Artificial Intelligence and Predictive Medicine: Predicting Thrombocytopenia in Patients with Atrial Fibrillation

(Amadou Diabagate, Katienefowa Sekou Koulibaly, Awa FOFANA and Doffou Jérôme Diako)

---

Abstract

This study aimed to develop and evaluate machine learning models capable of predicting thrombocytopenia in patients with atrial fibrillation, with a focus on African clinical settings where this complication is underexplored despite its clinical importance. Real-world data were obtained from the AFRICA registry, a large multicenter database encompassing diverse patient profiles. Six algorithms were implemented and compared, including Decision Tree, Random Forest, XGBoost, Support Vector Machine, K-Nearest Neighbors, and Multi-Layer Perceptron. Stratified cross-validation was used to ensure robust evaluation based on accuracy, F1-score, AUC-ROC, Log Loss, and Matthews Correlation Coefficient. Model interpretability was enhanced using the SHAP method to identify the most influential predictors. Tree-based models performed best. On cross-validation, XGBoost reached 97.4%, F1 0.92, and AUC-ROC 0.98, and its performance was confirmed on an independent 20% holdout set (accuracy 93.75%, F1 0.857, AUC-ROC 0.9803). SHAP analysis highlighted platelet count, hemoglobin, creatinine, and glycemia as the strongest predictors, alongside clinical factors such as amiodarone therapy, intensive care admission, and depressive symptoms. High-performance, interpretable machine learning models can accurately forecast thrombocytopenia in African patients with atrial fibrillation. These findings provide a solid basis for the development of clinical decision support systems aimed at improving patient management and treatment outcomes.
KEYWORDS
AFRICA registry, clinical decision, hematology, model interpretability, multicenter cohort, supervised learning

