A NOVEL HYBRID CNN-LSTM FRAMEWORK FOR PREDICTING ALZHEIMER’S PROGRESSION USING SMART IOT SENSORS

Abstract

Alzheimer’s disease (AD) progression prediction is critical for timely intervention and effective patient management. This study introduces a novel hybrid deep learning framework that integrates Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to analyze longitudinal data collected from smart Internet of Things (IoT) sensors. The proposed model leverages CNN’s capability to extract spatial features from sensor data and LSTM’s strength in capturing temporal dependencies, enabling comprehensive learning from the dynamic behavioral patterns indicative of AD progression. Smart IoT sensors continuously monitor patient activities and physiological indicators, providing rich, real-time datasets essential for early and accurate detection of disease stages. Experimental evaluation on real-world datasets demonstrates that the hybrid CNN-LSTM architecture achieves superior performance in predicting Alzheimer’s progression compared to traditional models, with improved accuracy and robustness. Furthermore, the integration of IoT sensor data enhances the model’s sensitivity to subtle changes in patient condition over time. This innovative framework exemplifies the potential of combining advanced deep learning techniques with smart healthcare technologies to revolutionize the early diagnosis and monitoring of Alzheimer’s disease, ultimately supporting personalized treatment strategies and improving patient outcomes.