Artificial intelligence algorithm can diagnose childhood autism from retinal photos with absolute 100% accuracy.
One study published at JAMA Network Open this month, showed that researchers were able to successfully diagnose childhood autism using photographs of the children's retinas, which they analyzed through a deep learning artificial intelligence algorithm.
To test the algorithm, the researchers recruited 958 participants, all with averages age the 7,8 years. They then photographed their retinas in a total of 1.890 images.
The study reports that half of the participants were already diagnosed with autism spectrum disorder (ASD) and that the other half were simply age- and gender-matched participants.
The AI system was then trained on 85% of the retinal images, along with symptom severity scores, to help it build models on which to base its diagnosis.
The remaining 15% of the eye photos were given to him for testing by the researchers.
The test is based on a new method scientists have found that allows them to access information about the brain by looking at the back of the eye, where the retina and optic nerve connect to the optic disc.
This method allowed scientists to develop a non-invasive means of rapidly diagnosing strokes by emitting a single, eye-safe, laser beam onto the retina.
During this trial phase of the study, the AI algorithm was able to diagnose participants with childhood autism with 100% accuracy.
In this research, the scientists narrowed down the data educations to children and teenagers between the ages of 4 and 18, but they believe it could work for younger children in future studies.
The findings support the use of AI technology as an objective screening tool for the early diagnosis of autism, especially when access to a specialist child psychiatrist is limited.
Science Alert reports that about one in 36 people are believed to be autistic. Therefore, being aware of this diagnosis as early as possible can make a huge difference to young people understanding themselves and living their lives better.
