Logo PTI Logo FedCSIS

Position Papers of the 21st Conference on Computer Science and Intelligence Systems

Annals of Computer Science and Information Systems, Volume 48

A Breakdown of Feature Importance Ranking and Trimming (FIRT) for Maternal Health Risk Prediction Using Physiological Factors and Machine Learning

, , , , , ,

DOI: http://dx.doi.org/10.15439/2026F2543

Citation: Shilpa Mahajan, , , , , ,

Full text

Abstract. This study examines the nuts and bolts of Machine Learning (ML) when identifying maternal health concerns using previously sampled indicators. The Feature Importance Ranking and Trimming (FIRT) exploits ANOVA F-tests, Mutual Information (MI), and``Recursive Feature Elimination'' (RFE) to reduce the feature vector (FV) entries (variables) and assess model soundness. The shorter, more expressive FV improves focal predictor identification, interpretability, and training efficiency, while reducing overfitting. Key methods for measuring FV entries' importance include (a) measuring a feature's total node impurity reduction (e.g., by evaluating entropy) across all possibilities, where a higher total reduction indicates greater appeal, (b) shuffling FV entries and measuring the breakdown in model accuracy, where a suggestive drop in performance indicates a highly noteworthy feature, and (c) gauging how much accuracy must be This study predicted maternal health risks using a Kaggle dataset from several sources. This study selected and ranked features following data preprocessing. The results show that FIRT, the ANOVA F-test, and RFE performed better than MI. After model training and testing, feature selection occurred. The models used in this study include K-nearest Neighbors (KNN), Random Forest (RF), ``Gradient Boosting'' (GB),``Decision Tree'' (DT),``Support Vector Classifier'' (SVC), and``Logistic Regression.'' The RF delivered the best accuracy (87\\%) and performance. Random forests have been the most accurate and robust models for predicting maternal well-being risk. Additional feature-selection rationales for ML models can boost future usefulness and efficacy. A large dataset spanning multiple states and nations can boost forecast accuracy.

References

  1. Agrawal, S., Maitra, N.: Prediction of adverse maternal outcomes in preeclampsia using a risk prediction model. The Journal of Obstetrics and Gynecology of India 66(Suppl 1), 104–111 (2016)
  2. Assaduzzaman, M., Al Mamun, A., Hasan, M.Z.: Early prediction of maternal health risk factors using machine learning techniques. In: 2023 international conference for advancement in technology (ICONAT). pp. 1–6. IEEE (2023)
  3. Bertini, A., Salas, R., Chabert, S., Sobrevia, L., Pardo, F.: Using machine learning to predict complications in pregnancy: a systematic review. Frontiers in bioengineering and biotechnology 9, 780389 (2022)
  4. Bogale, D.S., Abuhay, T.M., Dejene, B.E.: Predicting perinatal mortality based on maternal health status and health insurance service using homogeneous ensemble machine learning methods. BMC Medical Informatics and Decision Making 22(1), 341 (2022)
  5. Kang, B.S., Lee, S.U., Hong, S., Choi, S.K., Shin, J.E., Wie, J.H., Jo, Y.S., Kim, Y.H., Kil, K., Chung, Y.H., et al.: Prediction of gestational diabetes mellitus in asian women using machine learning algorithms. Scientific Reports 13(1), 13356 (2023)
  6. Kolukisa, B., Yavuz, L., Soran, A., Bakir-Gungor, B., Tuncer, D., Onen, A., Gungor, V.C.: Coronary artery disease diagnosis using optimized adaptive ensemble machine learning algorithm. International Journal of Bioscience, Biochemistry and Bioinformatics 10(1), 58–65 (2020)
  7. Mennickent, D., Rodriguez, A., Farias-Jofre, M., Araya, J., GuzmánGutiérrez, E.: Machine learning-based models for gestational diabetes mellitus prediction before 24–28 weeks of pregnancy: a review. Artificial Intelligence in Medicine 132, 102378 (2022)
  8. Mennickent, D., Rodrı́guez, A., Opazo, M.C., Riedel, C.A., Castro, E., Eriz-Salinas, A., Appel-Rubio, J., Aguayo, C., Damiano, A.E., Guzman-Gutierrez, E., et al.: Machine learning applied in maternal and fetal health: a narrative review focused on pregnancy diseases and complications. Frontiers in Endocrinology 14, 1130139 (2023)
  9. Myatra, S.N., Gangakhedkar, G.R., Savarkar, S.: Global health and women: Looking beyond maternal health. In: Diversity, Gender Equity and Inclusion in Critical Care and Perioperative Medicine: A New Guidance for Team Excellence, pp. 33–48. Springer (2025)
  10. Organization, W.H.: Guide for conducting national and subnational programme reviews for maternal, newborn, child and adolescent health. World Health Organization (2024)
  11. Patel, S.S.: Explainable machine learning models to analyse maternal health. Data & Knowledge Engineering 146, 102198 (2023)
  12. Pawar, L., Malhotra, J., Sharma, A., Arora, D., Vaidya, D.: A robust machine learning predictive model for maternal health risk. In: 2022 3rd international conference on electronics and sustainable communication systems (ICESC). pp. 882–888. IEEE (2022)
  13. Raza, A., Siddiqui, H.U.R., Munir, K., Almutairi, M., Rustam, F., Ashraf, I.: Ensemble learning-based feature engineering to analyze maternal health during pregnancy and health risk prediction. Plos one 17(11), e0276525 (2022)
  14. Schaefer, V., Hilfiker-Kleiner, D., Keil, C.: Hypertension in pregnancy. Cardiovascular Medicine 26(3), 80–86 (2023)
  15. Singh, A., Prakash, N., Jain, A.: Particle swarm optimization-based random forest framework for the classification of chronic diseases. IEEE Access 11, 133931–133946 (2023)
  16. Sweeting, A.N., Appelblom, H., Ross, G.P., Wong, J., Kouru, H., Williams, P.F., Sairanen, M., Hyett, J.A.: First trimester prediction of gestational diabetes mellitus: a clinical model based on maternal demographic parameters. Diabetes research and clinical practice 127, 44–50 (2017)
  17. Togunwa, T.O., Babatunde, A.O., Abdullah, K.u.R.: Deep hybrid model for maternal health risk classification in pregnancy: synergy of ann and random forest. Frontiers in Artificial Intelligence 6, 1213436 (2023)
  18. Ukrit, M.F., Jeyavathana, R.B., Rani, A.L., Chandana, V.: Maternal health risk prediction with machine learning methods. In: 2024 Second International Conference on Emerging Trends in Information Technology and Engineering (ICETITE). pp. 1–9. IEEE (2024)
  19. Wei, Q., Xiao, Y., Yang, T., Chen, J., Chen, L., Wang, K., Zhang, J., Li, L., Jia, F., Wu, L., et al.: Predicting autism spectrum disorder using maternal risk factors: a multi-center machine learning study. Psychiatry Research 334, 115789 (2024)