Time-Frequency Analysis of Electrohysterogram for Classification of Term and Preterm Birth

Abstract

Abstract:In this paper, a novel method for the classification of term and preterm birth is proposed based on time-frequency analysis of electrohysterogram (EHG) using multivariate empirical mode decomposition (MEMD). EHG is a promising study for preterm birth prediction, because it is lowcost and accurate compared to other preterm birth prediction methods, such as tocodynamometry (TOCO). Previous studies on preterm birth prediction applied prefilterings based on Fourier analysis of an EHG, followed by feature extraction and classification, even though Fourier analysis is suboptimal to biomedical signals, such as EHG, because of its nonlinearity and nonstationarity. Therefore, the proposed method applies prefiltering based on MEMD instead of Fourier-based prefilters before extracting the sample entropy feature and classifying the term and preterm birth groups. For the evaluation, the Physionet term-preterm EHG database was used where the proposed method and Fourier prefiltering-based method were adopted for comparative study. The result showed that the area under curve (AUC) of the receiver operating characteristic (ROC) was increased by 0.0351 when MEMD was used instead of the Fourier-based prefilter.

A Critical Look at Studies Applying Over-Sampling on the TPEHGDB Dataset

“…At the time of writing (December 2018), within all the 153 citations to the original paper, which introduced the TPEHGDB dataset, we have found three machine learning studies that were accessible and, to the best of our knowledge, had a sound evaluation methodology [28,22,30]. In the study of Sadi-Ahmed et al [30], all records taken before 26 weeks of gestation were filtered away from the dataset, resulting in a dataset of 138 recordings taken after the 26th week of gestation.…”

“…Nevertheless, the wavelet-based feature may be an interesting addition to the feature set. In the work of Ryu et al [28] a similar study is performed in which a feature based on Multivariate Emperical Mode Decomposition (MEMD) is proposed. They evaluate the added value of their feature, by subsampling a balanced dataset of 38 term and 38 preterm records, 100 times, from the original dataset.…”

Multivariate Time–Frequency Analysis of Electrohysterogram for Classification of Term and Preterm Labor

“…Moreover a study on preterm delivery detection using EMD achieved high classification accuracy and another study using MEMD showed that the classification using MEMD yields higher performance in compared to infinite impulse response (IIR) [46,47] Therefore, it is expected that MEMD would accurately extract the true uterine electrical activity from the multichannel uterine EMG signals, which gives a new prospect to the current researches on preterm birth prediction by improving their prefiltering performance. This paper is organized as follows: Sect.…”

Overly Optimistic Prediction Results on Imbalanced Data: a Case Study of Flaws and Benefits when Applying Over-sampling

“…At the time of writing (January 2020), out of the 160 citations to the original paper which introduced the TPEHGDB dataset, we have found three machine learning studies that were accessible, tackled preterm birth risk estimation and, to the best of our knowledge, had a sound evaluation methodology [35,36,37]. In the study of Sadi-Ahmed et al [37], all records taken before 26 weeks of gestation were filtered from the dataset, resulting in a dataset of 138 recordings taken after the 26th week of gestation.…”

“…Nevertheless, the wavelet-based feature may be an interesting and complementary addition to the feature set. In the work of Ryu et al [35] a similar study was performed in which they proposed a feature based on Multivariate Empirical Mode Decomposition (MEMD). They evaluated the added value of their feature, by subsampling a balanced dataset of 38 term and 38 preterm records, 100 times, from the original dataset.…”