Artificial Intelligence (AI) is being touted not only as a means to enhance business operations and supply chains but it is also expected to change the face of modern medicine. Italian researchers from the University of Bari have developed an artificial intelligence system that’s capable of diagnosing Alzheimer’s disease almost a decade before symptoms appear. According to a new study, the non-invasive detection system—a type of learning device—uses an algorithm to identify tiny structural changes in the brain and has been proven to be very successful in its early stage testing.
In one test, the algorithm evaluated brain scans from 148 subjects — 52 healthy, 48 with Alzheimer’s and 48 with mild cognitive impairment (MCI), which developed into Alzheimer’s over a period of three to nine years. The system was able to classify and distinguish between healthy and diseased brains with 86% accuracy and MCI-diagnosed and healthy subjects with 84% accuracy.
According to the researchers, “MCI detection suggested that the AI could identify small brain changes that lead to Alzheimer’s almost a decade before the appearance of clinical symptoms.”
Fifty million people suffer with Alzheimer’s disease worldwide and that number is expected to reach 75 million by 2030. Alzheimer’s disease is a devastating neurodegenerative disorder that is the leading cause of dementia for the elderly, leading to memory loss, severe cognitive function impairment and ultimately death.
As yet there is no cure but most doctors believe treatments like medication and deep brain stimulation can work better if started early; however, current early detection techniques are risky, invasive and expensive. For example, cerebrospinal fluid analyses and brain imaging can accurately predict the development of Alzheimer’s disease early but both are highly invasive and costly.
While this AI device currently utilizes a small data set, in time it could be expanded and developed into a fully functional, non-invasive tool for early Alzheimer’s detection.