Intelligent SMOTE Based Machine learning

Classification for Fetal State on Cardiotocography

Dataset
Pijush Dutta 
(

[email protected]
)
Global Institute of Management and Technology
https://orcid.org/0000-0001-5144-315X
Shobhandeb Paul 
Gurunanak Institute of Technology
Madhurima Majumder 
Mirmadan Mohanlal Government Polytechnic

Research Article

Keywords: Fetal cardiotocograph, machine learning, SMOTE, Feature selection, Confusion matrix,
performance metrics

Posted Date: November 17th, 2021

DOI: https://doi.org/10.21203/rs.3.rs-1040799/v1

License:


This work is licensed under a Creative Commons Attribution 4.0 International
License.
 
Read Full License

Page 1/17
Abstract
A major contributor to under-five mortality is the death of children in the 1st month of life. Intrapartum
complications are one of the major causes of perinatal mortality. Fetal cardiotocograph (CTGs) can be
used as a monitoring tool to identify high-risk women during labor. The objective of this study was to
study the precision of machine learning algorithm techniques on CTG data in identifying high-risk
fetuses. CTG data of 2126 pregnant women were obtained from the University of California Irvine
Machine Learning Repository. Out of 2126 CTG dataset 78% of them were normal, 14% were suspect, and
8 % had a pathological fetal state. To improve data imbalance SMOTE is applied followed by five
different machine learning classification models were trained using CTG data. Sensitivity, precision, and
F1 score for each class and overall accuracy of each model were obtained to predict normal, suspect, and
pathological fetal states. For the model validity two statistical parameters MCC & kappa (k) are used.
SMOTE based all the classification algorithm provides the higher degree of accuracy with minimum value
is 96% and RF algorithm had the highest prediction accuracy about 98.01% which is quite satisfactory.
Model validation statistical parameters MCC & kappa is maximum achieved by RF about 0.968 & 1 and
for SVC is 0.977 & 1 respectively. Finally proposed work also compared with previous state of art
techniques.

1. Introduction
Globally 2.4 million children died in the first month of life in 2019. There are approximately 7 000
newborn deaths every day, amounting to 47% of all child deaths under the age of 5-years, up from 40% in
1990. In a pregnancy cycle, the fetal heart rate (FHR) is one of the most important evidence about the
fetus[1]. The obstetricians are using cardiotocograph (CTG) to get information that includes FHR and
uterine contractions (UC) related to the fetus. The CTG intended for not only to get FHR, but also to
observe the mother’s contractions and other kinds of fetal monitoring [2]. CTG is a medical test utilized
throughout pregnancy which records UC and FHR. This test can be employed by either external or internal
techniques. With internal test, a catheter is located in the uterus after a precise quantity of expansion has
taken place. In external tests, a pair of sensor nodes is attached to the mother's stomach. The CTG data
usually represents two lines. The upper line records the FHR in beats per minute. The lower line records
uterine contractions[3]. In order to find fetal risks based on CTG, machine learning techniques turn out to
be an increasing trend to produce decision support systems in medicine. Different studies have carried
out for the classification of the CTG data[4].

The information taken from CTG is utilized for early identification of a pathological state and can help
the obstetrician to predict future problems and hinder before occurring a permanent impairment to the
fetus. Throughout the delivery of the baby who is showing to hypoxia can cause a temporary impairment
or death. Because of the wrong diagnosis of the FHR pattern recordings and inappropriate treatments
employed to the fetus can achieve more than half of these deaths[5]–[7]. While its practicality, there
might be some inconsistency in the success of CTG monitoring, predominantly in low-risk pregnancies. If

Page 2/17
there is an inaccurately evaluated fetal pain then, it might be results in useless treatments or if there is an
inappropriate investigation of fetal wellbeing then it might be excluded essential treatments [8].

