Artificial intelligence (AI) may soon be used to detect Alzheimer’s disease early in the course of disease say researchers. The results of the study were published in the latest issue of the journal Radiology.
The researchers from University of California at San Francisco, conducted a small but significant study where they used a self-learning computer programme to look at the brain scans of participants. The computer programme was designed to detect the early signs of neurodegenerative disease - Alzheimer’s. These features were usually too minute and subtle for human eyes to spot them. Results revealed that the programme could detect signs of Alzheimer’s disease in 40 patients six years (average 75.8 months) before symptoms of the disease manifested themselves and the disease could be formally diagnosed.
The team wrote, “There is wide recognition that deep learning may assist in addressing the increasing complexity and volume of imaging data, as well as the varying expertise of trained imaging physicians. The application of machine learning technology to complex patterns of findings, such as those found at functional PET imaging of the brain, is only beginning to be explored. We hypothesized that the deep learning algorithm could detect features or patterns that are not evident on standard clinical review of images and thereby improve the final diagnostic classification of individuals.”
The “deep learning algorithm” developed by the team was trained by showing it 2109 F-FDG PET (or fluorine 18 (18F) fluorodeoxyglucose positron emission tomography) scans from 1,002 patients. These PET scans can detect metabolic activities from all parts of the brain by showing uptake of a radioactive glucose compound that is administered by injection into the blood stream. The programme could detect changes in metabolic patterns of usage in the brain that could signal future development of Alzheimer’s. The algorithm trained itself by looking at these changes in all the scans that it was shown. Once it knew what it was looking for, it could detect all the early cases in the pilot study, the researchers explained. The pilot study was using 40 PET scans from 40 patients whose scans the algorithm had not encountered before. It predicted the early signs of the disease with 100 percent accuracy the team wrote. Dr Jae Ho Sohn, one of the co-authors of the study said, “We were very pleased with the algorithm's performance. It was able to predict every single case that advanced to Alzheimer's disease.”
The team explained that with such early detection, interventions and treatment could also begin early. This could not only slow down the progress of the disease but also halt the progression of degeneration.
Dr Carol Routledge, from the charity Alzheimer's Research UK explained that these neurodegenerative disorders usually begin up to two decades before the symptoms appear. This algorithm can help detect the disease early so treatment can start earlier before much damage is done she added. “This study highlights the potential of machine learning to assist with the early detection of diseases like Alzheimer's, but the findings will need to be confirmed in much larger groups of people before we can properly assess the power of this approach,” she said.
Camouflage detection boosts neural networks for brain tumor diagnosis