{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T22:22:55Z","timestamp":1790374975441,"version":"4.1.0"},"reference-count":45,"publisher":"MDPI AG","issue":"10","license":[{"start":{"date-parts":[[2025,10,14]],"date-time":"2025-10-14T00:00:00Z","timestamp":1760400000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100006109","name":"VEGA","doi-asserted-by":"publisher","award":["1\/0700\/23"],"award-info":[{"award-number":["1\/0700\/23"]}],"id":[{"id":"10.13039\/501100006109","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Pharmaceuticals"],"abstract":"<jats:p>Background: Despite the availability of direct oral anticoagulants, warfarin remains essential for mechanical valves, renal impairment, and resource-limited settings. Traditional dosing achieves therapeutic range in only 55\u201365% of patients, increasing bleeding and thrombotic complications. This systematic review evaluates the literature on machine learning (ML) approaches for warfarin dose prediction (2022\u20132025). Methods: We analysed 14 studies encompassing 122,400 patients across nine countries following PRISMA guidelines. Studies utilizing ML algorithms for warfarin dosing with quantifiable performance metrics were included. Risk of bias was assessed using PROBAST. Results: Reinforcement learning demonstrated superior performance, achieving an 80.8% excellent responder ratio versus 41.6% for standard practice and 99.5% safety responder ratio versus 83.1%. Support vector machines achieved R2 up to 0.98 in homogeneous populations. Mean absolute error ranged from 0.11 to 1.8 mg\/day, consistently outperforming traditional methods. Seven studies included external validation, whilst 78.6% were retrospective designs. Limited implementation studies showed therapeutic INR rates improving from 47.5% to 61.1%. Critically, only three studies (21.4%) reported any safety outcomes, with none adequately powered to detect differences in major bleeding events. Conclusions: While ML algorithms demonstrate improved dosing accuracy in retrospective analyses, the near-complete absence of adequately powered safety outcome data represents the primary barrier to clinical implementation. Without robust evidence on bleeding, thromboembolism, and mortality, the risk\u2013benefit profile remains unknown. Implementation requires addressing: the predominance of retrospective studies (78.6%), limited prospective validation, restricted geographic diversity (43% from China), absence of African and South American studies, and no new Hispanic population data. Multicentre prospective trials with safety endpoints, population-specific validation, and interpretable models are essential before widespread clinical adoption can be recommended.<\/jats:p>","DOI":"10.3390\/ph18101544","type":"journal-article","created":{"date-parts":[[2025,10,14]],"date-time":"2025-10-14T13:41:29Z","timestamp":1760449289000},"page":"1544","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Machine Learning for Warfarin Therapy: A Systematic Review"],"prefix":"10.3390","volume":"18","author":[{"given":"Pavol","family":"F\u00fcl\u00f6p","sequence":"first","affiliation":[{"name":"2nd Department of Cardiology, East Slovak Institute of Cardiovascular Diseases, Faculty of Medicine, Pavol Jozef \u0160af\u00e1rik University in Ko\u0161ice, Ondavsk\u00e1 8, 040 11 Ko\u0161ice, Slovakia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"\u0160tefan","family":"T\u00f3th","sequence":"additional","affiliation":[{"name":"Department of Gerontology and Geriatrics, University Hospital of St. Michael, Faculty of Medicine, Pavol Jozef \u0160af\u00e1rik University in Ko\u0161ice, Murga\u0161ova 1, 040 86 Ko\u0161ice, Slovakia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6946-3765","authenticated-orcid":false,"given":"Tibor","family":"Porub\u00e4n","sequence":"additional","affiliation":[{"name":"1st Department of Cardiology, East Slovak Institute of Cardiovascular Diseases, Faculty of Medicine, Pavol Jozef \u0160af\u00e1rik University in Ko\u0161ice, Ondavsk\u00e1 8, 040 11 Ko\u0161ice, Slovakia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zuzana","family":"F\u00fcl\u00f6pov\u00e1","sequence":"additional","affiliation":[{"name":"First Department of Internal Medicine, Louis Pasteur University Hospital, Faculty of Medicine, Pavol Jozef \u0160af\u00e1rik University in Ko\u0161ice, Trieda SNP 1, 040 11 Ko\u0161ice, Slovakia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Anna","family":"Borovsk\u00e1","sequence":"additional","affiliation":[{"name":"2nd Department of Cardiology, East Slovak Institute of Cardiovascular Diseases, Faculty of Medicine, Pavol Jozef \u0160af\u00e1rik University in Ko\u0161ice, Ondavsk\u00e1 8, 040 11 Ko\u0161ice, Slovakia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mariana","family":"Dvoro\u017e\u0148\u00e1kov\u00e1","sequence":"additional","affiliation":[{"name":"2nd Department of Cardiology, East Slovak Institute of Cardiovascular Diseases, Faculty of Medicine, Pavol Jozef \u0160af\u00e1rik University in Ko\u0161ice, Ondavsk\u00e1 8, 040 11 Ko\u0161ice, Slovakia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2025,10,14]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"823","DOI":"10.1053\/j.ajkd.2010.05.023","article-title":"Warfarin dosing in patients with impaired kidney function","volume":"56","author":"Limdi","year":"2010","journal-title":"Am. J. Kidney Dis."