Annals of Laboratory Medicine

Korean Society for Laboratory Medicine (대한진단검사의학회)
권호 표지
DOI QR코드

DOI QR Code

A Machine Learning Approach to Reference Interval Estimation for Red Cell Parameters in a South and East Asian Population

  • Received : 2025.01.13
  • Accepted : 2025.05.27
  • Published : 2026.01.01
Download PDF
⟨ Previous Next ⟩

Abstract

Background: Iron deficiency (ID) and hemoglobinopathies are highly prevalent in Southeast Asia. Accurate estimation of reference intervals (RIs) for red cell parameters is complicated by the need to exclude individuals with these conditions from the reference population. Indirect RI estimations using machine learning could help overcome these challenges. Methods: We developed a binary classification model using eXtreme Gradient Boosting (XGB) to distinguish normal individuals from those with ID, hemoglobinopathies, or other anemias. The model was trained on an annotated dataset comprising 5,520 complete blood count (CBC) results and validated with a holdout dataset of 2,367 CBC results. An independent dataset of 64,100 CBC results was used to identify individuals predicted to be normal, from which RIs were estimated using the refineR algorithm. Results: The XGB model achieved an area under the ROC of 0.97 (95% confidence interval: 0.96–0.97) for distinguishing between individuals with normal versus abnormal values. Among individuals within the independent dataset, 40,300 (62.9%) were predicted to be normal. The refineR-based reference limits (RLs) derived from this subset approximated those obtained through a direct approach. Improvements in the accuracy of indirect RL estimates were most evident for hematocrit, hemoglobin, and red cell concentrations. Conclusions: Combining XGB with refineR to indirectly derive RIs for red cell parameters improved the accuracy and yielded results comparable with those of directly derived RIs. A further benefit was the capacity to generate sex- and age-specific ranges, which has remained difficult to achieve through direct approaches.

Keywords

Acknowledgement

The authors extend their gratitude to Dr. Tong Dong Ling from the Department of Computer Science, Faculty of Information & Communication Technology, Universiti Tunku Abdul Rahman for her help in initiating the study and to the staff of the Laboratory Medicine Division, University Malaya Medical Centre for their contributions to the study.

