[ad_1]
AArtificial intelligence could detect early signs of Alzheimer's more than six years before the patient is normally diagnosed, according to research.
Scientists conducting a small pilot study have formed a self-learning computer program to recognize the characteristics of brain scans that are too subtle to be seen by humans.
The system was able to detect the onset of Alzheimer's disease in 40 patients on average more than six years before the formal diagnosis.
US researchers formed the "deep learning algorithm" using over 2,100 PET scanners (positron emission tomography) on 1,002 patients.
PET scans measure the metabolic activity of the brain by monitoring the absorption of a radioactive glucose compound injected into the bloodstream.
Research has linked the development of Alzheimer's disease with changes in metabolism in certain parts of the brain, but it can be difficult to detect.
After evaluating thousands of tests, the device learned to recognize the characteristics of the disease.
As a test, the algorithm received a set of 40 scans of 40 patients that had never been studied before. The detection of Alzheimer's disease has been 100% accurate on average more than six years before the patient's final diagnosis.
Dr. Jae Ho Sohn, a member of the University of California San Francisco team, said, "We are very pleased with the performance of the algorithm and have been able to predict each case. having evolved into Alzheimer's disease. "
[ad_2]
Source link
