Smart home Internet of Things-based behavioural analysis for early detection of cognitive decline: toward Saudi future vision.

Abstract

Background: The increasing integration of Internet of Things (IoT) technologies in smart-home environments has enabled continuous collection of behavioural data that can support cognitive health monitoring. Early identification of behavioural deviations associated with cognitive decline is critical for timely intervention and quality-of-life improvement among older adults. However, conventional clinical assessments are often episodic, subjective, and resource-intensive. The objective of this study is to develop a non-invasive, data-driven framework for analysing daily behavioural patterns from smart-home IoT data to support early cognitive-risk screening rather than clinical diagnosis.

Methods: This study proposes HEALNET (Home Environment Assisted Learning Network), a hybrid deep learning (DL) framework that integrates convolutional neural networks (CNNs) for spatial feature extraction, long short-term memory (LSTM) networks for temporal sequence modelling, and ensemble machine learning (ML) classifiers including Random Forest (RF) and support vector machine (SVM). The framework analyses longitudinal behavioural data collected from smart-home IoT sensors. Experimental evaluation was conducted using publicly available Centre for Advanced Studies in Adaptive Systems (CASAS) smart-home datasets.

Results: The proposed HEALNET framework achieved a classification accuracy of 94.2%, outperforming baseline ML and DL models. Results demonstrate that the integration of spatial, temporal, and statistical behavioural representations improves the detection of behavioural patterns associated with elevated cognitive-risk indicators.

Conclusions: The findings indicate that continuous, unobtrusive behavioural monitoring using smart-home IoT data can provide reliable indicators for cognitive-risk screening. HEALNET serves as a research-stage framework supporting data-driven behavioural analysis rather than clinical diagnosis and aligns with Saudi Vision objectives for digital health innovation and quality-of-life enhancement.