CTG data using three different machine learning techniques to predict fetal distress[9]. An employment of
statistical features extracted from Empirical Mode Decomposition (EMD)[10].The extracted features from
the sub-band decomposition classified as normal or risky. They achieved 86% accuracy for the test data.
Another study presented a two-steps examination of fetal heart rate data which permits for effective
prediction of the acidemia risk. The FHR signals are classified by Support Vector Machines (SVM), fuzzy,
Multilayer perceptron. A new model which utilize the artificial neural network (ANN) to classify the CTG
data[11]. The Recall and F-score were employed to assess the performance. Moreover, they proposed the
k-means clustering for the CTG classification. Adaptive neuro-fuzzy inference systems (ANFIS) is utilized
[12] for the CTG classification. Moreover, SVM and Genetic Algorithm (GA) based classification method
was implemented [13].

There several research done for the prediction of CTG on the basis of dataset & classify algorithms are
reviewed here. Eight different ML techniques are proposed to classify the normal & pathological fetal
state from 1831 dataset with 21 attributes Cardiotocograph dataset. From the result analysis is seen that
KNN achieved the maximum accuracy about 98.4% [6]. A hybrid feature selection PSO based ML
techniques proposed to predict the classification of fetal heart rate. Among all the ML it is seen that KNN
achieved the maximum accuracy about 83.3% [14]. A K Means ML classification algorithm proposed to
find out the accuracy from Cardiotocograph dataset contains 21 attributes. In first phase K means
algorithm used t eliminate the 7 attributes from the dataset & later on ML techniques are used to find out
the accuracy. From the result analysis was conclude that SVM achieved maximum accuracy about
90.64% [15]. A comparative study was performed between SVM & DT on the basis of same dataset used
in previous reference. Both the algorithm offers the quite satisfactory accuracy for the classification of
dataset [16]. RF classifier model has been used to classify the three different states of fetal state from
CTG dataset which contain 2126 dataset with 21 attributes. Maximum accuracy obtained by RF is 93.6 %
for seven potential attributes [17]. A correlation feature selection algorithm with four ML technique
proposed to classify the fetal state either in normal or pathological. Overall research performed for the
same dataset used in previous reference. From the result analysis it is also concludes that moreover all
the algorithms have same classification accuracy is about 94.7% [18]. Five different ML algorithms used
to classify the states of Fetal from CTG dataset which contain 2126 dataset with 21 attributes. Naïve
Bayes acquired the maximum accuracy about 82.32%[19]. A comparative study performed between
RBFN, DT, and NB & MLP to predict the states of fetal heat rate .Maximum accuracy obtained by NB
about 83.9 % when number of potential attributes is 15 [20]. Prediction of FHR has been performed by
hybrid ADB with SVM applied in a CTG datasets which contain 2126 dataset with 21 attributes. Overall
research performed into two stage, in first stage PCA used to sorted the potential attributes & later on
hybrid classification algorithm used [21]..Maximum accuracy obtained by the proposed model is 98.6%
for selected attributes.

Page 3/17
In this paper, several ensemble machine-learning models examined to classify the CTG data as unhealthy
or healthy based on the three obstetricians’ decisions. The contribution of this paper is to implement
Bagging ensemble method to classify the CTG data. To the best of the authors’ knowledge, the Bagging
ensemble classifiers have not been employed previously for the CTG classification. Hence, this paper
compares the performances of the single and ensemble learners in terms of F- measure, accuracy, and
ROC area. Hence, in section 2, materials and methods are presented. In section 3, results and discussion
presented. Section 4 is conclusion.

2. Methods
The dataset was obtained from the University of California Irvine Machine Learning Repository. It
comprised of 2126 pregnant women who were in the third trimester of pregnancy. The dataset consisted
of 35 attributes used in the measurements of FHR and uterine contractions (UCs) on CTG (Figure 1).

According to the Child Health and Human Development, the core risk variable used to derive the state of
fetus includes qualitative and quantitative descriptions of FHR and UCs [22]. The machine learning
algorithms used in this study were Decision Tree, Random Forest, KNN, SVC & Linear SVC. The current
dataset was split into training and testing folds using K-Fold Cross Validation technique to test the
performance of each machine learning model in the training phase.

2.1. Feature Selection Approaches:

The increase in diagnosis cost and the huge volume of data produced by different sources consist of the
number of attributes. All attributes may not be useful, thus it is necessary to remove them during data
preprocessing or feature selection. The feature selected attributes would, in turn, improve the
performance to build a better classification. The various feature selection methods such as embedded,
ensemble and hybrid methods, filter methods and wrapper methods have been applied to study the fetal
heart rate or CTG analysis[23]. In this research we used Symmetrical Uncertainty based feature selection
methods [24] [24] and five classification algorithms such as Decision Tree, Random Forest, KNN, SVC &
Linear SVC to study the CTG data analysis.

