Embolic signals characterization using wavelet networks

Matos, S.; Ruano, M. Graça; Ruano, Antonio; Evans, D. H.

http://hdl.handle.net/10400.1/2312

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
BioEng_paper.pdf		73.67 KB	Adobe PDF	Download

Send Feedback

Authors

Abstract(s)

Cerebral embolism represents a major cause of stroke. While embolic signal can be detected using transcranial Doppler (TCD) ultrasound, there are limitations in this technique that makes it difficult to differentiate between gaseous and solid emboli and artifacts. In this paper, we report the application of emergent signal processing techniques in the analysis and classification of embolic signals. The Wavelet Neural Network (WNN) is used to approximate the signals and the parameters from the wavelets that best fit each signal are used as inputs to train a Neural Network (NN) for classifying them as normal signals, or gaseous or solid embolic signals.

URI

http://hdl.handle.net/10400.1/2312

Citation

Matos, S.; Ruano, M. G.; Ruano, A. E.; Evans, D. H. Embolic Signals Characterization using Wavelet Networks, Trabalho apresentado em 6th Portuguese Conference on Biomedical Engineering (BioEng’2001), In 6th Portuguese Conference on Biomedical Engineering (BioEng’2001), Faro, 2001.