Alzheimers. Artificial intelligence detects disease by routine imaging of the brain

Scientists at Massachusetts General Hospital collected brain-level MRI data from patients with and without Alzheimer’s and then tested the model across five datasets to test its predictive ability for neurodegenerative disease. . Among all five datasets, validation of the method showed 90.2% accuracy in detecting Alzheimer’s disease.

06 MAR – By analyzing brain images collected as part of research, a group of scientists at Massachusetts General Hospital have developed an accurate method for detecting Alzheimer’s disease. The method, which uses complex algorithms and large datasets to ‘train’ the model, was described in PLoS ONE.

The team collected brain-level MRI data from patients with and without Alzheimer’s and then tested the model in five datasets to test its predictive ability for neurodegenerative disease. Overall, the research used 11,103 images from 2,348 patients at risk for Alzheimer’s disease and 26,892 images from 8,456 patients without Alzheimer’s disease. Among all five datasets, validation of the method showed 90.2% accuracy in detecting Alzheimer’s disease. And among the potential of the method, there is the ability to detect neurodegenerative disease independently of other variables such as age.

“This is one of the few studies that used routine MRI of the brain to attempt to detect dementia,” he comments Matthew Leming, who coordinated the study team – While a large number of deep learning studies have been conducted for the detection of Alzheimer’s from MRI of the brain, this study has made substantial steps towards actually performing this in real clinical settings versus to perfect laboratory settings. Our results – with cross-site, cross-time and cross-population generalizability – set a good precedent for the clinical use of this diagnostic technology”.

Source: PLoS One 2023

March 06, 2023
© breaking latest news

Other articles in Science and Drugs