Interpretable Machine Learning for Personalized Profiling of Mild Cognitive Impairment from Daily Activities

Continuous monitoring of individual daily activities is essential to detect mild cognitive impairment (MCI) wherein timely intervention can still be applied to prevent more severe mental decline. Recent approaches in predicting MCI are mostly considering digital biomarkers across individuals but often neglecting specific indicators from a single person over a long period of time. Making this personalized, dynamic, and highly noisy prediction model with irregular distribution of missing information to be explainable and actionable for clinical use, remains a challenge. This paper presents a study on a personalized MCI prediction and profiling from an in-home and mobile cognitive health monitoring integrating data on Activities of Daily Living (ADLs), digital biomarkers, and spatial-temporal features. To address the challenge, two interpretable neural network models for MCI detection are proposed: (i) STEM-MCI integrating spatial-temporal mobility data and ADL sequences; and (ii) Fusion ART, a multi-channel digital biomarker-based predictive model, leveraging multi-modal digital biomarkers and ADL features. Empirical evaluations demonstrate that Fusion ART, based on biomarker features, produces better performance than STEM-MCI, trained on spatial-temporal mobility data. However, augmenting them with ADL data improves predictive performance of both models. Beyond prediction, the models based on biomarkers enable interpretive analysis of internal representations, offering insights into behavioral patterns that differentiate MCI from Normal Cognition (NC).