Scientists in Toronto have harnessed the power of AI to unlock the secrets of Alzheimer’s disease. It may soon be possible, through a simple eye exam, to detect this mind-robbing illness 20 years before symptoms develop. Alzheimer’s disease affects more than seven million Americans.

According to The Wall Street Journal, the model, which was developed by RetiSpec, analyzes the results from an eye scanner. Researchers say that while the hallmarks of Alzheimer’s, such as brain inflammation and neurodegeneration, have been recognized and studied, the definitive cause of the disease is still unknown.

“There still remains a huge amount we fundamentally don’t know about the brain and how it works,” said Eliav Shaked, a biomedical engineer and co-founder of RetiSpec. “The power of AI is that it can connect the dots.”

The artificial intelligence-based algorithm is also being investigated by a California-based company, Neurovision, which aims to use machine learning to develop retinal scans and blood tests to identify people who are at risk of developing Alzheimer’s disease and dementia, says the Journal. Steven Verdooner, the co-founder of Neurovision, explains that the scanner can detect the buildup of certain proteins or blood vessels in the eye that have a twisted shape and other anomalies that may signal Alzheimer’s disease — details that are hard to detect with the human eye.

“The algorithm does a better job,” he said.

And at the University of Arizona College of Medicine in Tucson, assistant professor of neurology Rui Chang has developed an AI model that spots genetic changes and triggers linked to Alzheimer’s. While Chang compares the traditional approach to deciphering symptoms of Alzheimer’s as “looking at the forest one tree at a time,” AI can evaluate the whole forest and find patterns that people can’t. Chang says his AI model has saved so much time it has cut a decade off his research. He founded a company called Path Biotech which will start clinical trials based on his research next year. 

While spinal taps and PET scans are currently the gold standards for diagnosing Alzheimer’s, they are not widely available, and the scans are expensive. Artificial intelligence technology could not only speed up the process of diagnosis, but also make it cheaper. For example, RetiSpec’s AI device scans from a camera that attaches to machines already available in most optometrist’s office.