Sarfraz, M and Li, FF 2013, 'Independent component analysis for motion artifacts removal from electrocardiogram' , Global Perspectives on Artificial Intelligence (GPAI), 1 (4) , pp. 49-55.
- Published Version
Restricted to Repository staff only
Download (496kB) | Request a copy
A method of using Independent Component Analysis to remove motion induced artifacts in the signals picked up by ECG electrodes is developed in this paper. In a first aid setting, ECG electrodes on patients cannot always keep stationary, resulting in a large amount of contact noise in acquired signals. Similar problems occur in ECGs in motion, e.g. sports and ambulatory ECGs. The motion induced artifacts are known to undermine the arrhythmia recognition. An artificial neural system for automated ECG classification with an extra independent component analysis de-noising pre-processor is proposed and validated by pre-recorded real ECG and noise datasets. The proposed system shows improved recognition accuracy, providing a useful means to more accurately detect arrhythmia from ECGs in the presence of no trivial motion related noises.
|Schools:||Schools > School of Computing, Science and Engineering > Salford Innovation Research Centre (SIRC)|
|Journal or Publication Title:||Global Perspectives on Artificial Intelligence (GPAI)|
|Publisher:||Science and Engineering Publishing Company|
|Funders:||University of Salford|
|Depositing User:||FF Li|
|Date Deposited:||09 May 2016 08:56|
|Last Modified:||09 May 2016 08:56|
Actions (login required)
|Edit record (repository staff only)|