Email this article 
(Universitas Nusa Mandiri, Indonesia)(2) Mila Desi Anasanti
(Bart and London Genome Center, Queen Mary University of London, London, United Kingdom)*corresponding author
AbstractThe global concern over declining perinatal death rates, particularly in low- and middle-income nations, underscores the importance of adopting Cardiotocography (CTG) as a vital fetal monitoring method. Recent strides in machine learning (ML) present promising opportunities to enhance the accuracy of assessing fetal health, providing a viable alternative to traditional approaches. This study aims to evaluate various ML methodologies and feature selection techniques for predicting fetal health using CTG data. The primary objective is to improve ML algorithms' accuracy, precision, recall, and F1 score while selecting the most critical features. The dataset includes 2,126 expectant mothers in the third trimester, with 35 variables related to fetal heart rate (FHR) and uterine contractions (UC). Preprocessing involves feature scaling, data balancing, and outlier elimination. Additionally, a 10-fold stratified cross-validation approach is employed to ensure robust evaluation and generalizability of the model's performance. Six ML algorithms—Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), Naive Bayes (NB), Logistic Regression (LR), and K-Nearest Neighbors (KNN)—are employed, optimized through grid search cross-validation. The RF algorithm outperforms with an impressive 99% accuracy, closely followed by DT at 98.7%. Optimizing 15 features from the original 35 using Simultaneous Perturbation Feature Selection and Ranking (spFSR) yields a remarkable accuracy of 99%, mirroring the full feature set. This underscores the vital role of selected features in improving predictive power and overall model performance. The study emphasizes the efficacy of tree-based classification algorithms, especially RF, in predicting fetal health and highlights the impact of preprocessing on model performance. These findings suggest avenues for future research, including exploring alternative feature engineering methods and assessing algorithm performance in diverse scenarios.
KeywordsPerinatal mortality; cardiotocography (CTG); machine learning; fetal health prediction; feature selection;
|
DOIhttps://doi.org/10.29099/ijair.v8i1.1110 |
Article metrics10.29099/ijair.v8i1.1110 Abstract views : 864 | PDF views : 39 |
Cite |
Full Text Download
|
References
nina, “Levels and trends in child mortality,” UNICEF DATA. Accessed: Oct. 20, 2023. [Online]. Available: https://data.unicef.org/resources/levels-and-trends-in-child-mortality/
U. Syed et al., “Advancing maternal and perinatal health in low- and middle-income countries: A multi-country review of policies and programmes,” Front. Glob. Womens Health, vol. 3, 2022, Accessed: Dec. 23, 2023. [Online]. Available: https://www.frontiersin.org/articles/10.3389/fgwh.2022.909991
Y. G. Robi and T. M. Sitote, “Neonatal Disease Prediction Using Machine Learning Techniques,” J. Healthc. Eng., vol. 2023, pp. 1–16, Feb. 2023, doi: 10.1155/2023/3567194.
“Acta Obstet Gynecol Scand - 2015 - Goldenberg - Reducing stillbirths in low?income countries.pdf.”
Z. Alfirevic, G. M. Gyte, A. Cuthbert, and D. Devane, “Continuous cardiotocography (CTG) as a form of electronic fetal monitoring (EFM) for fetal assessment during labour,” Cochrane Database Syst. Rev., vol. 2019, no. 5, Feb. 2017, doi: 10.1002/14651858.CD006066.pub3.
D. Lewis, S. Downe, and FIGO Intrapartum Fetal Monitoring Expert Consensus Panel, “FIGO consensus guidelines on intrapartum fetal monitoring: Intermittent auscultation,” Int. J. Gynecol. Obstet., vol. 131, no. 1, pp. 9–12, Oct. 2015, doi: 10.1016/j.ijgo.2015.06.019.
G. A. Macones, G. D. V. Hankins, C. Y. Spong, J. Hauth, and T. Moore, “The 2008 National Institute of Child Health and Human Development workshop report on electronic fetal monitoring: update on definitions, interpretation, and research guidelines,” Obstet. Gynecol., vol. 112, no. 3, pp. 661–666, Sep. 2008, doi: 10.1097/AOG.0b013e3181841395.
R. M. Grivell, Z. Alfirevic, G. M. Gyte, and D. Devane, “Antenatal cardiotocography for fetal assessment,” Cochrane Database Syst. Rev., vol. 2019, no. 5, Sep. 2015, doi: 10.1002/14651858.CD007863.pub4.
A. Sheikhtaheri, M. R. Zarkesh, R. Moradi, and F. Kermani, “Prediction of neonatal deaths in NICUs: development and validation of machine learning models,” BMC Med. Inform. Decis. Mak., vol. 21, no. 1, p. 131, Apr. 2021, doi: 10.1186/s12911-021-01497-8.
N. Rahmayanti, H. Pradani, M. Pahlawan, and R. Vinarti, “Comparison of machine learning algorithms to classify fetal health using cardiotocogram data,” Procedia Comput. Sci., vol. 197, pp. 162–171, 2022, doi: 10.1016/j.procs.2021.12.130.
T. J. Park et al., “Machine Learning Model for Classifying the Results of Fetal Cardiotocography Conducted in High-Risk Pregnancies,” Yonsei Med. J., vol. 63, no. 7, p. 692, 2022, doi: 10.3349/ymj.2022.63.7.692.
S. Das, H. Mukherjee, K. Roy, and C. K. Saha, “Fetal Health Classification from Cardiotocograph for Both Stages of Labor—A Soft-Computing-Based Approach,” Diagnostics, vol. 13, no. 5, p. 858, Feb. 2023, doi: 10.3390/diagnostics13050858.
