E Oczeretko1, J Swiatecka, T Laudański. 1. Institute of Computer Science, University of Białystok, 15-887 Bialystok, Sosnowa 64 St., Poland. eddoczer@ii.uwb.edu.pl
Abstract
PURPOSE: Analysis of the uterine contractility in the nonpregnant states has provided information about physiological changes during menstrual cycle. There is need to develop methods of recording uterine activity as well as mathematical interpretation of recorded time series. Wavelets are a new powerful tool for signal and image processing. The aim of this study is an introductory view of Fourier (one of the fundamental methods of investigating of biomedical signals) and wavelet transforms applications in the analysis of uterine contractions. MATERIAL AND METHODS: Spontaneous uterine activity of healthy patient and patient with dysmenorrhea was recorded by micro-tip two sensors catheter (Millar Instruments, Inc. USA). After amplification analogue signals were converted to digital. Signals were analysed using Fourier and wavelet transforms. RESULTS: Contrary to the Fourier decomposition, which is global and provides the information integrated over the whole signal, the continuous and discrete wavelet transforms allow to extract local and global variations of the recorded contractions. From the analysis of the coefficients of the wavelet transform we can assess various pattern of propagation: normal propagation, simultaneous propagation and inverted propagation. CONCLUSIONS: This study is the introduction to the wavelet analysis of the uterine contraction signals. Wavelet transform provides insight into the structure of the time series at various scales. It allows to localise changes of the signal in time, providing additional information in comparison with the Fourier transform.
PURPOSE: Analysis of the uterine contractility in the nonpregnant states has provided information about physiological changes during menstrual cycle. There is need to develop methods of recording uterine activity as well as mathematical interpretation of recorded time series. Wavelets are a new powerful tool for signal and image processing. The aim of this study is an introductory view of Fourier (one of the fundamental methods of investigating of biomedical signals) and wavelet transforms applications in the analysis of uterine contractions. MATERIAL AND METHODS: Spontaneous uterine activity of healthy patient and patient with dysmenorrhea was recorded by micro-tip two sensors catheter (Millar Instruments, Inc. USA). After amplification analogue signals were converted to digital. Signals were analysed using Fourier and wavelet transforms. RESULTS: Contrary to the Fourier decomposition, which is global and provides the information integrated over the whole signal, the continuous and discrete wavelet transforms allow to extract local and global variations of the recorded contractions. From the analysis of the coefficients of the wavelet transform we can assess various pattern of propagation: normal propagation, simultaneous propagation and inverted propagation. CONCLUSIONS: This study is the introduction to the wavelet analysis of the uterine contraction signals. Wavelet transform provides insight into the structure of the time series at various scales. It allows to localise changes of the signal in time, providing additional information in comparison with the Fourier transform.
Authors: Piotr Pierzynski; Edward Oczeretko; Piotr Laudanski; Tadeusz Laudanski Journal: BMC Pregnancy Childbirth Date: 2007-06-01 Impact factor: 3.007