2.2. Data Sources:

The publicly UCI machine learning repository has been used to retrieve the Cardiotocograph (CTG)
dataset available at https://archive.ics.uci.edu/ml/datasets/Cardiotocography. The multivariate data
type consists of 2126 instances with 35 attributes, which are numeric. The class attribute consists of 3
distinct values, which are Normal, Suspect, and Pathologic. The frequencies of 2126 instances are as
follows: 1655 normal, 295 suspicious and 176 pathologic, indicating the uneven distribution of the
observations across the classes, which refers to class imbalance dataset. The imbalanced datasets
require special attention because the regular classifiers accuracies are inappropriate to use for class
imbalance [5], [25], [26], since these classifiers generally favor the majority class i.e., the class with a large
number of instances. The performance of the classifier can be improved by the ensemble of classifiers.
Page 4/17
However, the majority of ensembles is static and cannot be applied to imbalanced datasets [27]. Apart
from this, based on experimental results, it is known that the performance on the balanced dataset is
better than the imbalance dataset [28]. In the view of aforementioned sentences, the dataset used in this
study consists of 248 normal fetal state class randomly derived from 1655 instances from UCI repository
CTG dataset), keeping other class codes as the same (i.e., 295 suspicious and 176 pathologic) with 23
attributes is shown in Table 1

Table 1
Characteristics of dataset after SMOTE
  UCI machine repository Dataset after SMOTE

Attributes 35 23

Normal 1655 248

Suspicious 295 295

Pathologic 176 176

Total instances 2126 719

2.3 Attributes Description:

The dataset consists of 23 attributes. The predictable attribute is referred to “NSP: Fetal state class code
(N = normal; S = suspect; P = pathologic)”. The description of the attributes is shown in Table 2.

Table 2
Class description of fetal heart rate
Class values Description No of dataset

(N) Normal 248

(S) Suspicious 295

(P) Pathologic 176

The derived dataset consists of 719 instances with 23 attributes (Table 1) has been taken into
consideration to build a classification model after normalization of the data. Python-based Scikit learn
was used as an analytical tool. A total of seven machine learning (ML) techniques, each (refer to
literature review) was used to evaluate the performance of the classifiers and tools. Later, feature
selection was also implemented on the aforementioned dataset.

3. Result Analysis
To the best of author knowledge, most of the classification model studies have been carried out on the
UCI machine learning repository CTG dataset [29], [30]. Thus, there were no studies addressing the derived
dataset with the five machine learning techniques. To measure the performance of each classification
Page 5/17
algorithm, the accuracy has been taken into accordance. The key outcome of this study was to compare
major machine learning algorithms (listed above) with regard to their precision accuracy and sensitivity
to predict normal, suspect, or pathologic fetal state based on CTG attributes. Various statistical
techniques were used to compare the performance of the algorithms. These included precision, sensitivity
or recall, F1 score, and overall accuracy ([true positive + true negative]/[true positive + true negative +
false positive + false negative]).

On the provided dataset, the experiment is run, and the results are produced. Each experiment is
evaluated to stratified K-fold validation to ensure that the results are free of bias. The major goal is to
remove any bias in the outcomes, as feature engineering sometimes leads to the omission of specific
characteristics, which might affect overall prediction results. Furthermore, the process of feature
engineering is typically highly costly. Machine learning algorithms are provided raw data after some
preparation. The findings are then obtained and compared to current state-of-the-art systems. The
dataset was examined, and methods were used when needed, and the model was trained to improve the
precision.

