{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,28]],"date-time":"2026-07-28T13:05:16Z","timestamp":1785243916826,"version":"3.55.0"},"reference-count":48,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Med Inform"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Despite the increasing use of machine learning (ML) in clinical research, the early stages of data preparation, especially for structured clinical data, often receive limited methodological scrutiny. These datasets typically contain missing values, complex categorical variables, and imbalanced class distributions, all of which complicate downstream model development and interpretation.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study introduces a structured preprocessing framework designed to address common challenges in medical tabular data and to assess how preprocessing choices affect the stability and portability of predictive models across settings.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n We constructed a modular workflow comprising 3 components. First, preprocessing strategies included imputation for missing data, 3 types of categorical encoding (one-hot, frequency, and target), and resampling approaches for class imbalance (Synthetic Minority Over-sampling Technique [SMOTE] and Random Over Sampling Example [ROSE]). Second, 6 classification algorithms were used to evaluate performance patterns, including logistic regression (LGR), decision tree (DT), random forest, XGBoost (XGB), CatBoost (CAT), and light gradient-boosting machine (LightGBM). Third, we assessed cross-dataset portability using 2 datasets with distinct data-generating mechanisms: a registry for patients with end-stage renal disease (ESRD; n=412) and the population-based Behavioral Risk Factor Surveillance System (BRFSS) 2015 survey. For each dataset, we independently cleaned, standardized, encoded, tuned, and evaluated models using the same predefined hyperparameter search space, without cross-dataset feature matching or pooling the area under the ROC curve (AUC) calculations; the complete pipeline was then rerun on BRFSS as an external replication.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n One-hot encoding in combination with ROSE yielded the most consistent performance improvements in terms of AUC (0.940) and accuracy (0.932), particularly for classifiers sensitive to class distribution. Notably, ROSE enhanced sensitivity without substantially distorting the original data structure. Feature importance rankings also contributed to model interpretability, and performance trends were largely reproducible in cross-context application.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n Our findings suggest that preprocessing decisions often treated as ancillary play a central role in shaping model outcomes, especially in high-variance clinical datasets. The proposed framework offers a reproducible and adaptable tool for aligning data preparation with the unique demands of health care prediction tasks and may serve as a foundation for future efforts to standardize preprocessing in clinical ML workflows.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/75655","type":"journal-article","created":{"date-parts":[[2026,2,5]],"date-time":"2026-02-05T15:25:06Z","timestamp":1770305106000},"page":"e75655-e75655","source":"Crossref","is-referenced-by-count":1,"title":["Enhancing Model Generalizability in Medical Artificial Intelligence: Systematic Comparison of Categorical Encoding and Sampling Techniques for Imbalanced Data"],"prefix":"10.2196","volume":"14","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3592-4794","authenticated-orcid":false,"given":"Chien-wei","family":"Chuang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4446-0167","authenticated-orcid":false,"given":"Chung-Kuan","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-7379-9576","authenticated-orcid":false,"given":"Chao-Hsin","family":"Wu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2854-8361","authenticated-orcid":false,"given":"Ben-Chang","family":"Shia","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8278-0033","authenticated-orcid":false,"given":"Mingchih","family":"Chen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,4,13]]},"reference":[{"key":"R1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cmpb.2022.106773","article-title":"Diabetes mellitus prediction and diagnosis