Competition for the Diagnosis of Myopic Maculopathy by Artificial Intelligence Algorithms. JAMA Ophthalmology. doi.org/10.1001/jamaophthalmol.2024.3707