Related Experiment Video
Updated: Jul 30, 2025
03:37
Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
Published on: March 1, 2024
815
Intelligent Bi-LSTM with Architecture Optimization for Heart Disease Prediction in WBAN through Optimal Channel
Muthu Ganesh Veerabaku1, Janakiraman Nithiyanantham1, Shabana Urooj2
1Department of Electronics and Communication Engineering, K.L.N. College of Engineering, Pottapalayam 630612, India.
Biomedicines
|May 16, 2023
Summary
This study introduces an advanced Wireless Body Area Network (WBAN) system for accurate heart disease prediction. The novel approach enhances cardiovascular health monitoring through optimized data analysis for timely medical intervention.
Area of Science:
- Biomedical Engineering
- Health Informatics
- Artificial Intelligence in Healthcare
Background:
- Wireless Body Area Networks (WBAN) are emerging technologies in Wireless Sensor Networks (WSN) for enhanced healthcare systems.
- Continuous cardiovascular health monitoring is crucial for early disease detection, but conventional methods face limitations.
- Existing WBAN research focuses on routing, security, and energy efficiency, with a need for improved predictive capabilities.
Purpose of the Study:
- To propose a novel heart disease prediction model utilizing WBAN technology.
- To enhance the accuracy and efficiency of personal health monitoring systems for cardiovascular conditions.
- To leverage advanced AI algorithms for early detection and classification of heart diseases.
Main Methods:
- Gathering standard patient data for heart diseases from benchmark datasets via WBAN.
- Employing the Improved Dingo Optimizer (IDOX) algorithm for multi-objective channel selection in data transmission.
- Utilizing One Dimensional-Convolutional Neural Networks (ID-CNN) and Autoencoder for deep feature extraction.
- Performing optimal feature selection using the IDOX algorithm.
- Implementing heart disease prediction with Modified Bidirectional Long Short-Term Memory (M-BiLSTM), with hyperparameters tuned by IDOX.
Main Results:
- The proposed method accurately categorizes patient health status based on abnormal vital signs.
- The IDOX algorithm effectively optimizes channel selection and feature selection processes.
- Deep feature extraction using ID-CNN and Autoencoder enhances the predictive model's performance.
- M-BiLSTM, tuned by IDOX, demonstrates high accuracy in heart disease prediction.
Conclusions:
- The developed WBAN-based system offers a reliable solution for continuous cardiovascular health monitoring.
- The integration of IDOX, ID-CNN, Autoencoder, and M-BiLSTM significantly improves heart disease prediction accuracy.
- This approach facilitates timely and appropriate medical care by accurately identifying patients' health status.
