Non-invasive transabdominal uterine electromyography correlates with the strength of intrauterine pressure and is predictive of labor and delivery

Abstract

EMG measurements correlated strongly with the strength of contractions and therefore may be a valuable alternative to invasive measurement of intrauterine pressure. Unlike TOCO, transabdominal uterine EMG can be used reliably to predict labor and delivery.

Identification of Human Term and Preterm Labor using Artificial Neural Networks on Uterine Electromyography Data

“…When the average PS peak frequency, the standard deviation of the burst duration, and the ratio of these two were used as input parameters for the ANN, the training set was classified correctly for 30/38 (79%) of the TL patients, 5/6 (83%) of the PTL patients, 25 (Fig. 4).…”

“…Standard TOCO external mechanical force readings were also recorded simultaneously by a transducer belt around the patients' abdomens, but only to assist with additional confirmation of the presence of uterine contraction events. Since previous work with the crude TOCO instrument has indicated limited, if any, predictive utility, 25 no numerical data were utilized from that instrument for this study.…”

A comparison of various linear and non-linear signal processing techniques to separate uterine EMG records of term and pre-term delivery groups

“…The choice of multiple band-pass filters also enabled us to test whether the results yielded by the tested methods depend on the choice of the filtering bandwidth. 20 25 Fig. 2 The times of delivery in relation to the times of recording for the 300 EHG records.…”

“…A band-pass filter is needed. Various frequency bands, such as 0.08-4 Hz (using a Butterworth digital filter) [25], 0.05-4 Hz [16], 0.2-4 Hz [20], and filtering methods including wavelets [3] were used. It was recognized that the uterine EMG content ranges from 0 to\5 Hz [4].…”

“…The techniques studied which consider individual uterine contractions included the following: the calculation of the peak frequency of the power spectrum within bursts of activity [6]; the calculation of burst energy levels [20], the use of the peak frequency, the duration and number of bursts, the means and deviations of the frequency spectrum, combined with neural networks [16]; and the approaches of analyzing contractions using multiple techniques such as the median frequency, the peak frequency and the kurtosis and skewness coefficient, combined with principal component analysis [15]. Other approaches included calculating the root mean square value of the signals and the median frequency value of the power spectrum of the signals for whole 30-min records of uterine activity [13,25].…”