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"m3956","DOI":"10.1136\/bmj.m3956","article-title":"Warfarin in patients with mechanical heart valves","volume":"371","author":"Catterall","year":"2020","journal-title":"BMJ"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1002\/psp4.12740","article-title":"Stable warfarin dose prediction in sub-Saharan African patients: A machine-learning approach and external validation of a clinical dose-initiation algorithm","volume":"11","author":"Asiimwe","year":"2022","journal-title":"CPT Pharmacomet. Syst. Pharmacol."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"753","DOI":"10.1056\/NEJMoa0809329","article-title":"Estimation of the warfarin dose with clinical and pharmacogenetic data","volume":"360","author":"Klein","year":"2009","journal-title":"N. Engl. J. Med."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Takeuchi, F., McGinnis, R., Bourgeois, S., Barnes, C., Eriksson, N., Soranzo, N., Whittaker, P., Ranganath, V., Kumanduri, V., and McLaren, W. (2009). A genome-wide association study confirms VKORC1, CYP2C9, and CYP4F2 as principal genetic determinants of warfarin dose. PLoS Genet., 5.","DOI":"10.1371\/journal.pgen.1000433"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"226","DOI":"10.1097\/FPC.0b013e328326e0c7","article-title":"Validation of VKORC1 and CYP2C9 genotypes on interindividual warfarin maintenance dose: A prospective study in Chinese patients","volume":"19","author":"Huang","year":"2009","journal-title":"Pharmacogenet. Genom."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"781","DOI":"10.2217\/pgs.10.49","article-title":"Combined CYP2C9, VKORC1 and CYP4F2 frequencies among racial and ethnic groups","volume":"11","author":"Scott","year":"2010","journal-title":"Pharmacogenomics"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"563","DOI":"10.2217\/fca.12.31","article-title":"The future of warfarin pharmacogenetics in under-represented minority groups","volume":"8","author":"Cavallari","year":"2012","journal-title":"Future Cardiol."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1008","DOI":"10.1345\/aph.1E566","article-title":"The influence of ethnicity on warfarin dosage requirement","volume":"39","author":"Dang","year":"2005","journal-title":"Ann Pharmacother."},{"key":"ref_10","first-page":"876","article-title":"Patients\u2019 time in therapeutic range on warfarin among US patients with atrial fibrillation: Results from ORBIT-AF registry","volume":"169","author":"Pokorney","year":"2015","journal-title":"Am. Heart J."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"89","DOI":"10.2147\/JBM.S41404","article-title":"Current strategies to minimize the bleeding risk of warfarin","volume":"4","author":"Snipelisky","year":"2013","journal-title":"J. Blood Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"110","DOI":"10.1016\/j.thromres.2023.03.010","article-title":"Bleeding related to oral anticoagulants: Trends in US emergency department visits, 2016\u20132020","volume":"225","author":"Geller","year":"2023","journal-title":"Thromb. Res."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Zhang, F., Liu, Y., Ma, W., Zhao, S., Chen, J., and Gu, Z. (2022). Nonlinear Machine Learning in Warfarin Dose Prediction: Insights from Contemporary Modelling Studies. J. Pers. Med., 12.","DOI":"10.3390\/jpm12050717"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e44S","DOI":"10.1378\/chest.11-2292","article-title":"Oral anticoagulant therapy: Antithrombotic therapy and prevention of thrombosis","volume":"141","author":"Ageno","year":"2012","journal-title":"Chest"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2563","DOI":"10.1161\/CIRCULATIONAHA.107.737312","article-title":"Randomized trial of genotype-guided versus standard warfarin dosing in patients initiating oral anticoagulation","volume":"116","author":"Anderson","year":"2007","journal-title":"Circulation"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Liu, R., Li, X., Zhang, W., and Zhou, H.H. (2015). Comparison of nine statistical model based warfarin pharmacogenetic dosing algorithms using the racially diverse international warfarin pharmacogenetic consortium cohort database. