Innovative Machine Learning Strategies for Early Detection and Prevention of Pregnancy Loss - The Vitamin D Connection and Gestational Health

Aug 2, 2026

Early pregnancy loss (EPL) is a prevalent global health concern with major consequences for gestational well-being. This study applies machine learning to improve prediction of EPL and to separate typical pregnancies from those at increased risk in the first trimester. Multiple modeling approaches are assessed, ranging from traditional classifiers to deep learning and multilayer perceptron architectures, using confusion matrices, cross-validation, and feature-importance analysis. Advanced models outperform classical methods, achieving up to 98% accuracy with linear and quadratic discriminant analyses, while maternal serum vitamin D, prior pregnancy outcomes, and age emerge as key determinants, supporting improved gestational health outcomes for mothers and infants.

LanguageEnglish

CategoryHealthcare

FormatPDF

Pages47

Reading Time1 h 58 min

AccessFree to download

Abstract

Introduction

Background

Methods and Modeling Approaches

Classical vs advanced machine learning

Evaluation and Key Determinants

Confusion matrices and feature significance

Please cite the Published Version Suﬁan, Md Abu , Hamzi, Wahiba, Hamzi, Boumediene, Sagar, ASM Sharifuzzaman , Rahman, Mustaﬁzur, Varadarajan, Jayasree , Hanumanthu, Mahesh and Azad, Md Abul Kalam (2024) Innovative machine learning strategies for early detection and prevention of pregnancy loss: the Vitamin D connection and gestational health. Diagnostics, 14 (9) . 920 ISSN 2075-4418 DOI: [https://doi.org/10.3390/diagnostics14090920](https://doi.org/10.3390/diagnostics14090920) Publisher: MDPI AG Version: Published Version Downloaded from: [https://e-space.mmu.ac.uk/634613/](https://e-space.mmu.ac.uk/634613/) Usage rights: Creative Commons: Attribution 4 .0 Additional Information: This is an open access article which originally appeared in Diagnostics, published by MDPI Enquiries: If you have questions about this document, contact [openresearch@mmu.ac.uk. Please](openresearch@mmu.ac.uk. Please) include the URL of the record in e-space. If you believe that your, or a third party's rights have been compromised through this document please see our Take Down policy (available from [https://www.mmu.ac.uk/library/using-the-library/policies-and-guidelines](https://www.mmu.ac.uk/library/using-the-library/policies-and-guidelines)) diagnostics Article Innovative Machine Learning Strategies for Early Detection and Prevention of Pregnancy Loss: The Vitamin D Connection and Gestational Health Md Abu Sufian 1,2, Wahiba Hamzi 3, Boumediene Hamzi 4,5,6, A. S. M. Sharifuzzaman Sagar 7, Mustafizur Rahman 8, Jayasree Varadarajan 9, Mahesh Hanumanthu 2 and Md Abul Kalam Azad 10, * Citation: Sufian, M.A.; Hamzi, W.; Hamzi, B.; Sagar, A.S.M.S.; Rahman, M.; Varadarajan, J.; Hanumanthu, M.; Azad, M.A.K. Innovative Machine Learning Strategies for Early Detection and Prevention of Pregnancy Loss: The Vitamin D Connection and Gestational Health. Diagnostics 2024, 14, 920. [https://](https://)[ ](https://)[doi.org/10.3390/diagnostics14090920](doi.org/10.3390/diagnostics14090920) Academic Editors: Kwang-Sig Lee and Ki Hoon Ahn Received: 22 February 2024 Revised: 30 March 2024 Accepted: 23 April 2024 Published: 28 April 2024 Copyright: © 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license ([https://](https://)[ ](https://)[creativecommons.org/licenses/by/](creativecommons.org/licenses/by/)[ ](creativecommons.org/licenses/by/)[4.0/](4.0/)) . 1 IVR Low-Carbon Research Institute, Chang’an University, Xi’an 710018, China; [md.sufian@mail.bcu.ac.uk](md.sufian@mail.bcu.ac.uk) 2 School of Computing and Mathematical Sciences, University of Leicester, Leicester LE1 7RH, UK 3 Laboratoire de Biotechnologie, Environnement et Santé, Department of Biology, University of Blida, Blida 09000, Algeria 4 Department of Computing and Mathematical Sciences, California Institute of Technology, Caltech, CA 91125, USA 5 The Alan Turing Institute, London NW1 2DB, UK 6 Department of Mathematics, Gulf University for Science and Technology (GUST), Mubarak Al-Abdullah 32093, Kuwait 7 Department of AI and Robotics, Sejong University, Seoul 05006, Republic of Korea 8 Department of Industrial Engineering, Tsinghua University, Beijing 100084, China 9 Centre for Digital Innovation, Manchester Metropolitan University, Manchester M15 6BH, UK 10 Department of Medicine, Rangpur Medical College and Hospital, Rangpur 5400, Bangladesh * Correspondence: [drazad.2019@gmail.com](drazad.2019@gmail.com) Abstract: Early pregnancy loss (EPL) is a prevalent health concern with significant implications globally for gestational health. This research leverages machine learning to enhance the prediction of EPL and to differentiate between typical pregnancies and those at elevated risk during the initial trimester. We employed different machine learning methodologies, from conventional models to more advanced

FAQs

What problem does the research address?

The research targets early pregnancy loss (EPL) and aims to predict which pregnancies are at elevated risk during the initial trimester.

Which machine learning methods are compared?

The study compares conventional models with advanced approaches, including deep learning and multilayer perceptron models.

What factors are identified as key determinants of EPL?

Maternal serum vitamin D levels are highlighted, alongside prior pregnancy outcomes and maternal age as important contributors.

Aug 2, 2026

Upload by Ava Thompson