from a data preprocessing and machine learning perspective","volume":"220","author":"Olisah","journal-title":"Comput Methods Programs Biomed"},{"issue":"4","key":"R2","doi-asserted-by":"publisher","first-page":"808","DOI":"10.1055\/s-0041-1735184","article-title":"Systematic review of approaches to preserve machine learning performance in the presence of temporal dataset shift in clinical medicine","volume":"12","author":"Guo","journal-title":"Appl Clin Inform"},{"key":"R3","unstructured":"Anderson K . The role of data preprocessing in machine learning accuracy for heart disease prediction hybrid models for heart disease prediction: combining neural networks with traditional. 2024. URL: https:\/\/www.researchgate.net\/publication\/386177388_The_Role_of_Data_Preprocessing_in_Machine_Learning_Accuracy_for_Heart_Disease_Prediction_Hybrid_Models_for_Heart_Disease_Prediction_Combining_Neural_Networks_with_Traditional [Accessed 30-03-2026]"},{"key":"R4","volume":"17","author":"Rahman","journal-title":"Int J Artif Intell"},{"issue":"5","key":"R5","doi-asserted-by":"publisher","DOI":"10.2196\/36388","article-title":"Evaluation and mitigation of racial bias in clinical machine learning models: scoping review","volume":"10","author":"Huang","journal-title":"JMIR Med Inform"},{"issue":"1","key":"R6","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-019-0747-6","article-title":"Importance of medical data preprocessing in predictive modeling and risk factor discovery for the frailty syndrome","volume":"19","author":"Hassler","journal-title":"BMC Med Inform Decis Mak"},{"issue":"4","key":"R7","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics9040178","article-title":"A machine-learning-based prediction method for hypertension outcomes based on medical data","volume":"9","author":"Chang","journal-title":"Diagnostics (Basel)"},{"issue":"6","key":"R8","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics10060415","article-title":"Comparison between statistical models and machine learning methods on classification for highly imbalanced multiclass kidney data","volume":"10","author":"Jeong","journal-title":"Diagnostics (Basel)"},{"issue":"1","key":"R9","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-019-0681-4","article-title":"Machine learning in medicine: a practical introduction","volume":"19","author":"Sidey-Gibbons","journal-title":"BMC Med Res Methodol"},{"issue":"7","key":"R10","doi-asserted-by":"publisher","first-page":"4119","DOI":"10.3390\/app13074119","article-title":"Evaluation of synthetic categorical data generation techniques for predicting cardiovascular diseases and post-hoc interpretability of the risk factors","volume":"13","author":"Garc\u00eda-Vicente","journal-title":"Appl Sci (Basel)"},{"key":"R11","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2021.3102399","article-title":"A comparative performance analysis of data resampling methods on imbalance medical data","volume":"9","author":"Khushi","journal-title":"IEEE Access"},{"key":"R12","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2021.103980","article-title":"Deep learning for temporal data representation in electronic health records: a systematic review of challenges and methodologies","volume":"126","author":"Xie","journal-title":"J Biomed Inform"},{"key":"R13","doi-asserted-by":"publisher","DOI":"10.22067\/cke.2024.88940.1121","article-title":"Machine learning classifiers and data synthesis techniques to tackle with highly imbalanced COVID-19 data","volume":"7","author":"Naghipour","journal-title":"Comput Knowl Eng"},{"issue":"5","key":"R14","doi-asserted-by":"publisher","first-page":"2671","DOI":"10.1007\/s00180-022-01207-6","article-title":"Regularized target encoding outperforms traditional methods in supervised machine learning with high cardinality features","volume":"37","author":"Pargent","journal-title":"Comput Stat"},{"issue":"2","key":"R15","doi-asserted-by":"publisher","first-page":"472","DOI":"10.1080\/1540496X.2020.1825935","article-title":"Missing data preprocessing in credit classification: one-hot encoding or imputation?","volume":"58","author":"Yu","journal-title":"Emerg Mark Finance Trade"},{"issue":"23","key":"R16","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics14232634","article-title":"The impact of the SMOTE method on machine learning and ensemble learning performance results in addressing class imbalance in data used for