3.1 Performance Measure 

One of the most blatant misrepresentations about machine learning model assessment is that every
dataset, regardless of its type, can be quantified using the same evaluation matrices. The majority of
machine learning models are judged on their accuracy[31]–[39]. When working with an unbalanced
dataset, this deliberate proves to be deceiving. As a result, several appropriate assessment matrices, as
well as accuracy, are employed. Precision, recall, the F1 measure, and the ROC curve were used to
evaluate the proposed study[40], [41]. The accuracy ratio is the number of correct predictions divided by
the total number of inputs. The confusion matrices are obtained by calculating the true positive (TP), true
negative (TN), false positive (FP), and false negative (FN) values (FN). TP/ (FN + TP) and FP/ (FP + TN)
are two considerations that are computed as TP/(FN + TP) and FP/(FP + TN). Another statistic common
screening a model's classification accuracy is the receiver operation curve (ROC)[42].

Table 3.  Performance of  SMOTE based Classifier Algorithms

Algorithm Accuracy  Precision  Recall F1 Score AUC

Decision Tree 96.8% 96.8% 96.5% 96.4% 0.89
Random Forest 98.01% 97.8% 97.7% 97.5% 0.96
KNN 96.2% 96.2% 96% 96% 0.92
SVC 97.7% 97.5% 97% 97% 0.96
Linear SVC 97% 97% 97% 97% 0.64

Table 3 shows the tabular form result analysis of Average Accuracy, Precision ,F1 score , Precision Area
under ROC & Computational Time for SMOTE based Logistic  Regression , Random Forest , Decision tree,

Page 6/17
KNN & SVM models when trained and tested on the Tabular data consisting of actual 540 records.
Results are obtained after principle component analysis & SMOTE. Average parameters are calculated for
both negative & positive classes’ cases. We found that SMOTE based Random Forest performed best
among the entire SMOTE based algorithm with the Average Accuracy, precision, Recall & F1 Score values
of 98.01%, 97.8%, 97.7% & 97.5 % respectively. However the SMOTE based Random Forest and SMOTE
based KNN & SVM have least computational time & maximum area under ROC with the values of 0.10
sec & 96% respectively.

Figure 3 shows graphical comparison for the entire SMOTE based machine learning algorithm on the
basis of computational time. From the plot we found that SMOTE based Random Forest have least
computational time of 0.010 sec while SMOTE based Decision Tree have a maximum time of 0.031 sec. 

3.2. Classification model evaluation

 The reason for the assessment of a classification model is to achieve a solid evaluation of the model
that is known as the model’s predictive performance. Diverse execution parameters can be utilized. 

Provided that the model is dependent on training set and has speculation property which is basis for the
quality assessment. For any assessment measure, it is imperative to recognize its incentive for a specific
dataset performance, particularly the training set performance, and its true generalization performance.
The created model’s training performance is determined by assessing the model on the training set.
However, the aim of classification models is not to categorize the training data. Suitable evaluation
processes are required to dependably evaluate the unfamiliar values of the assumed performance
measures on the whole domain [43], [44]. 

3.3 Statistical Measurement 

The Mathews correlation coefficient (MCC) is a metric for evaluating binary classification quality[45]–
[47]. The Matthews correlation coefficient is a contingency matrix technique of calculating the Pearson
product-moment correlation coefficient between actual and predicted values that is unaffected by the
unbalanced datasets issue. MCC is the only binary classification rate that awards a high score only if the
binary predictor accurately predicts the majority of positive and negative data instances. It has a range of
[1, +1], with extreme values of –1 and +1 for perfect misclassification and perfect classification,
respectively, and for coin tossing classifier MCC=0. Equation (1) is demonstrating the MCC.

Page 7/17
The kappa (k) statistic is a key parameter for judging the model’s consistency[48]–[50]. It compares the
outcome of the suggested model to the outcome of the randomly classified technique. The kappa
statistic’s value ranged from 0 to 1. The model’s expected effect is represented by a value near to 1,
whereas 0 indicates that the model is flawed. (2), (3), and (4) demonstrate the kappa statistic’s equation.

In present research range of kappa value 0.702 to 1 indicates proposed model attains great consistency.
Both the values of MCC & kappa for all the algorithms are shown in Table 4.

Table 5. Statistical Measure after experimentation

Algorithm MCC Kappa(k)
Decision Tree 0.741 0.958
Random Forest 0.968 1
KNN 0.789 0.95
SVC 0.957 1
Linear SVC 0.541 0.857

3.4 Comparison with existing system

The proposed work’s findings are being compared to the results of other state-of-the-art existing system in
order to ascertain the proposed work’s trustworthiness.

