IOT-enabled wearable device and cloud framework for chronic disease identification and health monitoring using LST-BCLNM

The continuous monitoring of healthcare data addresses health anomalies and the timely reduction of the chronic disease risk. None of the existing works considered in mitigating the interference of Internet of Things (IoT) healthcare data meant for actual disease identification, besides monitoring. Therefore, this paper presents the mitigation of IoT data interference and healthcare data analysis. Initially, the data from IoT-based wearable devices, healthcare apps, and patient symptoms data are collected. Then, the interference between the IoT data is mitigated using the Sparsity-inducing Reinforcement Learning Regularized Least Mean Square (S-2(RL)MS) technique. Next, the data is secured using the Hyperbolic Tangent Elliptic Logistic Curve Cryptography (HTELCC). Also, the data is transmitted to the edge node based on Ontology Weighted Fair Queuing (OWFQ). In the edge node, data is filtered and aggregated. Then, the features are extracted, and by utilizing Long Short-Term Batch-Centric Layer Normalization Memory (LST-BCLNM), the anomaly is detected. For the presence of disease (anomaly), information is secured and sent to the cloud server. Apache Kafka processes big data, while LST-BCLNM extracts features and classifies chronic diseases accurately. Finally, to access the healthcare data, the doctor/healthcare provider registers and logs into the cloud server. Experimental evaluation demonstrated the effectiveness of the proposed IoT-cloud healthcare framework. The LST-BCLNM model achieved 97.58% classification accuracy, 96.56% precision, 97.24% recall, and 98.95% F-measure, while HTELCC attained 98.82% security and S-2(RL)MS achieved a 98.51% convergence rate. These results confirm improved healthcare data reliability, security, and chronic disease monitoring performance.