predicting total testosterone deficiency in type 2 diabetes patients","volume":"14","author":"Kivrak","journal-title":"Diagnostics (Basel)"},{"issue":"9","key":"R17","doi-asserted-by":"publisher","first-page":"1242","DOI":"10.5551\/jat.63870","article-title":"Left ventricular hypertrophy geometry and vascular calcification co-modify the risk of cardiovascular mortality in patients with end-stage kidney disease: a retrospective cohort study","volume":"30","author":"Chao","journal-title":"J Atheroscler Thromb"},{"key":"R18","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijcha.2024.101505","article-title":"Combinations of valvular calcification and serum alkaline phosphatase predict cardiovascular risk among end-stage kidney disease patients","volume":"54","author":"Chao","journal-title":"IJC Heart Vasculature"},{"issue":"1","key":"R19","doi-asserted-by":"publisher","first-page":"112","DOI":"10.1093\/bioinformatics\/btr597","article-title":"MissForest--non-parametric missing value imputation for mixed-type data","volume":"28","author":"Stekhoven","journal-title":"Bioinformatics"},{"issue":"8","key":"R20","doi-asserted-by":"publisher","DOI":"10.1136\/bmjopen-2013-002847","article-title":"Comparison of imputation methods for missing laboratory data in medicine","volume":"3","author":"Waljee","journal-title":"BMJ Open"},{"key":"R21","doi-asserted-by":"publisher","DOI":"10.1109\/IEMBS.2009.5334548","article-title":"A frequency based encoding technique for transformation of categorical variables in mixed IVF dataset","volume":"2009","author":"Uyar","journal-title":"Annu Int Conf IEEE Eng Med Biol Soc"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","first-page":"79","DOI":"10.32614\/RJ-2014-008","article-title":"ROSE: a package for binary imbalanced learning","volume":"6","author":"Lunardon","journal-title":"R J"},{"issue":"1","key":"R23","doi-asserted-by":"publisher","first-page":"92","DOI":"10.1007\/s10618-012-0295-5","article-title":"Training and assessing classification rules with imbalanced data","volume":"28","author":"Menardi","journal-title":"Data Min Knowl Disc"},{"key":"R24","doi-asserted-by":"publisher","DOI":"10.1613\/jair.953","article-title":"SMOTE: synthetic minority over-sampling technique","volume":"16","author":"Chawla","journal-title":"JAIR"},{"key":"R25","doi-asserted-by":"crossref","unstructured":"Kamalov F Leung HH Cherukuri AK . Keep it simple: random oversampling for imbalanced data. Presented at: 2023 Advances in Science and Engineering Technology International Conferences (ASET); Feb 20-23, 2023:1-4; Dubai, United Arab Emirates. 2023.[doi: 10.1109\/ASET56582.2023.10180891]","DOI":"10.1109\/ASET56582.2023.10180891"},{"key":"R26","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.604","article-title":"Selective oversampling approach for strongly imbalanced data","volume":"7","author":"Gnip","journal-title":"PeerJ Comput Sci"},{"key":"R27","doi-asserted-by":"crossref","unstructured":"Nguyen S Schumacher P Olinsky A Quinn J . Effects of resampling techniques on imbalanced data classification: a new under-resampling method. In: Advances in Business and Management Forecasting. Emerald Publishing Limited; 2021:51-70. [doi: 10.1108\/S1477-407020210000014005]","DOI":"10.1108\/S1477-407020210000014005"},{"key":"R28","doi-asserted-by":"crossref","unstructured":"Berrar D . Cross-validation. In: Reference Module in Life Sciences. Elsevier; 2019:542-545. [doi: 10.1016\/B978-0-12-809633-8.20349-X]","DOI":"10.1016\/B978-0-12-809633-8.20349-X"},{"key":"R29","unstructured":"Therneau T Atkinson B Ripley B Ripley MB . Package \u2018rpart. rpart: Recursive Partitioning and Regression Trees. 2015. URL: https:\/\/cran.r-project.org\/web\/packages\/rpart\/rpart.pdf [Accessed 18-03-2026]"},{"issue":"2","key":"R30","doi-asserted-by":"publisher","first-page":"45","DOI":"10.32614\/RJ-2009-016","article-title":"Rattle: a data mining GUI for R","volume":"1","author":"Williams","journal-title":"R J"},{"issue":"1","key":"R31","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random Forests","volume":"45","author":"Breiman","journal-title":"Mach Learn"},{"key":"R32","unstructured":"Chen T He T Benesty M et al. Xgboost: extreme gradient boosting. r package version 0.4-2. Vol 1. Tianqi Chen; 2015:1-4. URL: https:\/\/cran.r-project.org\/web\/packages\/xgboost\/index.html [Accessed 30-03-2026]"},{"key":"R33","unstructured":"Dorogush AV Ershov V Gulin A . CatBoost: gradient boosting with categorical features support. arXiv. Preprint posted online on Oct 24, 2018. URL: https:\/\/arxiv.org\/abs\/1810.11363 [Accessed 30-03-2026] [doi: 10.48550\/arXiv.1810.11363]"},{"key":"R34","doi-asserted-by":"publisher","DOI":"10.5555\/3294996.3295074","article-title":"LightGBM: a highly efficient gradient boosting decision tree","author":"Ke","journal-title":"Adv Neural Inf Process Syst"},{"issue":"5","key":"R35","doi-asserted-by":"crossref","DOI":"10.18637\/jss.v028.i05","volume":"28","author":"Kuhn","journal-title":"J Stat Soft"},{"key":"R36","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2105-12-77","article-title":"pROC: an open-source package for R and S+ to analyze and compare ROC curves","volume":"12","author":"Robin","journal-title":"BMC Bioinformatics"},{"issue":"7","key":"R37","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.64771","article-title":"Pathogenesis and mechanism of uremic vascular calcification","volume":"16","author":"Shen","journal-title":"Cureus"},{"issue":"4","key":"R38","doi-asserted-by":"publisher","DOI":"10.1002\/jbm4.10610","article-title":"The calcified vasculature in chronic kidney disease secretes factors that inhibit bone mineralization","volume":"6","author":"Mace","journal-title":"JBMR Plus"},{"issue":"12","key":"R39","doi-asserted-by":"publisher","DOI":"10.3390\/toxins12120812","article-title":"Uremic vascular calcification: the pathogenic roles and gastrointestinal decontamination of uremic toxins","volume":"12","author":"Chao","journal-title":"Toxins (Basel)"},{"key":"R40","doi-asserted-by":"publisher","DOI":"10.23876\/j.krcp.23.290","article-title":"Effect of the geometry and severity of left ventricular hypertrophy on cardiovascular mortality in dialysis patients","author":"Huang","journal-title":"Kidney Res Clin Pract"},{"issue":"1","key":"R41","doi-asserted-by":"publisher","DOI":"10.1186\/s12882-021-02321-1","article-title":"Prognostic significance of concentric left ventricular hypertrophy at peritoneal dialysis initiation","volume":"22","author":"Tomura","journal-title":"BMC Nephrol"},{"issue":"9","key":"R42","doi-asserted-by":"publisher","DOI":"10.3390\/biomedicines10092094","article-title":"Atherosclerosis specific features in chronic kidney disease (CKD)","volume":"10","author":"Poznyak","journal-title":"Biomedicines"},{"key":"R43","unstructured":"Lacount S Tannock LR . Dyslipidemia in chronic kidney disease. In: Endotext. 2025. [Medline: 26247091]"},{"issue":"5","key":"R44","doi-asserted-by":"publisher","first-page":"592","DOI":"10.1161\/CIRCRESAHA.123.324001","article-title":"Cardiovascular consequences of uremic metabolites: an overview of the involved signaling pathways","volume":"134","author":"Curaj","journal-title":"Circ Res"},{"key":"R45","doi-asserted-by":"publisher","DOI":"10.1016\/j.atherosclerosis.2019.12.016","article-title":"Chronic kidney disease and outcomes of lower extremity revascularization for peripheral artery disease","volume":"297","author":"Smilowitz","journal-title":"Atherosclerosis"},{"issue":"1","key":"R46","doi-asserted-by":"publisher","DOI":"10.1186\/s12882-019-1499-3","article-title":"Effect of antiplatelet therapy on cardiovascular and kidney outcomes in patients with chronic kidney disease: a systematic review and meta-analysis","volume":"20","author":"Su","journal-title":"BMC Nephrol"},{"issue":"1-2","key":"R47","doi-asserted-by":"publisher","first-page":"158","DOI":"10.1159\/000503776","article-title":"Intradialytic hypotension: mechanisms and outcome","volume":"49","author":"Sars","journal-title":"Blood Purif"},{"issue":"7","key":"R48","doi-asserted-by":"publisher","DOI":"10.1161\/JAHA.121.023466","article-title":"Left ventricular hypertrophy and biomarkers of cardiac damage and stress in aortic stenosis","volume":"11","author":"Stein","journal-title":"J Am Heart Assoc"}],"container-title":["JMIR Medical Informatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e75655","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:06:46Z","timestamp":1782842806000},"score":1,"resource":{"primary":{"URL":"https:\/\/medinform.jmir.org\/2026\/1\/e75655"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,13]]},"references-count":48,"URL":"https:\/\/doi.org\/10.2196\/75655","relation":{},"ISSN":["2291-9694"],"issn-type":[{"value":"2291-9694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,13]]},"article-number":"v14i4e75655"}}