Table 6.  Comparative study between proposed models with existing model

Page 8/17
 Reference  Algorithm used Outcomes from the research 
[14] PSO based KNN & SVM  Overall accuracy for PSO
feature selection based KNN
achieved the maximum accuracy
88.5%
[15] SVM & hybrid K means SVM  Maximum accuracy obtained by
the K means SVM with 90.64% ,
where k=10
[17] Random Forest  Random Forest with seven
important feature  classify the
CTG data with maximum
accuracy 93.6%
[18] Bagging approach with three different All   the proposed algorithm
decision tree algorithms : Random forest , achieved overall accuracy was
REP Tree & J48 & correlation feature about to 90%.
selection were used 
[20] Naive Bayes, Decision Tree, Multi Layer Maximum accuracy obtained by
Perceptron and Radial Basis Function Decision Tree   for 15 potential
attributes about 93.3%
Present Decision Tree, Random Forest , SVC , KNN, Maximum accuracy obtained by
Research  Linear SVC  SMOTE based Random Forest is
about 98.01% for 23 attributes

4. Conclusion
Classification of accuracy from CTG dataset is a one of major challenges in the medical diagnosis
system. Delayed detection of pathologic fetal state based on CTG attributes may caused serious health
issue of mother & baby so early diagnosis is important. In modern research for early detection in medical
diagnosis ML techniques have been introduced. ML techniques are the subsection of AI which has
capability to learn the large amount of unlabeled & unstructured data in few seconds. In this research we
proposed existing techniques for diagnosis of early detection of pathologic fetal state on CTG datasets.

In last decades there are several research performed for the detection of pathologic fetal state in terms of
accuracy. All the approaches used same dataset (CTG) for their training & testing model. In Ml there are
number shortcomings for the prediction of Diabetes like accuracy & identification of potential attributes
etc. Hence, a model must be designed in such a manner in future so that it able to overcome these
shortcomings. In this research we performed the overall research into three stages: In first stage
imbalance CTG data oversampled by SMOTE, in second stage we used hyper parameter tuning of
training dataset to reduce the model's complexity & make a trade-off between these components and
final stage we applied six Machine learning technique for testing data classification. For the classification
of fetal state of CTG dataset we used five different ML techniques, DT, SVC, KNN, RF & Linear SVC. 

All the proposed model are evaluated on the basis of confusion matrix parameters like  accuracy, recall,
precision, F1 score, and AUC, computational time parameters like  training & testing time & statistical
parameters for model consistency like MCC & k.  Here we identify the best model on the basis of six
parameters which is explain in previous sections,  SMOTE based RF model achieved best accuracy of

Page 9/17
98.01%  as compared to other models used. Further maximum AUC provided by RF & SVC model
indicates it is optimal classifiers for CTG dataset. However both RF & SVC also achieved the maximum
value of MCC & k indicating higher degree of model consistency. On the contrary, future aspects of
present research can be upgraded by the collecting the large real-time dataset from IoT based device &
prediction can be done the performance of ML techniques by means of reduction of signal bandwidth &
less computational time. 

