What question did this study set out to answer?

The research aims to develop a system integrating wearables and AI to detect cognitive decline in older adults early and accurately.

February 21, 2026

AI Enhanced Wearable Systems for Proactive Detection of Cognitive Decline in Elderly People

Key Points

The research aims to develop a system integrating wearables and AI to detect cognitive decline in older adults early and accurately.
Integrated wearable technology for tracking physiological and lifestyle data
Utilized surveys and sensors for data collection
Employed exploratory data analysis for pattern recognition
Applied Long Short-Term Memory (LSTM) networks for predictive modeling
Evaluated model performance using Mean Squared Error (MSE) and Receiver Operating Characteristic (ROC) metrics.
Achieved real-time monitoring and early intervention for cognitive health
Provided personalized lifestyle recommendations based on continuous data trends
Enhanced predictive accuracy due to integration of multiple data sources

Abstract

Cognitive impairment in older adults, including neurodegenerative conditions like Alzheimer's and Parkinson's, is a growing concern for global healthcare systems. Despite increasing prevalence, early diagnosis remains difficult due to the limitations of conventional, subjective, and delayed techniques. The use of wearable technology offers new opportunities for real-time, continuous health tracking, enabling earlier detection of cognitive decline before clinical symptoms emerge. This paper presents a method that integrates wearable technology and artificial intelligence (AI) to track and predict cognitive decline. It wearables (such as heart rate, sleep cycles, and physical activity) with lifestyle factors (like combines wearable-derived features from stress, diet, and exercise) collected through surveys and sensors. The data undergoes pre-processing, including handling missing values, outliers, and feature scaling, followed by exploratory data analysis (EDA) to uncover patterns. The pre-processed wearable and lifestyle features are organized into sequential time-series representations, which are subsequently fed into a Long Short-Term Memory (LSTM) network. This modeling stage enables the system to capture long-term temporal dependencies within continuous wearable data, thereby linking real-time data acquisition directly to predictive cognitive decline assessment. "The model's performance is evaluated using metrics such as Mean Squared Error (MSE) and Receiver Operating Characteristic (ROC) scores to accurately predict and classify cognitive decline. This system provides real-time monitoring and early intervention, offering personalized recommendations-such as adjusting sleep, exercise, and diet-based on individual trend data. By converging wearable technology and AI, the system not only enhances predictive precision but also empowers elderly individuals to take proactive steps in maintaining their cognitive health, ultimately reducing the burden on caregivers and healthcare systems. This work aligns with the International Journal of Humanoid Robotics by advancing AI-enhanced wearable robotic systems for continuous human-robot interaction and proactive cognitive health monitoring in elderly populations through intelligent time-series modeling.

KI fragen

Bookmark