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0135784"},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Ma, Z., Wang, P., Gao, Z., Wang, R., and Khalighi, K. (2018). Ensemble of machine learning algorithms using the stacked generalization approach to estimate the warfarin dose. PLoS ONE, 13.","DOI":"10.1371\/journal.pone.0205872"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Lee, K.E., Chang, B.C., Kim, H.O., Yoon, I.K., Lee, N.R., Park, H.Y., and Gwak, H.S. (2021). Development of a system to support warfarin dose decisions using deep neural networks. Sci. Rep., 11.","DOI":"10.1038\/s41598-021-94305-2"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"W1","DOI":"10.7326\/M18-1377","article-title":"PROBAST: A Tool to Assess Risk of Bias and Applicability of Prediction Model Studies: Explanation and Elaboration","volume":"170","author":"Moons","year":"2019","journal-title":"Ann. Intern. Med."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"1722","DOI":"10.1093\/jamia\/ocac088","article-title":"Optimizing the dynamic treatment regime of in-hospital warfarin anticoagulation in patients after surgical valve replacement using reinforcement learning","volume":"29","author":"Zeng","year":"2022","journal-title":"J. Am. Med. Inform. Assoc."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"5219","DOI":"10.1109\/JBHI.2025.3545384","article-title":"Warfarin Dose Management Using Offline Deep Reinforcement Learning","volume":"29","author":"Ji","year":"2025","journal-title":"IEEE J. Biomed. Health Inform."},{"key":"ref_22","first-page":"45","article-title":"Optimizing warfarin dosing for patients with atrial fibrillation using machine learning","volume":"7","author":"Petch","year":"2024","journal-title":"NPJ Digit. Med."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Guo, B., Chen, C., Jia, J., Zheng, J., Song, Y., Liu, T., Zhang, K., Li, Y., and Dong, R. (2025). A Prediction Model of Stable Warfarin Doses in Patients After Mechanical Heart Valve Replacement Based on a Machine Learning Algorithm. Front. Pharmacol., 16.","DOI":"10.31083\/RCM33425"},{"key":"ref_24","unstructured":"Ganji, M., Karimi, M., Yazdi, S.M.A., Ebrahimi, A., Shirsalimi, N., Ghomi, F., Jahangiri, S., and Mahjub, H. (2025). Machine learning-based models for predicting warfarin maintenance dose and investigating factors affecting INR status in patients with cardiovascular diseases: A cross-sectional study. BMC Med. Inform. Decis. Mak., 25."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Wang, M., Qian, Y., Yang, Y., Chen, H., and Rao, W.F. (2024). Improved stacking ensemble learning based on feature selection to accurately predict warfarin dose. Front. Cardiovasc. Med., 10.","DOI":"10.3389\/fcvm.2023.1320938"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Choi, H., Kang, H.J., Ahn, I., Gwon, H., Kim, Y., Seo, H., Cho, H.N., Han, J., Kim, M., and Kee, G. (2023). Machine learning models to predict the warfarin discharge dosage using clinical information of inpatients from South Korea. Sci. Rep., 13.","DOI":"10.1038\/s41598-023-49831-6"},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Kuang, Y., Liu, Y., Pei, Q., Ning, X., Zou, Y., Liu, L., Song, L., Guo, C., Sun, Y., and Deng, K. (2022). Long Short-Term Memory Network for Development and Simulation of Warfarin Dosing Model Based on Time Series Anticoagulant Data. Front. Cardiovasc. Med., 9.","DOI":"10.3389\/fcvm.2022.881111"},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"e2612","DOI":"10.7717\/peerj-cs.2612","article-title":"Application of machine learning techniques for warfarin dosage prediction: A case study on the MIMIC-III dataset","volume":"11","author":"Wani","year":"2025","journal-title":"PeerJ Comput. Sci."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"e47262","DOI":"10.2196\/47262","article-title":"Evaluation of machine learning approaches for predicting warfarin discharge dose in cardiac surgery patients: Retrospective algorithm development and validation study","volume":"7","author":"Dryden","year":"2023","journal-title":"JMIR Cardio"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Dai, M.F., Li, S.Y., Zhang, J.F., Wang, B.Y., Zhou, L., Yu, F., Xu, H., Ge, W.H., Nasiri, H., and Liu, K. (2022). Warfarin anticoagulation management during the COVID-19 pandemic: The role of internet clinic and machine learning. Comput. Biol. Med., 150.","DOI":"10.3389\/fphar.2022.933156"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Bontempi, M., Borgese, L., Visani, A., Giavaresi, G., and Cosmi, B. (2025). Semi-empirical Anticoagulation Model (SAM): Dose prediction during warfarin therapy. Comput. Biol. Med., 168.","DOI":"10.1016\/j.compbiomed.2025.110010"},{"key":"ref_32","first-page":"717","article-title":"Development and validation of a mobile application based on a machine learning model to aid in predicting dosage of vitamin K antagonists among Indian patients post mechanical heart valve replacement","volume":"12","author":"Amruthlal","year":"2022","journal-title":"J. Pers. Med."