References
1. M. F. Hurtado-Sánchez, D. Pérez-Melero, A. Pinto-Ibáñez, E. González-Mesa, J. Mozas-Moreno, and A.
Puertas-Prieto, “Characteristics of Heart Rate Tracings in Preterm Fetus,” Medicina (Mex.), vol. 57, no.
6, p. 528, 2021.
2. Y. Lu, Y. Qi, and X. Fu, “A framework for intelligent analysis of digital cardiotocographic signals from
IoMT-based foetal monitoring,” Future Gener. Comput. Syst., vol. 101, pp. 1130–1141, 2019.
3. S. Chan, M. Arjuna, and N. L. Nik Ahmad Zuky, “Cardiotocography waveform analysis using image
extraction technique,” 2020.
4. H. Sahin and A. Subasi, “Classification of the cardiotocogram data for anticipation of fetal risks
using machine learning techniques,” Appl. Soft Comput., vol. 33, pp. 231–238, 2015.
5. F. Marzbanrad, L. Stroux, and G. D. Clifford, “Cardiotocography and beyond: a review of one-
dimensional Doppler ultrasound application in fetal monitoring,” Physiol. Meas., vol. 39, no. 8,
p. 08TR01, 2018.
6. A. Subasi, B. Kadasa, and E. Kremic, “Classification of the cardiotocogram data for anticipation of
fetal risks using bagging ensemble classifier,” Procedia Comput. Sci., vol. 168, pp. 34–39, 2020.
7. V. Nagendra, H. Gude, D. Sampath, S. Corns, and S. Long, “Evaluation of support vector machines
and random forest classifiers in a real-time fetal monitoring system based on cardiotocography
data,” in 2017 IEEE Conference on Computational Intelligence in Bioinformatics and Computational
Biology (CIBCB), 2017, pp. 1–6.
8. M. Belzile, A. Pouliot, A. Cumyn, and A. M. Côté, “Renal physiology and fluid and electrolyte disorders
in pregnancy,” Best Pract. Res. Clin. Obstet. Gynaecol., vol. 57, pp. 1–14, 2019.
9. A. E. Permanasari and A. Nurlayli, “Decision tree to analyze the cardiotocogram data for fetal distress
determination,” in 2017 international conference on sustainable information engineering and
technology (SIET), 2017, pp. 459–463.
10. S. Aziz, M. U. Khan, Z. A. Choudhry, A. Aymin, and A. Usman, “ECG-based biometric authentication
using empirical mode decomposition and support vector machines,” in 2019 IEEE 10th Annual
Information Technology, Electronics and Mobile Communication Conference (IEMCON), 2019,
pp. 0906–0912.
11. H. Ellethy, S. S. Chandra, and F. A. Nasrallah, “The detection of mild traumatic brain injury in
paediatrics using artificial neural networks,” Comput. Biol. Med., vol. 135, p. 104614, 2021.

