,
Research Scholar, School of Computer Science and Engineering, Vellore Institute of Technology (VIT), Vellore, Tamil Nadu, India
Professor, School of Computer Science and Engineering, Vellore Institute of Technology (VIT), Vellore, Tamil Nadu, India
Background: Healthcare Industrial Internet of Things (IIoT) application majorly focuses on clinical decision support, remote healthcare services, intelligent medical data management and patient monitoring to save the life of patients. The existing techniques of anomaly detection have issues of temporal misalignment, noisy physiological signals and poor handling of missing values. Objectives: To design an adaptive healthcare IIoT framework for accurately identifying the anomalies, predicting data degradation risks, improving the data quality and ensures efficient healthcare analytics within the intelligent
edge-cloud. Proposed Methodology: Initially, the healthcare data were collected and processed using Synthetic Quality Degradation Injection where, missing values, timestamp misalignment, Gaussian noise, sensor drift, packet loss, and outliers were introduced to simulate IIoT sensing and communication failures. Adaptive Context-Aware Data Ingestion (ACADI) prioritized and synchronized data, while Dynamic Residual Quality Normalization (DRQN) performed missing-value estimation, denoising, error correction, normalization, and outlier removal. These pre-processed data were given to the Temporal Multi-Scale Healthcare Feature Fusion (TMHFF) for extracting and fusing the short-term, long-term, and
frequency-domain healthcare features. Then, the fused features are fed into the Reinforced Adaptive Hybrid Long Short-Term Memory Autoencoder (RAHLA) model for detecting the anomalies with the help of Bidirectional LSTM, Sparse Autoencoder Reconstruction, and Reinforcement Learning-based threshold optimization. Predictive Evolutionary Quality Estimator (PEQE) evaluates the quality degradation risks of healthcare data and Self-Healing Intelligent Remediation Engine (SHIRE) reconstructs the degraded healthcare streams. Finally, Meta-Adaptive Feedback Optimization (MAFO) optimized model parameters, and Latency-Aware Adaptive Edge Orchestration (LAAEO) enabled efficient latency aware cloud-edge task orchestration. Result: The proposed framework achieved an Accuracy value of 98.37%, Precision of 99.17% and Latency of 1.79s, which describes better enhancement of healthcare data quality and the performance of anomaly detection. Conclusion: The proposed framework provides scalable, reliable and intelligent healthcare IIoT data quality management for healthcare analytics and monitoring applications.
This is an open access article distributed under the Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
0
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.