Grant: DRA: Canada Impact+ Research Training Awards - Doctoral

"Mild Cognitive Impairment (MCI) is a global priority, representing a critical window for intervention before dementia, affecting 10-20% of adults over age 65. While physical activity delays cognitive decline, management is complicated when MCI is comorbid with cardiovascular disease (CVD). CVD requires careful exercise prescription due to activity limitations and complex medications. Current clinical guidelines address each condition in isolation, leaving patients without support to navigate competing recommendations, leading to decision paralysis. Building on my completed preliminary groundwork regarding these gaps, this study develops a personalized intelligent interactive decision support system. It synthesizes MCI and CVD guidelines with patient variables to generate actionable exercise recommendations accounting for cognitive status, cardiovascular risk, functional capacity, and preferences, improving decision quality and self-management. This study uses a three-phase design. Phase 1 (Evidence Synthesis): A scoping review extracting high-quality guidelines for physical activity in MCI-CVD comorbidity. Phase 2 (Algorithm Development): A Delphi process with 15-20 experts establishing consensus-based decision rules. The system operationalizes these rules by treating guidelines as datasets and preferences as variables. A Machine Learning (ML) inference engine synthesizes this complex multi-source data to generate tailored recommendations. Phase 3 (User Testing): A quasi-experimental study (n=40) evaluating tool acceptability, usability, and impacts on decision quality and exercise adherence in older adults."