Page 10/17
12. Y. Fei et al., “Automatic Classification of Antepartum Cardiotocography Using Fuzzy Clustering and
Adaptive Neuro-Fuzzy Inference System,” in 2020 IEEE International Conference on Bioinformatics
and Biomedicine (BIBM), 2020, pp. 1938–1942.
13. A. O. de Carvalho Filho, A. C. Silva, A. C. de Paiva, R. A. Nunes, and M. Gattass, “Computer-aided
diagnosis of lung nodules in computed tomography by using phylogenetic diversity, genetic
algorithm, and SVM,” J. Digit. Imaging, vol. 30, no. 6, pp. 812–822, 2017.
14. G. Georgoulas, C. Stylios, V. Chudacek, M. Macas, J. Bernardes, and L. Lhotska, “Classification of
fetal heart rate signals based on features selected using the binary particle swarm algorithm,” in
World Congress on Medical Physics and Biomedical Engineering 2006, 2007, pp. 1156–1159.
15. N. Chamidah and I. Wasito, “Fetal state classification from cardiotocography based on feature
extraction using hybrid K-Means and support vector machine,” in 2015 international conference on
advanced computer science and information systems (ICACSIS), 2015, pp. 37–41.
16. D. Jagannathan and M. Phil, “Cardiotocography-a comparative study between support vector
machine and decision tree algorithms,” Int. J. Trend Res. Dev., vol. 4, no. 1, 2017.
17. M. Arif, “Classification of cardiotocograms using random forest classifier and selection of important
features from cardiotocogram signal,” Biomater. Biomech. Bioeng., vol. 2, no. 3, pp. 173–183, 2015.
18. S. A. A. Shah, W. Aziz, M. Arif, and M. S. A. Nadeem, “Decision trees based classification of
cardiotocograms using bagging approach,” in 2015 13th international conference on frontiers of
information technology (FIT), 2015, pp. 12–17.
19. D. Bhatnagar and P. Maheshwari, “Classification of cardiotocography data with WEKA,” Int. J.
Comput. Sci. Netw.-IJCSN, vol. 5, no. 2, 2016.
20. V. Subha, D. Murugan, J. Rani, K. Rajalakshmi, and T. Tirunelveli, “Comparative analysis of
classification techniques using Cardiotocography dataset,” Int. Jour Res. Inf. Technol., vol. 1, no. 12,
pp. 274–280, 2013.
21. Y. Zhang and Z. Zhao, “Fetal state assessment based on cardiotocography parameters using PCA
and AdaBoost,” in 2017 10th international congress on image and signal processing, BioMedical
engineering and informatics (CISP-BMEI), 2017, pp. 1–6.
22. Z. Hoodbhoy, M. Noman, A. Shafique, A. Nasim, D. Chowdhury, and B. Hasan, “Use of machine
learning algorithms for prediction of fetal risk using cardiotocographic data,” Int. J. Appl. Basic Med.
Res., vol. 9, no. 4, p. 226, 2019.
23. I. Rafique, M. Dilawar, A. Umer, and M. A. Hassan, “Classification of Cardiotocography Data for Fetal
Health Using Feature Selection Techniques,” in Computer Science On-line Conference, 2021, pp. 34–
44.
24. M. S. Devi, S. Sridevi, K. K. Bonala, R. H. Dadi, and K. V. K. Reddy, “Oversampling Response Stretch
based Fetal Health Prediction using Cardiotocographic Data,” Ann. Romanian Soc. Cell Biol.,
pp. 1448–1464, 2021.
25. J. Xu, Z. Chen, J. Zhang, Y. Lu, X. Yang, and A. Pumir, “Realistic preterm prediction based on
optimized synthetic sampling of EHG signal,” Comput. Biol. Med., vol. 136, p. 104644, 2021.
Page 11/17
26. K. Madasamy and M. Ramaswami, “Data imbalance and classifiers: impact and solutions from a big
data perspective,” Int. J. Comput. Intell. Res., vol. 13, no. 9, pp. 2267–2281, 2017.
27. B. Krawczyk, A. Cano, and M. Woźniak, “Selecting local ensembles for multi-class imbalanced data
classification,” in 2018 International joint conference on neural networks (IJCNN), 2018, pp. 1–8.
28. D. Ballabio, F. Grisoni, and R. Todeschini, “Multivariate comparison of classification performance
measures,” Chemom. Intell. Lab. Syst., vol. 174, pp. 33–44, 2018.
29. K. Agrawal and H. Mohan, “Cardiotocography analysis for fetal state classification using machine
learning algorithms,” in 2019 International Conference on Computer Communication and Informatics
(ICCCI), 2019, pp. 1–6.
30. N. Sevani, I. Hermawan, and W. Jatmiko, “Feature Selection based on F-score for Enhancing CTG
Data Classification,” in 2019 IEEE International Conference on Cybernetics and Computational
Intelligence (CyberneticsCom), 2019, pp. 18–22.
31. P. Dutta and A. Kumar, “Application of an ANFIS model to optimize the liquid flow rate of a process
control system,” Chem. Eng. Trans., vol. 71, pp. 991–996, 2018.
32. P. Dutta and A. Kumar, “Design an intelligent calibration technique using optimized GA-ANN for liquid
flow control system,” J. Eur. Systèmes Autom., vol. 50, no. 4–6, p. 449, 2017.
33. P. Dutta and A. Kumar, “Design an intelligent flow measurement technique by optimized fuzzy logic
controller,” J. Eur. Systèmes Autom., vol. 51, no. 1–3, p. 89, 2018.
34. P. Dutta and A. Kumar, “Intelligent calibration technique using optimized fuzzy logic controller for
ultrasonic flow sensor,” Math. Model. Eng. Probl., vol. 4, no. 2, pp. 91–94, 2017.
35. P. Dutta and A. Kumar, “Modeling and optimization of a liquid flow process using an artificial neural
network-based flower pollination algorithm,” J. Intell. Syst., vol. 29, no. 1, pp. 787–798, 2020.
36. S. Mandal, P. Dutta, and A. Kumar, “Modeling of liquid flow control process using improved versions
of elephant swarm water search algorithm,” SN Appl. Sci., vol. 1, no. 8, pp. 1–16, 2019.
37. P. Dutta and A. Kumar, “Modelling of Liquid Flow control system Using Optimized Genetic Algorithm,”
Stat. Optim. Inf. Comput., vol. 8, no. 2, pp. 565–582, 2020.
38. P. Dutta, R. Agarwala, M. Majumder, and A. Kumar, “PARAMETERS EXTRACTION OF A SINGLE DIODE
SOLAR CELL MODEL USING BAT ALGORITHM, FIREFLY ALGORITHM & CUCKOO SEARCH
OPTIMIZATION,” Ann. Fac. Eng. Hunedoara, vol. 18, no. 3, pp. 147–156, 2020.
39. P. Dutta, S. K. Biswas, S. Biswas, and M. Majumder, “Parametric optimization of Solar Parabolic
Collector using metaheuristic Optimization”.
40. P. Dutta, S. Paul, and A. Kumar, “Comparative analysis of various supervised machine learning
techniques for diagnosis of COVID-19,” in Electronic Devices, Circuits, and Systems for Biomedical
Applications, Elsevier, 2021, pp. 521–540.
41. P. Dutta, S. Paul, A. J. Obaid, S. Pal, and K. Mukhopadhyay, “Feature Selection based Artificial
Intelligence Techniques for the Prediction of COVID like Diseases,” in Journal of Physics: Conference
Series, 2021, vol. 1963, no. 1, p. 012167.

