Overview
We apply machine learning models and advanced statistical methods to various datasets in the field of Alzheimer’s Disease (AD) and Dementia to study crucial topics such as disease progression, prognostic and diagnostic power of biomarkers, cognitive decline etc. Some of the novel modelling woks in our team include developing risk scores for amyloid positivity, comparative prognostic value of blood-based biomarkers against imaging, and building predictive models using baseline imaging data to enrich recruitment of AD clinical trials. The datasets we work with include both population-based cohort studies such as Alzheimer’s Disease Neuroimaging Initiative (ADNI) and Harvard Aging Brain Study (HABS) as well placebo and treatment arms of randomized clinical trials in AD such as Anti-Amyloid Treatment in Asymptomatic Alzheimer’s (A4) study and EXPEDITION 1,2&3
Advanced Computational Methods
We leverage a variety of computational techniques in our research. Our toolkit includes cutting-edge machine learning algorithms and sophisticated statistical methods such as Cox survival analysis and mixed effects linear models to analyze complex datasets.
Clinical Relevance of Models
Our computational models are highly pertinent to clinical practices. We develop and refine models that assist in the diagnosis of Alzheimer’s disease, predict its progression, and provide valuable prognostic information. These models are tailored to enhance clinical decision-making and improve patient outcomes.
Diverse Data Sets
Our research utilizes an extensive range of datasets. This includes data from prominent clinical trials (e.g., A4, Expedition, etc.), longitudinal studies (e.g., ADNI), and cross-sectional studies. These rich datasets enable us to understand Alzheimer’s disease from multiple perspectives and across different stages of its progression.
Enrichment of Clinical Trials
We apply our machine learning models not only to analyze the outcomes of clinical trials but also to enhance their execution. By using these models, we improve patient recruitment processes, ensuring that clinical trials are populated with suitable candidates, which in turn augments the robustness and reliability of trial results.
Our lab is dedicated to advancing the understanding of Alzheimer’s disease through innovative data science approaches, with the ultimate goal of driving meaningful improvements in clinical practice and patient care.