Determining a diagnosis of age related macular degeneration (AMD) is difficult work. An ophthalmologist or related lab researchers (aka eye graders) must diligently inspect the retina for the slightest abnormalities. A false positive diagnosis is possible so always getting a second opinion is smart. Either way, AMD testing is serious business and, until now, only your ophthalmologist could perform it.
Although many tests have made it to the residential market, since COVID-19, more rapid options are becoming available. With the same medical technology as a pregnancy test, soon you may be able to use a rapid test for AMD detection in your own home. One such test is currently coming to market for office and hospital implementation claiming a detection of AMD and two other eye diseases in just under two minutes.
Selena+
A team of researchers from the Singapore Eye Research Institute and National University of Singapore’s School of Computing developed a software capable of storing and searching over 500,000 retinal images. This is considered ingenious as images of all kinds of eye diseases have been stored for years. Access to this material for a comparison test in under two minutes far exceeds the ability of humans.
The software is called Selena+ which stands for Singapore Eye Lesion Analyzer Plus. It is one of many new deep learning tools being applied to macular degeneration. Selena+ detects three eye diseases:
It is believed that as computerized deep learning tools continue to progress, diagnosis and treatment of AMD as well as many other systemic diseases will be possible.
According to co-founder Dr. Daniel Ting, an associate consultant at Singapore’s National Eye Centre,
“It [Selena+] could potentially reduce 75% of the workload performed by the professional graders, enabling them to re-focus their work on the more sophisticated and complex tasks,”
EyRis Eye Camera
Selena+ is linked to a special “eye camera” developed by the startup company EyRis. It is capable of controlling a very important part of the process known as eye-movement-contingent display (EMCD). This is the ability to be able to study a patient’s eye with accurate control of the position and motion of stimulus on the retina.
Data from this recorded stimulus may bring more accurate readings to determine specific AMD diagnoses. This, in turn, allows a more targeted treatment protocol to slow progression. It also offers a study of psychophysical experiments on fixational eye movements which may help some people manage or overcome various psychological disorders.
Deep Learning
As the EyeRis camera implements into mainstream ophthalmology practices, it comes with an impressive round of studies. One such study which consisted of 120,656 manually graded color fundus [interior surface of the eye opposite the lens and includes the retina, optic disc, macula, fovea, and posterior pole] images from 3,654 Age-Related Eye Disease Study (AREDS) participants showed distinct results.
The study published the journal Ophthalmology was titled, ‘A Deep Learning Algorithm for Prediction of Age-Related Eye Disease Study Severity Scale for Age-Related Macular Degeneration from Color Fundus Photography.’ It concluded that,
“Our deep learning algorithm revealed a weighted κ outperforming human graders in the AREDS study and is suitable to classify AMD fundus images in other datasets using individuals >55 years of age.”
As more deep learning tools are utilized, positive study results of diseases such as macular degeneration may lead the way for answers of other diseases that have eluded researchers for centuries. Once these tools are able to determine rapid early detection, scientists and medical personnel alike will be able to stay one step ahead of damage rather than wait until accumulation becomes devastatingly irreversible.
Self-Learning AI
The Selena+ software is also self-learning AI (artificial intelligence). This means that as it compares a patient’s eye map to the thousands of others in its archive, it also notates the patient’s specific diagnosis. Once this is recorded it is added to the Selena+ database enabling it to advance in storing current AMD detection alongside archival data. It is believed that this technology will make eye grading a thing of the past. Other AI advancements in the battle to cure macular degeneration include:
Eye Graders No Longer Needed
Using eye graders to determine macular degeneration and many other eye diseases has been strenuous business. As reported by the Singapore’s Ocular Reading Centre,
“According to Haslina Hamzah, senior manager of SORC, and also a founding member of EyRis, her centre receives more than 4,000 images a day – all of which need to be processed by just eight to 10 staff who grade eye conditions. These graders sit in a darkened room staring at retinal images on a screen, which they need to scan for abnormalities. They are typically not allowed to work for more than half a day as the job is too strenuous…Selena+ is set to replace the role of the primary graders.”
These practitioners are not in jeopardy of losing their jobs as their expertise can be utilized elsewhere to strengthen the continued fight against AMD.
This rapid test for AMD detection is a perfect example of the continued progress researchers are making. In the not too distant future you may be able to visit your eye doctor for a checkup, have an artificial intelligence system diagnosis an eye disease like macular degeneration long before symptoms appear, and immediately begin treatment to stop AMD in its tracks.
Sources:
https://kr-asia.com/this-startup-can-detect-eye-diseases-in-under-two-minutes-startup-stories
https://pubmed.ncbi.nlm.nih.gov/17958145/
https://aplab.bcs.rochester.edu/assets/download/PDFs/articles/SantiniEtAl07.pdf
https://www.aaojournal.org/article/S0161-6420(17)33064-6/fulltext
https://www.sciencedaily.com/releases/2020/12/201218112519.htm
https://tvst.arvojournals.org/article.aspx?articleid=2765234