Page 12/17
42. P. DUTTA and A. KUMAR, “Flow sensor Analogue: Realtime prediction Analysis using SVM & KNN,”
presented at the Emerging trends in Engineering and Science (ETES 2018), 2018.
43. F. DIMAIO et al., “Accounting for Safety Barriers Degradation in the Risk Assessment of Oil and Gas
Systems by Multistate Bayesian Networks,” Reliab. Eng. Syst. Saf., vol. 216, p. 107943, 2021.
44. P. Schneider and K. Böttinger, “High-performance unsupervised anomaly detection for cyber-physical
system networks,” in Proceedings of the 2018 workshop on cyber-physical systems security and
privacy, 2018, pp. 1–12.
45. D. Chicco and G. Jurman, “The advantages of the Matthews correlation coefficient (MCC) over F1
score and accuracy in binary classification evaluation,” BMC Genomics, vol. 21, no. 1, pp. 1–13,
2020.
46. D. Chicco, N. Tötsch, and G. Jurman, “The Matthews correlation coefficient (MCC) is more reliable
than balanced accuracy, bookmaker informedness, and markedness in two-class confusion matrix
evaluation,” BioData Min., vol. 14, no. 1, pp. 1–22, 2021.
47. C. Halimu, A. Kasem, and S. S. Newaz, “Empirical comparison of area under ROC curve (AUC) and
Mathew correlation coefficient (MCC) for evaluating machine learning algorithms on imbalanced
datasets for binary classification,” in Proceedings of the 3rd international conference on machine
learning and soft computing, 2019, pp. 1–6.
48. H. Wu, S. Yang, Z. Huang, J. He, and X. Wang, “Type 2 diabetes mellitus prediction model based on
data mining,” Inform. Med. Unlocked, vol. 10, pp. 100–107, 2018.
49. Y. Zheng, R. Zhang, M. Huang, and X. Mao, “A pre-training based personalized dialogue generation
model with persona-sparse data,” in Proceedings of the AAAI Conference on Artificial Intelligence,
2020, vol. 34, no. 05, pp. 9693–9700.
50. G. Qian, W.-S. Lei, M. Niffenegger, and V. F. González-Albuixech, “On the temperature independence of
statistical model parameters for cleavage fracture in ferritic steels,” Philos. Mag., vol. 98, no. 11,
pp. 959–1004, 2018.

Declarations
Funding:
For this research authors does not get any fund.

Conflict of Interest:

All the author declares that there has no conflict of interest.

Ethical approval:

This article does not contain any studies with human participants or animals performed by any of the
authors.

Page 13/17
Figures

Figure 1

Attributes of CTG datasets

Page 14/17
Figure 2

Flowchart of the proposed model

Page 15/17
Figure 3

graphical comparison for the entire SMOTE based machine learning algorithm on the basis of
computational time

Figure 4

Page 16/17
Comparative study of proposed ML techniques on ROC Curve

Figure 5

Comparative study based on Statistical parameters

Figure 6

Comparative study between state of art techniques with present research

Page 17/17