A new artificial intelligence (AI) algorithm developed by Canadian researchers can detect evidence of cognitive decline in brain MRI scans, genetics and clinical data, and may predict whether findings will lead to Alzheimer’s disease five years before symptoms appear.
“At the moment, there are limited ways to treat Alzheimer’s disease and the best evidence that we have is for prevention,” Mallar Chakravarty, PhD, a computational neuroscientist at the Douglas Mental Health University Institute in Canada, said. “Our AI methodology could have significant implications as a ‘doctor’s assistant’ that would help detect and treat the disease.”
For their study, Chakravarty and colleagues from the University of Toronto and the Center for Addiction and Mental Health in Toronto trained their AI algorithm with data from the Alzheimer’s Disease Neuroimaging Initiative through the National Institutes of Health’s National Institute on Aging.
The data came from more than 800 geriatric patients ranging from normal and healthy to those experiencing mild cognitive impairment and Alzheimer’s disease. The researchers found their algorithm accurately predicted cognitive decline leading to Alzheimer’s disease in cohorts by analyzing brain MRI scans, genetics and clinical data.
Study results were then replicated for accuracy on an independently collected sample from the Australian Imaging and Biomarkers Lifestyle Study of Ageing, according to the researchers.
In the pursuit of predictive biomarkers identification, several studies have reported varying neuroanatomical patterns associated with functional and cognitive decline in Alzheimer’s disease. The lack of localized atrophy could be attributed to cognitive reserve, genetics, or environmental factors. As a result, local anatomical features, such as hippocampal volume, may be insufficient for predicting future clinical decline at a single subject-level. Thus, models incorporating an ensemble of imaging features, clinical and genotypic information have been proposed.
However, in such multimodal models, the performance gains offered by the imaging data are unclear. Particularly, insight into prediction improvement from magnetic resonance (MR) images is crucial, as it may aid decision-making regarding the necessity of the MR acquisition (a relatively expensive, time-consuming, and possibly stressful requirement) for a given subject in the aims of improving prognosis. Furthermore, there is increasing interest in incorporating data from multiple time points (i.e. follow-up patient visits), in an effort to improve long-term prognosis. However, this is a challenging task requiring longitudinally consistent feature selection and mitigation of missing time points.
The overarching goal of this work is to predicting symptom progression in AD that addresses the aforementioned challenges. The contributions of this work are two-fold. First, we present a novel data-driven approach for modeling long-term symptom trajectories derived solely from clustering of longitudinal clinical assessments. We show that the resultant trajectory classes represent relatively stable and declining trans-diagnostic subgroups of the subject population. Second, we present a novel machine-learning (ML) model called longitudinal Siamese network (LSN) for prediction of these symptom trajectories based on multimodal and longitudinal data.
We envision two overarching clinical uses for the presented work. First our trajectory modeling efforts provide a symptom-centric prognostic objective across AD and prodromal population. Then the prediction model facilitates decision-making pertaining to frequency and types of assessments that should be conducted in preclinical and prodromal individuals. If an individual is predicted to be in fast-decline, then this may lead a clinician to recommend more frequent assessments, additional MRI sessions, and preventative therapeutic interventions. Conversely, if an individual is deemed as being stable (in spite of an MCI diagnosis) this may lead a clinician to reduce the frequency of assessments. Another application would be assisting clinical trial recruitment based on the projected rate of decline. This information can help identify the suitable candidates who are at high risk for decline over the next few years.
In summary, we presented a longitudinal framework that provides a data-driven, flexible way of modeling and predicting disease progression. We introduced a novel LSN model that combines clinical and MR data from two timepoints and provides state of the art predictive performance. We demonstrated the robustness of the model via successful cross-validation using three different ADNI cohorts with varying data acquisition protocol and scanner resolutions. We also verified the generalizability of LSN on a replication AIBL dataset. Lastly, we provide an example use case that could further help clinicians identify subjects that would benefit the most from LSN model predictions. We believe this work will further motivate the exploration of multi-modal, longitudinal models that would improve the prognostic predictions and patient care in AD.