STRIDE: Systematic Radar Intelligence Analysis for ADRD Risk Evaluation with Gait Signature Simulation and Deep...

Area of Science:

• Biomedical Engineering
• Artificial Intelligence
• Neuroscience

Background:

• Abnormal gait is a key non-cognitive biomarker for Alzheimer's disease (AD) and AD-related dementia (ADRD).
• Early detection of ADRD is crucial for timely intervention and management.
• Non-wearable technologies offer potential for unobtrusive patient monitoring.

Purpose of the Study:

• To design STRIDE, a system integrating micro-Doppler radar and AI for ADRD risk assessment.
• To develop and validate a deep learning (DL) classification framework for gait signature analysis.
• To demonstrate the feasibility of using radar-based gait analysis for early ADRD risk evaluation.

Main Methods:

• Development of a STRIDE "digital-twin" using human walking and micro-Doppler radar simulation models.
• Simulation of individuals with ADRD under various conditions using established walking parameters.
• Generation of micro-Doppler signatures from simulated gait using electromagnetic scattering and Doppler shift models.
• Application of a band-dependent DL framework for ADRD risk prediction.

Main Results:

• Successful generation of a comprehensive gait signature dataset through simulation.
• Demonstration of the effectiveness of the DL framework in predicting ADRD risks.
• Validation of STRIDE's feasibility for evaluating ADRD risk via gait analysis.

Conclusions:

• STRIDE, integrating micro-Doppler radar and AI, is a feasible approach for ADRD risk assessment.
• The developed DL framework shows effectiveness in analyzing gait signatures for risk prediction.
• This non-wearable technology holds potential for early, non-invasive detection of ADRD.