PDF

References

Bishop, C.M. (1995). Neural Networks for Pattern Recognition. Oxford University Press.
Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.  DOI: 10.1023/A:1010933404324
Breiman, L., Friedman, J., Olshen, R.A. and Stone, C.J. (2017). Classification and Regression Trees. Chapman and Hall/CRC. DOI: 10.1201/9781315139470
Chen, T. and Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794). DOI: 10.1145/2939672.2939785
Cheng, Y., Chen, C., Yang, J., Yang, H., Fu, M., Zhong, X., Wang, B., He, M., Hu, Z., Zhang, Z., Jin, X., Kang, Y. and Wu, Q. (2021). Using machine learning algorithms to predict hospital-acquired thrombocytopenia after operation in the intensive care unit: A retrospective cohort study. Diagnostics, 11(9), 1614. DOI: 10.3390/diagnostics11091614
Chugh, S.S., Havmoeller, R., Narayanan, K., Singh, D., Rienstra, M., Benjamin, E.J., Gillum, R.F., Kim, Y.-H., McAnulty, J.H., Zheng, Z.-J., Forouzanfar, M.H., Naghavi, M., Mensah, G.A., Ezzati, M. and Murray, C.J.L. (2014). Worldwide epidemiology of atrial fibrillation: A global burden of disease 2010 study. Circulation, 129(8), 837–47. DOI: 10.1161/CIRCULATIONAHA.113.005119
Cortes, C. and Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 73–297. DOI: 10.1007/BF00994018
Ding, L. and Tang, M. (2025). Global, regional and national burden of atrial fibrillation and flutter attributable to metabolic risks from 1990 to 2021: Analysis of data from the global burden of disease study. EP Europace, 27(1), euaf085.269. DOI: 10.1093/europace/euaf085.269
Diop, K.R., Samb, C.A.B., Kane, A., Mingou, J.S., Beye, S.M., Diouf, Y. and Adoubi, A. K. (2022). Atrial fibrillation in three cardiological reference centers in Dakar: Senegal data from the AFRICA register survey. The Pan African Medical Journal, 43(n/a), 112. DOI: 10.11604/pamj.2022.43.112.31397
Ettarh, R. (2016). Patterns of international collaboration in cardiovascular research in sub-saharan Africa. Cardiovascular Journal of Africa, 27(3), 436–45. DOI: 10.5830/CVJA-2015-082
Ghanima, W. and Cooper, N. (2024). Could machine learning revolutionize how we treat immune thrombocytopenia? British Journal of Haematology, 205(3), 770–1. DOI: 10.1111/bjh.19684
Haibo, H. and Garcia, E.A. (2009). Learning from imbalanced data. IEEE Transactions on Knowledge and Data Engineering, 21(9), 1263–84. DOI: 10.1109/TKDE.2008.239
Haroon, M., Dimitrios, K., Giacinto, L.P., Drew, P. and Frederick, C. (2024). Can machine learning assist in diagnosis of primary immune thrombocytopenia? A feasibility study. Diagnostics, 14(13), 1352. DOI: 10.3390/diagnostics14131352
Hindricks, G., Potpara, T., Dagres, N., Arbelo, E., Bax, J.J., Blomström-Lundqvist, C., Boriani, G., Castella, M., Dan, G.A., Dilaveris, P.E., Fauchier, L., Filippatos, G., Kalman, J.M., La Meir, M., Lane, D.A., Lebeau, J.P., Lettino, M., Lip, G.Y.H., Pinto, F.J., Thomas, G.N., Valgimigli, M., Van Gelder, I.C., Van Putte, B.P. and Watkins, C.L. (2021). 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation. European Heart Journal, 42(5), 373–498. DOI: 10.1093/eurheartj/ehaa612
Iijima, R., Tokue, M., Nakamura, M., Yasuda, S., Kaikita, K., Akao, M., Ako, J., Matoba, T., Miyauchi, K., Hagiwara, N., Kimura, K., Hirayama, A., Matsui, K., Ogawa, H. and AFIRE Investigators. (2023). Thrombocytopenia as a bleeding risk factor in atrial fibrillation and coronary artery disease: Insights from the AFIRE study. Journal of the American Heart Association, 12(20), e031096. DOI: 10.1161/JAHA.123.031096
Iyengar, V., Patell, R., Ren, S., Ma, S., Pinson, A., Barnett, A., Elavalakanar, P., Kazi, D.S., Neuberg, D. and Zwicker, J.I. (2023). Influence of thrombocytopenia on bleeding and vascular events in atrial fibrillation. Blood Advances, 7(24), 7516–24. DOI: 10.1182/bloodadvances.2023011235
James, G., Witten, D., Hastie, T. and Tibshirani, R. (2013). An Introduction to Statistical Learning: With Applications in R (Vol. 103). New York: Springer. DOI: 10.1007/978-1-4614-7138-7
January, C.T., Wann, L.S., Calkins, H., Chen, L.Y., Cigarroa, J.E., Cleveland, J.C., Ellinor, P.T., Ezekowitz, M.D., Field, M.E., Furie, K.L., … (2019). AHA/ACC/HRS focused update on atrial fibrillation: Management guidelines. Circulation, 140(2), e125–e151. DOI: 10.1161/CIR.0000000000000665
Jiang, X., Wang, Y., Pan, Y. and Zhang, W. (2022). Prediction models for sepsis-associated thrombocytopenia risk in intensive care units based on a machine learning algorithm. Frontiers in Medicine, 9(n/a), 837382. DOI: 10.3389/fmed.2022.837382
Lip, G., Gue, Y., Zhang, J., Chao, T., Calkins, H. and Potpara, T. (2022). Stroke prevention in atrial fibrillation. Trends in Cardiovascular Medicine, 32(8), 501–10. DOI: 10.1016/j.tcm.2021.10.001
Liu, T., Krentz, A., Lu, L. and Curcin, V. (2025). Machine learning based prediction models for cardiovascular disease risk using electronic health records data: Systematic review and meta-analysis. European Heart Journal-Digital Health, 6(1), 7–22. DOI: 10.1093/ehjdh/ztae080
Lu, J., Bisson, A., Bennamoun, M., Zheng, Y., Sanfilippo, F.M., Hung, J., Briffa, T., McQuillan, B., Stewart, J., Figtree, G., Huisman, M.V., Dwivedi, G. and Lip, G.Y. (2024). Predicting multifaceted risks using machine learning in atrial fibrillation: Insights from GLORIA-AF study. European Heart Journal – Digital Health, 5(3), 235–46. DOI: 10.1093/ehjdh/ztae010
Lundberg, S.M. and Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30(n/a), 4765–74. DOI: 10.48550/arXiv.1705.07874
Mendis, S., Puska, P. and Norrving, B. (2011). Global Atlas on Cardiovascular Disease Prevention and Control. World Health Organization.
Miotto, R., Wang, F., Wang, S., Jiang, X. and Dudley, J.T. (2018). Deep learning for healthcare: Review, opportunities and challenges. Briefings in Bioinformatics, 19(6), 1236–46. DOI: 10.1093/bib/bbx044
Moulis, G., Comont, T., Germain, J., Sommet, A., Lapeyre-Mestre, M., Beyne-Rauzy, O. and Adoue, D. (2020). Significance of antinuclear antibodies in primary immune thrombocytopenia: Results of the CARMEN registry. Blood Advances, 4(9), 1974–7. DOI: 10.1182/bloodadvances.2020001664
Pisters, R., Lane, D.A., Nieuwlaat, R., de Vos, C.B., Crijns, H.J.G.M. and Lip, G.Y.H. (2010). The HAS-BLED score: A novel tool to assess bleeding risk in patients with atrial fibrillation. Chest, 138(5), 1093–100. DOI: 10.1378/chest.10-0134
Rahul, C., Mehdi, N., Floyd, W.T., Walid, F.G., Wei-Hsuan, L.-C., Rohit, C., Anahita, D., Kevin, P.B., Paul, A.G., Matthew, D.N., Sandeep, J., Aditya, B., Suresh, R.M., Yanshan, W., Matthew, E.H. and Samir, S. (2025). Machine learning predicts bleeding risk in atrial fibrillation patients on direct oral anticoagulant. The American Journal of Cardiology, 244(n/a), 58–66. DOI: 10.1016/j.amjcard.2025.02.030

Rajkomar, A., Dean, J. and Kohane, I. (2019). Machine learning in medicine. The New England Journal of Medicine, 380(14), 1347–58. DOI: 10.1056/NEJMra1814259
Stambler, B.S. and Ngunga, L.M. (2015). Atrial fibrillation in sub-Saharan Africa: Epidemiology, unmet needs and treatment options. International Journal of General Medicine, 8(n/a), 231–42. DOI: 10.2147/IJGM.S84537
Stasi, R. (2012). Immune thrombocytopenia: Pathophysiologic and clinical update. Seminars in Thrombosis and Hemostasis, 38(5), 454–62. DOI: 10.1055/s-0032-1305780
Verheugt, F.W. and Granger, C.B. (2015). Oral anticoagulants for stroke prevention in atrial fibrillation: Current status, special situations and unmet needs. The Lancet, 386(9990), 303–10. DOI: 10.1016/S0140-6736(15)60245-8
Yeh, Y.H., Chan, Y.H., Chen, S.W., Chang, S.H., Wang, C.L., Kuo, C.T. and Chao, T.F. (2022). Oral anticoagulant use for patients with atrial fibrillation with concomitant anemia and/or thrombocytopenia. The American Journal of Medicine, 135(8), e248-56. DOI: 10.1016/j.amjmed.2022.03.011