Binary Whale Optimization Algorithm with Bidirectional Long Short-Term Memory Blockchain-based Privacy-Preserving for Healthcare Data in Internet of Things

From the past few years, rapid integration of Internet of Things (IoT) devices into the healthcare sector led advancements in patient care and data management. However, traditional approaches for blockchain-integrated data fusion for secure health informatics faced challenges like inefficient featur...

Full description

Saved in:
Bibliographic Details
Published in:2025 3rd International Conference on Data Science and Information System (ICDSIS) pp. 1 - 5
Main Authors: Shaikh, Anwar Ahamed, Riadhusin, Rame, K, Sudhakar, Subalakshmi, R., S, Mahesh
Format: Conference Proceeding
Language:English
Published: IEEE 16.05.2025
Subjects:
Accuracy
Bidirectional long short term memory
binary whale optimization algorithm
Bioinformatics
block chain
Blockchains
Feature extraction
Internet of Things
Intrusion detection
Medical services
Pattern recognition
request pattern recognition
Whale optimization algorithms
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:From the past few years, rapid integration of Internet of Things (IoT) devices into the healthcare sector led advancements in patient care and data management. However, traditional approaches for blockchain-integrated data fusion for secure health informatics faced challenges like inefficient feature selection and high computational overhead. Therefore, this research proposes Binary Whale Optimization Algorithm based Bidirectional Long Short-Term Memory (BWOA based BiLSTM) for optimizing feature selection and enhancing predictive accuracy in blockchain-integrated data fusion for secure health informatics. Initially, data is collected from WUSTL-EHMS-2020 dataset which consists of sensor-based healthcare monitoring data like environmental parameters, energy harvesting data, device utilization metrics, and patient-related health signals. Next, blockchain-enabled request and transaction encryption is implemented to strengthen the security of data, which ensures immutable and transparent record-keeping. Then, a request pattern recognition mechanism is presented by using multiple data sources for detecting and preventing potential unauthorized access attempts. Finally, optimized feature selection with BWOA and BiLSTM network is integrated to improve the accuracy and efficiency of intrusion detection. The proposed BWAO based BiLSTM achieved better results in terms of accuracy (99.2%), precision (98.8%) and recall (98.7%) respectively when compared with existing BiLSTM
DOI:10.1109/ICDSIS65355.2025.11071121