Title :
A modified independent component analysis algorithm for extrating the fetal electrocardiogram
Author :
Yang Junchao ; Peng Cheng ; Qian Xiang
Author_Institution :
Shenzhen Key Lab. for Minimal Invasive Med. Technol., Tsinghua Univ., Shenzhen, China
Abstract :
Fetal electrocardiogram can provide the information about the electrical activities of the fetal heart and was suggested to be a promising diagnostic tool in clinical applications. In order to extract the fetal electrocardiogram from the mixed electrocardiograms recorded on the mother´s abdomen, strong maternal electrocardiogram interference should be considered primarily comparing to other noises. In this paper, a modified independent component analysis algorithm was proposed to eliminate the maternal electrocardiogram. Different from the conventional independent component analysis model, the proposed algorithm was simplified to a single parameter optimization problem for the two-channel electrocardiogram record. The results indicated that the proposed algorithm can extract the fetal electrocardiogram directly, avoiding the uncertainties in the conventional independent component analysis method. Therefore, the proposed algorithm was suggested to be an effective and robust approach for future applications.
Keywords :
electrocardiography; independent component analysis; medical signal processing; obstetrics; fetal electrocardiogram; maternal electrocardiogram; modified independent component analysis; single parameter optimization problem; two-channel electrocardiogram record; Algorithm design and analysis; Electrocardiography; Independent component analysis; Noise; Optimization; Signal processing algorithms; Uncertainty; biomedical signal processing; fetal electrocardiogram; independent component analysis;
Conference_Titel :
Audio, Language and Image Processing (ICALIP), 2014 International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4799-3902-2
DOI :
10.1109/ICALIP.2014.7009760