Summary: McGill University researchers have developed a deep learning algorithm capable of detecting signatures of dementia in patients two years before the onset of symptoms by reviewing a single PET brain scan.
Source: McGill University.
Imagine if doctors could determine, many years in advance, who is likely to develop dementia. Such prognostic capabilities would give patients and their families time to plan and manage treatment and care. Thanks to artificial intelligence research conducted at McGill University, this kind of predictive power could soon be available to clinicians everywhere.
Scientists from the Douglas Mental Health University Institute’s Translational Neuroimaging Laboratory at McGill used artificial intelligence techniques and big data to develop an algorithm capable of recognizing the signatures of dementia two years before its onset, using a single amyloid PET scan of the brain of patients at risk of developing Alzheimer’s disease. Their findings appear in a new study published in the journal Neurobiology of Aging.
Dr. Pedro Rosa-Neto, co-lead author of the study and Associate Professor in McGill’s departments of Neurology & Neurosurgery and Psychiatry, expects that this technology will change the way physicians manage patients and greatly accelerate treatment research into Alzheimer’s disease.
“By using this tool, clinical trials could focus only on individuals with a higher likelihood of progressing to dementia within the time frame of the study. This will greatly reduce the cost and the time necessary to conduct these studies,” adds Dr. Serge Gauthier, co-lead author and Professor of Neurology & Neurosurgery and Psychiatry at McGill.
Amyloid as a biomarker of dementia
Scientists have long known that a protein known as amyloid accumulates in the brain of patients with mild cognitive impairment (MCI), a condition that often leads to dementia. Though the accumulation of amyloid begins decades before the symptoms of dementia occur, this protein couldn’t be used reliably as a predictive biomarker because not all MCI patients develop Alzheimer’s disease.
To conduct their study, the McGill researchers drew on data available through the Alzheimer’s Disease Neuroimaging Initiative (ADNI), a global research effort in which participating patients agree to complete a variety of imaging and clinical assessments.
Sulantha Mathotaarachchi, a computer scientist from Rosa-Neto’s and Gauthier’s team, used hundreds of amyloid PET scans of MCI patients from the ADNI database to train the team’s algorithm to identify which patients would develop dementia, with an accuracy of 84%, before symptom onset. Research is ongoing to find other biomarkers for dementia that could be incorporated into the algorithm in order to improve the software’s prediction capabilities.
“This is an example how big data and open science brings tangible benefits to patient care,” says Dr. Rosa-Neto, who is also director of the McGill University Research Centre for Studies in Aging.
While new software has been made available online to scientists and students, physicians won’t be able to use this tool in clinical practice before certification by health authorities. To that end, the McGill team is currently conducting further testing to validate the algorithm in different patient cohorts, particularly those with concurrent conditions such as small strokes.
Funding: Canadian Consortium on Neurodegeneration in Aging, Canadian Institutes of Health Research supported this research.
Source: Justin Dupuis – McGill University
Image Source: NeuroscienceNews.com image is credited to the researchers.
Original Research: Full open access research for “Identifying incipient dementia individuals using machine learning and amyloid imaging” by Sulantha Mathotaarachchi, Tharick A. Pascoal, Monica Shin, Andrea L. Benedet, Min Su Kang, Thomas Beaudry, Vladimir S. Fonov, Serge Gauthier, Pedro Rosa-Neto, for the Alzheimer’s Disease Neuroimaging Initiative1 in Neurobiology of Aging. Published online July 10 2017 doi:10.1016/j.neurobiolaging.2017.06.027
Identifying incipient dementia individuals using machine learning and amyloid imaging
Identifying individuals destined to develop Alzheimer’s dementia within time frames acceptable for clinical trials constitutes an important challenge to design studies to test emerging disease-modifying therapies. Although amyloid-β protein is the core pathologic feature of Alzheimer’s disease, biomarkers of neuronal degeneration are the only ones believed to provide satisfactory predictions of clinical progression within short time frames. Here, we propose a machine learning–based probabilistic method designed to assess the progression to dementia within 24 months, based on the regional information from a single amyloid positron emission tomography scan. Importantly, the proposed method was designed to overcome the inherent adverse imbalance proportions between stable and progressive mild cognitive impairment individuals within a short observation period. The novel algorithm obtained an accuracy of 84% and an under-receiver operating characteristic curve of 0.91, outperforming the existing algorithms using the same biomarker measures and previous studies using multiple biomarker modalities. With its high accuracy, this algorithm has immediate applications for population enrichment in clinical trials designed to test disease-modifying therapies aiming to mitigate the progression to Alzheimer’s disease dementia.
“Identifying incipient dementia individuals using machine learning and amyloid imaging” by Sulantha Mathotaarachchi, Tharick A. Pascoal, Monica Shin, Andrea L. Benedet, Min Su Kang, Thomas Beaudry, Vladimir S. Fonov, Serge Gauthier, Pedro Rosa-Neto, for the Alzheimer’s Disease Neuroimaging Initiative1 in Neurobiology of Aging. Published online July 10 2017 doi:10.1016/j.neurobiolaging.2017.06.027