International Journal of Innovative Research in Computer and Communication Engineering
ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines
| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |
| TITLE | Detection of Perinatal Oxygen Deprivation Using Cardiotocography Signals: A Machine Learning Approach |
|---|---|
| ABSTRACT | Perinatal oxygen deprivation, widely known as neonatal asphyxia, constitutes one of the foremost contributors to preventable neonatal mortality and long-term neurodevelopmental impairment worldwide. Cardiotocography (CTG) is the primary clinical instrument for intrapartum fetal surveillance; however, its manual interpretation is afflicted by substantial inter-clinician variability, limiting its effectiveness as a reliable screening tool. This paper presents an automated detection approach of perinatal oxygen deprivation by applying and systematically comparing six supervised machine learning classifiers — Gradient Boosting (GB), Random Forest (RF), Decision Tree (DT), Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Logistic Regression (LR) — on the UCI Cardiotocography benchmark dataset (2,126 records, 21 features). The Synthetic Minority Over-sampling Technique (SMOTE) was employed to rectify the inherent class imbalance, equalizing all three fetal state classes to 935 training samples each. Gradient Boosting achieved the highest classification accuracy of 96.71% with a macro-AUC of 0.9975 and Pathologic-class sensitivity of 96.00%. Random Forest yielded near-equivalent performance (96.48% accuracy, AUC 0.9954) with superior model interpretability through Gini-based feature importance analysis. Abnormal Short-Term Variability (ASTV) and FHR histogram statistics emerged as the most discriminative predictors of fetal compromise. The proposed framework demonstrates competitive performance compared to existing methods on this benchmark while maintaining full transparency — a prerequisite for clinical deployment. |
| AUTHOR | PROF. MANJULA P, USHA K, YASHASWINI A N, SINCHANA S H, SIDDESH R, CHANDRAKANTHA C V Assistant Professor, Dept. of CS&E., Jain Institute of Technology, Davangere, Karnataka, India. UG Student, Dept. of CS&E, Jain Institute of Technology, Davangere, Karnataka, India. |
| VOLUME | 184 |
| DOI | DOI: 10.15680/IJIRCCE.2026.1405007 |
| pdf/7_Detection of Perinatal Oxygen Deprivation Using Cardiotocography Signals A Machine Learning Approach.pdf | |
| KEYWORDS | |
| References | [1] WHO. (2023). Newborn mortality. WHO Global Health Observatory. https://www.who.int/news-room/fact-sheets/detail/newborn mortality. [2] Devane, D., Lalor, J. G., Daly, S., et al. (2017). Cardiotocography versus intermittent auscultation of fetal heart on admission to labour ward. Cochrane Database Syst. Rev., (1), CD005122. [3] Magenes, G., Signorini, M. G., & Cerutti, S. (2000). Identification of fetal sufferance antepartum through multiparametric analysis and SVM. Proc. 22nd Annual IEEE EMBC, Chicago, 2246–2249. [4] Ayres-de-Campos, D., Bernardes, J., Garrido, A., et al. (2000). SisPorto 2.0: A program for automated analysis of cardiotocograms. J. Maternal-Fetal Med., 9(5), 311–318. [5] Hoodbhoy, Z., Noman, M., Shafique, A., et al. (2019). Use of machine learning for prediction of fetal risk using cardiotocographic data. Int. J. Applied & Basic Medical Research, 9(4), 226–230. [6] Huang, M. L., & Hsu, Y. Y. (2020). Fetal distress prediction using discriminant analysis, decision tree, and artificial neural network. J. Biomedical Science and Engineering, 5(9), 526–533. [7] Aeberhard, J. L., et al. (2021). Machine learning for antepartum fetal surveillance. Acta Obstet. Gynecol. Scand., 100(6), 1053–1061. [8] Zhao, Z., Zhang, Y., & Comert, Z. (2022). Fetal state assessment from cardiotocogram data using artificial neural network. IEEE Access, 7, 52368–52377. [9] Petrozziello, A., Jordanov, I., Chandraharan, E., et al. (2023). Multimodal CNN for fetal monitoring during labor. Artificial Intelligence in Medicine, 132, 102389. [10] Ayres-de-Campos, D., Spong, C. Y., Chandraharan, E., & FIGO Panel. (2015). FIGO consensus guidelines on intrapartum fetal monitoring: Cardiotocography. Int. J. Gynecol. Obstet., 131(1), 13–24. |