},{"key":"ref_33","first-page":"145","article-title":"Machine learning guided prediction of warfarin blood levels for personalized medicine based on clinical longitudinal data from cardiac surgery patients: A prospective observational study","volume":"22","author":"Xue","year":"2024","journal-title":"J. Transl. Med."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Roche-Lima, A., Roman-Santiago, A., Feliu-Maldonado, R., Rodriguez-Maldonado, J., Nieves-Rodriguez, B.G., Carrasquillo-Carrion, K., Ramos, A.S., Cruz, I.A., Gonzalez-Sanchez, J.A., and Hernandez-O\u2019Farrill, E. (2020). Machine learning algorithm for predicting warfarin dose in Caribbean Hispanics using pharmacogenetic data. Front. Pharmacol., 10.","DOI":"10.3389\/fphar.2019.01550"},{"key":"ref_35","doi-asserted-by":"crossref","unstructured":"Duconge, J., Ramos, A.S., Claudio-Campos, K., Rivera-Miranda, G., Berm\u00fadez-Bosch, L., Renta, J.Y., Cadilla, C.L., Cruz, I., Feliu, J.F., and Vergara, C. (2016). A novel admixture-based pharmacogenetic approach to refine warfarin dosing in Caribbean Hispanics. PLoS ONE, 11.","DOI":"10.1371\/journal.pone.0145480"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"160035","DOI":"10.1038\/sdata.2016.35","article-title":"MIMIC-III, a freely accessible critical care database","volume":"3","author":"Johnson","year":"2016","journal-title":"Sci. Data"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"649","DOI":"10.1038\/s41569-021-00549-w","article-title":"Pharmacogenetics to guide cardiovascular drug therapy","volume":"18","author":"Duarte","year":"2021","journal-title":"Nat. Rev. Cardiol."},{"key":"ref_38","unstructured":"(2025, August 25). Available online: https:\/\/www.fda.gov\/medical-devices\/software-medical-device-samd\/artificial-intelligence-software-medical-device."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"645469","DOI":"10.1155\/2012\/645469","article-title":"Long-Term Costs of Ischemic Stroke and Major Bleeding Events among Medicare Patients with Nonvalvular Atrial Fibrillation","volume":"2012","author":"Mercaldi","year":"2012","journal-title":"Cardiol. Res. Pract."},{"key":"ref_40","unstructured":"Alkhaldi, N., and ITRex Group (2025, August 25). Assessing the Cost of Implementing AI in Healthcare [Internet]. Available online: https:\/\/itrexgroup.com\/blog\/assessing-the-costs-of-implementing-ai-in-healthcare\/."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"206","DOI":"10.1038\/s42256-019-0048-x","article-title":"Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead","volume":"1","author":"Rudin","year":"2019","journal-title":"Nat. Mach. Intell."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"18069","DOI":"10.1007\/s00521-019-04051-w","article-title":"The importance of interpretability and visualization in machine learning for applications in medicine and health care","volume":"32","author":"Vellido","year":"2020","journal-title":"Neural Comput. Appl."},{"key":"ref_43","unstructured":"Lundberg, S.M., and Lee, S.I. (2017, January 4\u20139). A unified approach to interpreting model predictions. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA."},{"key":"ref_44","doi-asserted-by":"crossref","unstructured":"Ribeiro, M.T., Singh, S., and Guestrin, C. (2016, January 13\u201317). \u201cWhy should I trust you?\u201d: Explaining the predictions of any classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, CA, USA.","DOI":"10.1145\/2939672.2939778"},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AIstatement: Updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"}],"container-title":["Pharmaceuticals"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8247\/18\/10\/1544\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,28]],"date-time":"2025-10-28T11:51:54Z","timestamp":1761652314000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8247\/18\/10\/1544"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,14]]},"references-count":45,"journal-issue":{"issue":"10","published-online":{"date-parts":[[2025,10]]}},"alternative-id":["ph18101544"],"URL":"https:\/\/doi.org\/10.3390\/ph18101544","relation":{},"ISSN":["1424-8247"],"issn-type":[{"value":"1424-8247","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,10,14]]}}}