Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM66683
Volume:
10 |
8 |
Aug 2026
Pages:
1-9
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Science and Engineering
Bhopal Institute of Technology & Science
Computer Science and Engineering
Bhopal Institute of Technology & Science
Cardiotocography (CTG) is a widely used non-invasive technique for monitoring fetal heart rate and uterine contractions during pregnancy, providing critical information to detect fetal distress and prevent adverse outcomes. This study benchmarks multiple machine learning and ensemble techniques for automated, accurate classification of fetal health status using CTG-derived features from the publicly available UCI Cardiotocography dataset, comprising 2126 records labeled as Normal, Suspect, or Pathological. Following preprocessing with feature scaling and an 80:20 stratified train-test split, five base classifiers were trained and evaluated: Logistic Regression, Support Vector Machine (SVM), Random Forest, K-Nearest Neighbors (KNN), and XGBoost. To further improve predictive performance, two ensemble strategies combining XGBoost and Random Forest were implemented: a soft-voting Blending classifier and a Stacking classifier with Logistic Regression as the meta-learner, validated using 5-fold cross-validation. Model performance was assessed using accuracy, precision, recall, F1-score, confusion matrices, ROC-AUC, and the Matthews Correlation Coefficient (MCC) to account for class imbalance. Results show XGBoost achieved the best standalone performance, with 96.00% accuracy and an MCC of 0.8893, slightly outperforming both the Blending and Stacking ensembles (95.77% accuracy, MCC of 0.8839 and 0.8829 respectively), while Random Forest followed closely at 95.00% accuracy. Logistic Regression yielded the weakest baseline at 88.00% accuracy. Although ensemble methods did not surpass standalone XGBoost in this experiment, Stacking demonstrated stable generalization with a mean cross-validation accuracy of 94.29%. These findings indicate that gradient boosting effectively captures the complex, imbalanced, multi-class structure of CTG data and holds strong potential for integration into clinical decision-support systems for early, reliable detection of fetal distress.
KEYWORDS
This study benchmarked five machine learning classifiers and two ensemble techniques for fetal health classification using cardiotocogram data, grounding each method in its underlying mathematical formulation and evaluating performance using accuracy, precision, recall, F1-score, confusion matrices, and the Matthews Correlation Coefficient. XGBoost emerged as the most accurate and reliable standalone model, achieving 96.00% accuracy and an MCC of 0.8893, slightly ahead of the Blending and Stacking ensembles, and achieved perfect sensitivity on the clinically critical Pathological class. These findings highlight the strength of regularized gradient boosting for structured, imbalanced clinical datasets and support its potential integration into CTG-based clinical decision-support systems. Future work should explore systematic hyperparameter optimization, external validation on independent CTG datasets, class-imbalance-aware resampling, and interpretability methods to increase clinical trust and support real-world adoption.
[1] World Health Organization, “Stillbirth,” WHO Fact Sheets, Geneva, Switzerland, 2023. [Online]. Available: https://www.who.int/news-room/fact-sheets/detail/stillbirth [2] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, 1995. [3] T. Cover and P. Hart, “Nearest neighbor pattern classification,” IEEE Transactions on Information Theory, vol. 13, no. 1, pp. 21–27, 1967. [4] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [5] L. Breiman, J. H. Friedman, R. A. Olshen, and C. J. Stone, Classification and Regression Trees. Belmont, CA: Wadsworth International Group, 1984. [6] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, 2016, pp. 785–794. [7] D. H. Wolpert, “Stacked generalization,” Neural Networks, vol. 5, no. 2, pp. 241–259, 1992. [8] T. G. Dietterich, “Ensemble methods in machine learning,” in Multiple Classifier Systems, LNCS vol. 1857, 2000, pp. 1–15. [9] B. W. Matthews, “Comparison of the predicted and observed secondary structure of T4 phage lysozyme,” Biochimica et Biophysica Acta, vol. 405, no. 2, pp. 442–451, 1975. [10] T. Fawcett, “An introduction to ROC analysis,” Pattern Recognition Letters, vol. 27, no. 8, pp. 861–874, 2006.
Publication Date
8/13/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
2
Total views since publication
PDF Downloads
7
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...