PROJECT SUMMARY (See instructions): Declines in cognitive and motor function are early harbingers of pending Alzheimer's Disease and related dementia (AD/ADRD). New technologies for characterizing mobility and cognition have made significant inroads within the younger, healthier population segments but have not permeated all demographics, especially among older adults. Additionally, current approaches for monitoring cognition and mobility are limited to periodic assessments in laboratory and clinical settings, which fail to capture the subtle gradation and continuous deterioration of these domains. To address this gap, we propose a multi-modal system of wearable sensors that will provide data to enable breakthroughs in statistical machine learning and health informatics, allowing for a nuanced, continuous assessment of mobility and cognition. Critically, our combined hardware/software system will address the unique challenges of an aging population including the need for a health monitoring system which is both comfortable and easy to use. In Aim 1, we will develop a system of wearable eutectogel sensor patches that are flexible and breathable, transparent, and inconspicuous. These individual sensor devices, designed for long-term wear, will provide direct, continuous, real-time monitoring of eight biophysical domains (gait, posture, head motion, heart rate variability, respiration, location, orientation, and movement). In Aim 2, building on state-of-the-art methods for the modeling and analysis of multiple time series, we will develop statistical machine learning algorithms and models for the estimation of cognitive function and mobility over time given data from the gel-biosensors. To validate the resulting engineered system, we undertake a feasibility study to determine the potential of the analytics-enhanced, integrated biosensor system and machine learning algorithms to predict cognitive function and mobility in 20 older men and women, starting in a controlled laboratory setting and extending into real-world scenarios. We anticipate our gel-based biosensor system and machine learning models will advance personalized sensing and health informatics with long term, direct application to detecting and predicting early development and progression of AD/ADRD that increasing numbers of older adults are now facing.