A Machine Learning Framework for Anticipating Ferritin Deficiency Utilizing Clinical Biomarkers and Gradient Boosting Models
DOI:
https://doi.org/10.29304/jqcsm.2026.18.22698Keywords:
Machine Learning, XGBoost, LightGBM, CatBoostAbstract
Ferritin deficiency is a public health problem that causes. It means people are iron deficient and that can have serious health consequences. We need to catch people at risk early and make sure that these problems do not happen.
This study takes these insights and uses a machine learning framework to predict whether an individual will have ferritin levels. We use data including who people are, what their blood contains and their diets. We extracted this information from NHANES data.
We ran three different machine learning systems. XGBoost, LightGBM and CatBoost. A sample of 1318 people were tested. The main problem was that there were no subjects with ferritin deficiency in the group. We used person measures such as area under the curve and accuracy to take a look on how the systems worked. CatBoost did the best of the bunch. It was good at detecting whom had deficiency and whom not. The area under curve of CatBoost system was 0.946 which’s very good. It also scored with a recall of 64.77% and an F1-score of 0.60 So that it was not bad at identifying deficiency. CatBoost system reduced the number of negatives by 38% compared to basic system. LightGBM system did well too. The Catboost system performed better than the others in identifying individuals with ferritin deficiency.
The results indicate that machine learning systems can help us identification of the people who are at risk of health problems. To make sense of results, we also need to view them in the context of something that some health problemsre not common
Downloads
References
Efros, O., Soffer, S., Mudrik, A., Robinson, R., Kenet, G., Nadkarni, G.N., and Klang, E.: ‘Predictive machine-learning model for screening iron deficiency without anaemia: a retrospective cohort study’, BMJ Open, 2025, 15, (8), pp. e097016
Kurstjens, S., de Bel, T., van der Horst, A., Kusters, R., Krabbe, J., and van Balveren, J.: ‘Automated prediction of low ferritin concentrations using a machine learning algorithm’, Clin Chem Lab Med, 2022, 60, (12), pp. 1921-1928
Rojanaphan, P.: ‘Automated Nutrient Deficiency Detection and Recommendation Systems Using Deep Learning in Nutrition Science’, International Journal of Scientific Research and Management (IJSRM), 2024, 12, (11), pp. 1746-1763
Das, A., Bai, C.H., Chang, J.S., Huang, Y.L., Wang, F.F., Hsu, C.Y., Chen, Y.C., and Chao, J.C.: ‘A ferritin-related dietary pattern is positively associated with iron status but negatively associated with vitamin D status in pregnant women: a cross-sectional study’, Eur J Nutr, 2024, 64, (1), pp. 30
Revesai, Z., and Kogeda, O.P.: ‘Knowledge-Guided Neural Networks for Transparent Micronutrient Deficiency Detection’, in Editor (Ed.)^(Eds.): ‘Book Knowledge-Guided Neural Networks for Transparent Micronutrient Deficiency Detection’ (2025, edn.), pp. 1-6
A. Ali Heydari, Naghmeh Rezaei, Javier L. Prieto†, Shwetak N. Patel, Ahmed A. Metwally: ‘Lifestyle-Informed Personalized Blood Biomarker Prediction via Novel Representation Learning ’, 2024, (arXiv:2407.07277v1 [cs.LG] 9 Jul 2024)
Sawicki, C., Haslam, D., and Bhupathiraju, S.: ‘Utilising the precision nutrition toolkit in the path towards precision medicine’, Proc Nutr Soc, 2023, 82, (3), pp. 359-369
Obermeyer, Z., and Emanuel, E.J.: ‘Predicting the Future — Big Data, Machine Learning, and Clinical Medicine’, N Engl J Med, 2016, 375, (13), pp. 1216–1219
] Rajkomar, A., Dean, J., and Kohane, I.: ‘Machine Learning in Medicine’, N Engl J Med, 2019, 380, (14), pp. 1347–135810
Guyon, I., and Elisseeff, A.: ‘An Introduction to Variable and Feature Selection’, J Mach Learn Res, 2003, 3, pp. 1157–1182
X. Yao, T. Li, and J. J. Liu, “Application of Machine Learning Model in Fraud Identification: A Comparative Study of CatBoost, XGBoost and LightGBM,” Mar. 2025, doi: 10.20944/preprints202503.1199.v1.
M. R. Johnson, E. S. A. Sanders, D. Miller, S. Ramírez, and D. R. Adams, “Comparative Study of Efficient Machine Learning Models for Real-Time Fraud Detection: CatBoost, XGBoost and LightGBM,” Sep. 2025, doi: 10.21203/rs.3.rs-7539803/v1.
Bentéjac, C., Csörgő, A., & Martínez-Muñoz, G. (2021). A comparative analysis of gradient boosting algorithms. Artificial Intelligence Review, 54(3), 1937–1967. https://doi.org/10.1007/S10462-020-09896-5
Д. С. Пономарев, “Сопоставление библиотек для создания моделей машинного обучения на основе методов градиентного бустинга,” Современные инновации, системы и технологии, vol. 5, no. 2, pp. 3001–3006, Apr. 2025, doi: 10.47813/2782-2818-2025-5-2-3001-3006.
R. D. Cançado, L. A. C. Leite, and M. Múñoz, “Defining Global Thresholds for Serum Ferritin: A Challenging Mission in Establishing the Iron Deficiency Diagnosis in This Era of Striving for Health Equity,” Diagnostics, vol. 15, no. 3, p. 289, Jan. 2025, doi: 10.3390/diagnostics15030289.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Zahraa Kadhim AlSendi
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
