Improved ICA algorithm for ECG feature extraction and R‐peak detection

Summary Electrocardiogram (ECG) signal transmission and monitoring plays a paramount role in long‐term cardiac monitoring and analysis to provide remote health care in time, especially for the postoperative people and people in remote areas. The accuracy of ECG signals is of fundamental importance i...

Full description

Saved in:
Bibliographic Details
Published in:International journal of adaptive control and signal processing 2021-01, Vol.35 (1), p.38-50
Main Authors: Jayasanthi, M., Ramamoorthy, V., Parthiban, A.
Format: Article
Language:English
Subjects:
Adaptive algorithms
ECG
Electrocardiography
Feature extraction
Hardware
Independent component analysis
long‐term monitoring
Power consumption
Remote monitoring
R‐peak detection
Signal monitoring
Signal transmission
Telemedicine
VLSI
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Summary Electrocardiogram (ECG) signal transmission and monitoring plays a paramount role in long‐term cardiac monitoring and analysis to provide remote health care in time, especially for the postoperative people and people in remote areas. The accuracy of ECG signals is of fundamental importance in cardiac diagnosis like R‐peak detection. So we need to incorporate analytical methods in existing healthcare systems, to capture more meaningful ECG components and to represent physical cardiac sources more clearly. With this aim, hardware optimized FCAICA (fast confluence adaptive independent component analysis) algorithm is proposed to extract pure ECG components from the ECG mixtures. The extracted signals are then subjected to R‐peak detection for further analysis. The proposed improved fast confluence adaptive independent component analysis (IFCAICA) method occupies less hardware resources, consumes low power, improves accuracy, and sensitivity in R‐peaks detection. In 0.18 nm technology, the IFCAICA consumes 10.13 mW of power and operates with 3.4 MHz operating frequency.
ISSN:0890-6327
1099-1115
DOI:10.1002/acs.3186