References

  1. Shen JZ, Hill BC, Polhill SR, Evans P, Galloway DP, Johnson RB, et al. Optimization of laboratory ordering practices for complete blood count with differential. Am J Clin Pathol 2019;151:306–15. https://doi.org/10.1093/ajcp/aqy146
  2. Rivera AKB, Latorre AAE, Nakamura K, Seino K. Using complete blood count parameters in the diagnosis of iron deficiency and iron deficiency anemia in Filipino women. J Rural Med 2023;18:79–86. https://doi.org/10.2185/jrm.2022-047
  3. Buttarello M. Laboratory diagnosis of anemia: are the old and new red cell parameters useful in classification and treatment, how? Int J Lab Hematol 2016;38(S1):123–32. https://doi.org/10.1111/ijlh.2016.38.issue-S1
  4. Winichagoon P, Kumbunlue R, Sirankapracha P, Boonmongkol P, Fucharoen S. Discrimination of various thalassemia syndromes and iron deficiency and utilization of reticulocyte measurements in monitoring response to iron therapy. Blood Cells Mol Dis 2015;54:336–41. https://doi.org/10.1016/j.bcmd.2015.01.010
  5. Feng P, Li Y, Liao Z, Yao Z, Lin W, Xie S, et al. An online alpha-thalassemia carrier discrimination model based on random forest and red blood cell parameters for low HbA2 cases. Clin Chim Acta 2022;525:1–5. https://doi.org/10.1016/j.cca.2021.12.003
  6. Tuo Y, Li Y, Li Y, Ma J, Yang X, Wu S, et al. Global, regional, and national burden of thalassemia, 1990–2021: a systematic analysis for the global burden of disease study 2021. EClinicalMedicine 2024;72:102619.
  7. Kattamis A, Forni GL, Aydinok Y, Viprakasit V. Changing patterns in the epidemiology of β-thalassemia. Eur J Haematol 2020;105:692–703. https://doi.org/10.1111/ejh.v105.6
  8. Alwi ZB and Syed-Hassan SR. Thalassemia in Malaysia. Hemoglobin 2022;46:45–52. https://doi.org/10.1080/03630269.2022.2057326
  9. Safiri S, Kolahi AA, Noori M, Nejadghaderi SA, Karamzad N, Bragazzi NL, et al. Burden of anemia and its underlying causes in 204 countries and territories, 1990-2019: results from the Global Burden of Disease Study 2019. J Hematol Oncol 2021;14:185. https://doi.org/10.1186/s13045-021-01202-2
  10. Ambayya A, Su AT, Osman NH, Nik-Samsudin NR, Khalid K, Chang KM, et al. Haematological reference intervals in a multiethnic population. PLoS One 2014;9:e91968. https://doi.org/10.1371/journal.pone.0091968
  11. Ichihara K, Ozarda Y, Barth JH, Klee G, Qiu L, Erasmus R, et al. A global multicenter study on reference values: 1. Assessment of methods for derivation and comparison of reference intervals. Clin Chim Acta 2017;467:70–82. https://doi.org/10.1016/j.cca.2016.09.016
  12. Ozarda Y, Sikaris K, Streichert T, Macri J, IFCC Committee on Reference intervals and Decision Limits (C-RIDL). Distinguishing reference intervals and clinical decision limits – a review by the IFCC Committee on Reference Intervals and Decision Limits. Crit Rev Clin Lab Sci 2018;55:420–31. https://doi.org/10.1080/10408363.2018.1482256
  13. Gao J and Liu W. Advances in screening of thalassaemia. Clin Chim Acta 2022;534:176–84. https://doi.org/10.1016/j.cca.2022.08.001
  14. Nah EH, Kim S, Cho S, Cho HI. Complete blood count reference intervals and patterns of changes across pediatric, adult, and geriatric ages in Korea. Ann Lab Med 2018;38:503–11. https://doi.org/10.3343/alm.2018.38.6.503
  15. Haeckel R, Adeli K, Jones G, Sikaris K, Wosniok W. Definitions and major prerequisites of direct and indirect approaches for estimating reference limits. Clin Chem Lab Med 2023;61:402–6. https://doi.org/10.1515/cclm-2022-1061
  16. Rusch JA, van der Westhuizen DJ, Gill RS, Louw VJ. Diagnosing iron deficiency: controversies and novel metrics. Best Pract Res Clin Anaesthesiol 2023;37:451–67. https://doi.org/10.1016/j.bpa.2023.11.001
  17. Lim E, Miyamura J, Chen JJ. Racial/ethnic-specific reference intervals for common laboratory tests: a comparison among Asians, Blacks, Hispanics, and White. Hawaii J Med Public Health 2015;74:302–10.
  18. Meyer A, Müller R, Hoffmann M, Skadberg Ø, Ladang A, Dieplinger B, et al. Comparison of three indirect methods for verification and validation of reference intervals at eight medical laboratories: a European multicenter study. J Lab Med 2023;47:155–63. https://doi.org/10.1515/labmed-2023-0042
  19. Ammer T, Schützenmeister A, Prokosch HU, Rauh M, Rank CM, Zierk J. refineR: a novel algorithm for reference interval estimation from realworld data. Sci Rep 2021;11:16023. https://doi.org/10.1038/s41598-021-95301-2
  20. Chen T and Guestrin C. XGBoost: a scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Association for Computing Machinery, San Francisco, 2016; 785–94.
  21. Omuse G, Maina D, Mwangi J, Wambua C, Radia K, Kanyua A, et al. Complete blood count reference intervals from a healthy adult urban population in Kenya. PLoS One 2018;13:e0198444. https://doi.org/10.1371/journal.pone.0198444
  22. Al-Mawali A, Pinto AD, Al-Busaidi R, Al-Lawati RH, Morsi M. Comprehensive haematological indices reference intervals for a healthy Omani population: first comprehensive study in Gulf Cooperation Council (GCC) and Middle Eastern countries based on age, gender and ABO blood group comparison. PLoS One 2018;13:e0194497. https://doi.org/10.1371/journal.pone.0194497
  23. Phillips R, Wood H, Weaving G, Chevassut T. Changes in full blood count parameters with age and sex: results of a survey of almost 900 000 patient samples from primary care. Br J Haematol 2021;192:e102–5. https://doi.org/10.1111/bjh.v192.4
  24. Schipper A, Rutten M, van Gammeren A, Harteveld CL, Urrechaga E, Weerkamp F, et al. Machine learning-based prediction of hemoglobinopathies using complete blood count data. Clin Chem 2024;70:1064–75. https://doi.org/10.1093/clinchem/hvae081
  25. Lee JY, Choi H, Park JW, Son BR, Park JH, Jang LC, et al. Age-related changes in mean corpuscular volumes in patients without anaemia: an analysis of large-volume data from a single institute. J Cell Mol Med 2022;26:3548–56. https://doi.org/10.1111/jcmm.v26.12
  26. Biino G, Santimone I, Minelli C, Sorice R, Frongia B, Traglia M, et al. Ageand sex-related variations in platelet count in Italy: a proposal of reference ranges based on 40987 subjects' data. PLoS One 2013;8:e54289. https://doi.org/10.1371/journal.pone.0054289
  27. Balduini CL and Noris P. Platelet count and aging. Haematologica 2014;99:953–5. https://doi.org/10.3324/haematol.2014.106260
  28. Coşkun A, Carobene A, Kilercik M, Serteser M, Sandberg S, Aarsand AK, et al. Within-subject and between-subject biological variation estimates of 21 hematological parameters in 30 healthy subjects. Clin Chem Lab Med 2018;56:1309–18. https://doi.org/10.1515/cclm-2017-1155