Artificial intelligence (AI) can be used to detect the early symptoms of Alzheimer's disease before the patient is diagnosed.
Researchers conducting small pilot studies have trained a computer program for self-learning to recognize warning traits in too-subtle brain scans so that people can see them. The system was able to detect the onset of Alzheimer's in 40 patients on average over six years before they were diagnosed earlier.
British expert AI, Professor Noel Sharkey, from the University of Sheffield, said about the results in the United States: "This is a task in which deep learning is limited – finding high-level patterns in the data relatively small, the result is so promising it would be worth a much larger study. "
American scientists have trained the "deep learning algorithm" using over 2,100 volumes of PET (positron emission tomography) from 1,002 patients.
In one test, the algorithm received a set of 40 scans from 40 patients that had never been studied before.
It turned out that the detection of Alzheimer's disease in 100 percent was accurate on average over six years before the final diagnosis of the patient.
The results of the study were published in the latest edition of the journal Radiology.