J. B. D. Campos, “Cardiotocography.” UCI Machine Learning Repository, 2000. doi: 10.24432/C51S4N.
“Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD),” National Institutes of Health (NIH). Accessed: Dec. 23, 2023. [Online]. Available: https://www.nih.gov/about-nih/what-we-do/nih-almanac/eunice-kennedy-shriver-national-institute-child-health-human-development-nichd
“3.2 - Identifying Outliers: IQR Method | STAT 200.” Accessed: Dec. 24, 2023. [Online]. Available: https://online.stat.psu.edu/stat200/lesson/3/3.2
N. Shrestha, “Detecting Multicollinearity in Regression Analysis,” Am. J. Appl. Math. Stat., vol. 8, pp. 39–42, Jun. 2020, doi: 10.12691/ajams-8-2-1.
“Technologies | Free Full-Text | Effect of Data Scaling Methods on Machine Learning Algorithms and Model Performance.” Accessed: Dec. 24, 2023. [Online]. Available: https://www.mdpi.com/2227-7080/9/3/52
A. I. Naimi, R. W. Platt, and J. C. Larkin, “Machine Learning for Fetal Growth Prediction,” Epidemiol. Camb. Mass, vol. 29, no. 2, pp. 290–298, Mar. 2018, doi: 10.1097/EDE.0000000000000788.
Z. Arain, S. Iliodromiti, G. Slabaugh, A. L. David, and T. T. Chowdhury, “Machine learning and disease prediction in obstetrics,” Curr. Res. Physiol., vol. 6, p. 100099, May 2023, doi: 10.1016/j.crphys.2023.100099.
Y. Yin and Y. Bingi, “Using Machine Learning to Classify Human Fetal Health and Analyze Feature Importance,” BioMedInformatics, vol. 3, no. 2, Art. no. 2, Jun. 2023, doi: 10.3390/biomedinformatics3020019.
T. Tiwari, T. Tiwari, and S. Tiwari, “How Artificial Intelligence, Machine Learning and Deep Learning are Radically Different?,” Int. J. Adv. Res. Comput. Sci. Softw. Eng., vol. 8, p. 1, Mar. 2018, doi: 10.23956/ijarcsse.v8i2.569.
A. Kuzu and Y. Santur, “Early Diagnosis and Classification of Fetal Health Status from a Fetal Cardiotocography Dataset Using Ensemble Learning,” Diagnostics, vol. 13, no. 15, p. 2471, Jul. 2023, doi: 10.3390/diagnostics13152471.
“Basic Tenets of Classification Algorithms K-Nearest-Neighbor, Support Vector Machine, Random Forest and Neural Network: A Review.” Accessed: Dec. 07, 2023. [Online]. Available: https://www.scirp.org/journal/paperinformation?paperid=104256
“Fig. 8. Artificial neural network (ANN) and long short-term memory...,” ResearchGate. Accessed: Dec. 07, 2023. [Online]. Available: https://www.researchgate.net/figure/Artificial-neural-network-ANN-and-long-short-term-memory-LSTM-network-training-and_fig6_351535690
G. S. Prayoga, “Logistic Regression: A Classifier With a Sense of Regression,” Medium. Accessed: Dec. 16, 2023. [Online]. Available: https://python.plainenglish.io/logistic-regression-a-classifier-with-a-sense-of-regression-83ce49ba3f5b
“Decision Tree,” GeeksforGeeks. Accessed: Dec. 24, 2023. [Online]. Available: https://www.geeksforgeeks.org/decision-tree/
D. Lazar, “Understanding Logistic Regression,” Medium. Accessed: Dec. 16, 2023. [Online]. Available: https://towardsdatascience.com/understanding-logistic-regression-81779525d5c6
M. Jain, “Support Vector Machine vs K Nearest Neighbours,” Stack Overflow. Accessed: Dec. 07, 2023. [Online]. Available: https://stackoverflow.com/q/19421954
“Bayes’ Theorem: The Idea Behind Naive Bayes Algorithm | by Soner Y?ld?r?m | Towards Data Science.” Accessed: Dec. 16, 2023. [Online]. Available: https://towardsdatascience.com/bayes-theorem-the-idea-behind-naive-bayes-algoritm-f7068834a4d7
“What Is a K-Nearest Neighbor Algorithm? | Built In.” Accessed: Dec. 24, 2023. [Online]. Available: https://builtin.com/machine-learning/nearest-neighbor-algorithm
Mila Desi Anasanti, Khairunisa Hilyati, and Annisa Novtariany, “The Exploring feature selection techniques on Classification Algorithms for Predicting Type 2 Diabetes at Early Stage,” J. RESTI Rekayasa Sist. Dan Teknol. Inf., vol. 6, no. 5, pp. 832–839, Nov. 2022, doi: 10.29207/resti.v6i5.4419.
M. Hossin and S. M.N, “A Review on Evaluation Metrics for Data Classification Evaluations,” Int. J. Data Min. Knowl. Manag. Process, vol. 5, pp. 01–11, Mar. 2015, doi: 10.5121/ijdkp.2015.5201.
“A Generic Machine Learning Approach for IoT Device Identification | IEEE Conference Publication | IEEE Xplore.” Accessed: Jan. 13, 2024. [Online]. Available: https://ieeexplore.ieee.org/document/9702983
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
________________________________________________________
The International Journal of Artificial Intelligence Research
Organized by: Departemen Teknik Informatika
Published by: STMIK Dharma Wacana
Jl. Kenanga No.03 Mulyojati 16C Metro Barat Kota Metro Lampung
Email: jurnal.ijair@gmail.com
This work is licensed under Creative Commons Attribution-ShareAlike 4.0 International License.