Weighted Rough Set Theory for Fetal Heart Rate Classification

Weighted Rough Set Theory for Fetal Heart Rate Classification

S. Udhaya Kumar (Department of Computer Science and Engineering, Sona College of Technology, Salem, India), Ahmad Taher Azar (Robotics and Internet-of-Things Lab (RIOTU), Prince Sultan University, Riyadh, Kingdom of Saudi Arabia & Faculty of Computers and Artificial Intelligence, Benha University, Banha, Egypt), H. Hannah Inbarani (Department of Computer Science Periyar University, Salem, India), O. Joseph Liyaskar (Government Mohan Kumaramangalam Medical College, Salem, India) and Khaled Mohamad Almustafa (College of Computer and Information Sciences, Prince Sultan University, Riyadh, Saudi Arabia)
Copyright: © 2019 |Pages: 19
DOI: 10.4018/IJSKD.2019100101
OnDemand PDF Download:
No Current Special Offers


A novel weighted rough set-based classification approach is introduced for the evaluation of fetal nature acquired from a CardioTocoGram (CTG) signal. The classification is essential to anticipate newborn's well-being, particularly for the life-threatening cases. CTG monitoring comprises of electronic fetal heart rate (FHR), fetal activities and the uterine contraction (UC) signals. These signals are extensively used as a part of the pregnancy and give extremely significant data on fetal health. The obtained data from these recordings can be utilized to anticipate the condition of the newborn baby, which gives an open door for early medication before perpetual deficiency to the fetus. The dimension of the obtained features from CTG is high and decreases the accuracy of classification algorithms. In this article, supervised particle swarm optimization (PSO) with a rough set-based dimensionality reduction method is used to find a minimal set of significant features from CTG extracted features. The proposed weighted rough set classifier (WRSC) method is utilized for predicting the fetal condition as normal and pathological states. The performance of the proposed WRSC algorithm is compared with various classification algorithms such as bijective soft set neural network classifier (BISONN), rough set-based classifier (RST), multi-layered perceptron (MLP), decision table (DT), Java repeated incremental pruning (JRIP) classifier, J48 and Naïve Bayes (NB) classifiers. The experimental results demonstrated that the proposed algorithm is capable of forecasting the fetal state with 98.5% classification accuracy, and the results show that the proposed classification algorithm performed considerably superior than other classification techniques.
Article Preview

1. Introduction

During pregnancy, exact prediction of fetal health with obtaining information is very critical, and Obstetricians ought to utilize secondary information associated with a fetal state, where the most essential technique is fetal heart rate (FHR) prediction. FHR is equipped tool to avoid the possible long-term burdens, for example, neonatal neuro development inability, cerebral paralysis and fetal hypoxia (Spilka et al., 2012; Doria et al., 2007). The establishment of Cardiotocography (CTG) in the 1960’s was accompanied by phenomenal expectation since it presented an innovative continuous recording and the monitoring of the fetal heart rate (FHR) and the uterine contractions (UC) during pregnancy (Alfirevic et al., 2006). Figure 1 shows a characteristic of the CTG signal in which FHR and UC are at the higher and lower portions of the figure. Traditionally, a graphical representation of fetal heart rate (FHR) signal is externally investigated by medical specialist, whose job is to recognize and to classify the signal patterns. The analysis of heart rate signals acquired by fetal monitoring depends mostly on the definition of the basal level of the FHR signal and its variability. The basal level of FHR signal, called the baseline, is reflected as the running average heart rate in the absence of external stimuli throughout the periods of fetal rest (Rooth et al., 1987). The FHR inconsistency is well-defined in the characteristic of its transient evolution (acceleration) or reduction (deceleration). Accelerations are the outcome of fetal activities to check the fetal health and decelerations are the indications of fetal suffering from the hazard of fetal hypoxia (Palomäki et al., 2006; Ocak, 2013).

Figure 1.

CTG signal (top part is FHR and the lower one is UC)


During the crucial time of pregnancy, FHR is utilized as the primary screening method of the fetal acid-base balance (van Geijn, 1996). The visual investigation of FHR signal does not ensure a right evaluation of the fetal state, and the exactness of the interpretation relies upon the expert’s involvement. It was determined by Steer (2008) that the fault of CTG in the poorest principles of interpretation and the commitment of the human element is exhibited by high intra-and inter-observer inconsistency. To decrease the inconsistency and increase the efficiency of the CTG interpretations, various techniques are applied by researchers to develop computer aided diagnosis systems for the investigation and the prediction of CTG signals.

The initial work for automatic analysis was totally in view of clinical rules for CTG evaluation (Rooth et al., 1987). Recently, techniques derived from adults Heart Rate Variability (HRV) research were likewise utilized for the FHR examination (Gonçalves et al., 2006). The statistical report of CTG tracings was utilized in Gonçalves et al. (2006), Magenes et al. (2000) and Salamalekis et al. (2002). A hybridization of neural systems and fuzzy models, namely neuro-fuzzy systems were also employed for the classification of fetal cardiotocograms (Czabanski et al., 2008). A fuzzy inference system with artificial neural network (ANBLIR) and epsilon-insensitive learning (Łęski, 2003) was utilized for the forecast of fetal result on the basis of FHR signal analysis (Gonçalves et al., 2008, 2010). The epsilon-insensitive learning, engaging the principles of the statistical learning theory resulted in high prediction accuracy (Vapnik, 1999; Vapnik & Vapnik, 1998).

Complete Article List

Search this Journal:
Open Access Articles: Forthcoming
Volume 14: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 13: 4 Issues (2021)
Volume 12: 4 Issues (2020)
Volume 11: 4 Issues (2019)
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing