{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T18:20:56Z","timestamp":1788459656976,"version":"build-2803163510"},"reference-count":38,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T00:00:00Z","timestamp":1788393600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Class imbalance and limited minority-class samples remain major challenges for developing reliable clinical diagnostic models, as insufficient observations often fail to capture complex minority-class distributions. Existing augmentation methods either rely on heuristic interpolation or suffer from instability when learning sparse and heterogeneous medical tabular data. This study proposes IFC-HFlowVAE, an iterative feedback and consensus framework built upon a flow-enhanced heterogeneous variational autoencoder for minority-class data generation. The proposed framework first models mixed-type clinical attributes through HFlowVAE and then introduces a self-circulating refinement strategy to progressively improve generated samples. To alleviate degradation during iterative refinement, SMOTE-generated samples are incorporated as structural references, followed by a localized GMM-based refinement procedure that guides samples toward more representative minority-class regions. Comprehensive experiments on seven clinical tabular datasets demonstrate that IFC-HFlowVAE improves downstream classification performance compared with existing augmentation approaches. Furthermore, fidelity analyses show that the proposed framework better preserves challenging minority-class characteristics, particularly skewed and multimodal numerical distributions.<\/jats:p>","DOI":"10.1371\/journal.pone.0357260","type":"journal-article","created":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T17:53:09Z","timestamp":1788457989000},"page":"e0357260","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["IFC-HFlowVAE: A self-enhancing generative framework with structural anchoring for imbalanced clinical data augmentation"],"prefix":"10.1371","volume":"21","author":[{"given":"Lu","family":"Yuwen","sequence":"first","affiliation":[{"name":"School of Big Data & Software Engineering, Chongqing University","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4515-3292","authenticated-orcid":true,"given":"Shuyu","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Big Data & Software Engineering, Chongqing University","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,3]]},"reference":[{"key":"pone.0357260.ref001","first-page":"100700","article-title":"Projected rapid growth in diabetes disease burden and economic burden in China: a spatio-temporal study from 2020 to 2030","volume":"33","author":"J Liu","year":"2023","journal-title":"Lancet Reg Health West Pac"},{"issue":"8","key":"pone.0357260.ref002","doi-asserted-by":"crossref","first-page":"957","DOI":"10.3390\/bioengineering10080957","article-title":"RGSB-UNet: Hybrid Deep Learning Framework for Tumour Segmentation in Digital Pathology Images","volume":"10","author":"T Zhao","year":"2023","journal-title":"Bioengineering (Basel)"},{"issue":"1","key":"pone.0357260.ref003","doi-asserted-by":"crossref","first-page":"16","DOI":"10.3390\/bioengineering11010016","article-title":"RGGC-UNet: Accurate Deep Learning Framework for Signet Ring Cell Semantic Segmentation in Pathological Images","volume":"11","author":"T Zhao","year":"2023","journal-title":"Bioengineering (Basel)"},{"issue":"3","key":"pone.0357260.ref004","doi-asserted-by":"crossref","first-page":"393","DOI":"10.3390\/bioengineering10030393","article-title":"GSN-HVNET: A Lightweight, Multi-Task Deep Learning Framework for Nuclei Segmentation and Classification","volume":"10","author":"T Zhao","year":"2023","journal-title":"Bioengineering (Basel)"},{"key":"pone.0357260.ref005","doi-asserted-by":"crossref","first-page":"110853","DOI":"10.1016\/j.dib.2024.110853","article-title":"Heterogeneous data integration: Challenges and opportunities","volume":"56","author":"IM Putrama","year":"2024","journal-title":"Data Brief"},{"issue":"4","key":"pone.0357260.ref006","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3343440","article-title":"A systematic review on imbalanced data challenges in machine learning: applications and solutions","volume":"52","author":"H Kaur","year":"2019","journal-title":"ACM Computing Surveys"},{"issue":"2","key":"pone.0357260.ref007","article-title":"Suboptimal capability of individual machine learning algorithms in modeling small-scale imbalanced clinical data of local hospital","volume":"19","author":"G Li","year":"2024","journal-title":"PLoS One"},{"key":"pone.0357260.ref008","doi-asserted-by":"crossref","first-page":"112677","DOI":"10.1016\/j.asoc.2024.112677","article-title":"Deep generative approaches for oversampling in imbalanced data classification problems: A comprehensive review and comparative analysis","volume":"170","author":"M Hayaeian Shirvan","year":"2025","journal-title":"Applied Soft Computing"},{"key":"pone.0357260.ref009","doi-asserted-by":"crossref","first-page":"228","DOI":"10.1016\/j.inffus.2022.08.017","article-title":"A unifying view of class overlap and imbalance: Key concepts, multi-view panorama, and open avenues for research","volume":"89","author":"MS Santos","year":"2023","journal-title":"Information Fusion"},{"key":"pone.0357260.ref010","doi-asserted-by":"crossref","unstructured":"Prati RC, Batista GE, Monard MC. Class imbalances versus class overlapping: an analysis of a learning system behavior. In: 2004.","DOI":"10.1007\/978-3-540-24694-7_32"},{"issue":"2","key":"pone.0357260.ref011","doi-asserted-by":"crossref","first-page":"827","DOI":"10.3390\/make6020039","article-title":"Impact of Nature of Medical Data on Machine and Deep Learning for Imbalanced Datasets: Clinical Validity of SMOTE Is Questionable","volume":"6","author":"S Gholampour","year":"2024","journal-title":"MAKE"},{"key":"pone.0357260.ref012","doi-asserted-by":"crossref","first-page":"1430245","DOI":"10.3389\/fdgth.2024.1430245","article-title":"A review on over-sampling techniques in classification of multi-class imbalanced datasets: insights for medical problems","volume":"6","author":"Y Yang","year":"2024","journal-title":"Front Digit Health"},{"key":"pone.0357260.ref013","doi-asserted-by":"crossref","first-page":"321","DOI":"10.1613\/jair.953","article-title":"SMOTE: Synthetic Minority Over-sampling Technique","volume":"16","author":"NV Chawla","year":"2002","journal-title":"jair"},{"key":"pone.0357260.ref014","doi-asserted-by":"crossref","unstructured":"Han H, Wang WY, Mao BH. Borderline-SMOTE: A New Over-Sampling Method in Imbalanced Data Sets Learning. In: 2005.","DOI":"10.1007\/11538059_91"},{"key":"pone.0357260.ref015","unstructured":"He H, Bai Y, Garcia EA, Li S. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In: Ieee, 2008."},{"key":"pone.0357260.ref016","doi-asserted-by":"crossref","first-page":"184","DOI":"10.1016\/j.ins.2014.08.051","article-title":"SMOTE\u2013IPF: Addressing the noisy and borderline examples problem in imbalanced classification by a re-sampling method with filtering","volume":"291","author":"JA S\u00e1ez","year":"2015","journal-title":"Information Sciences"},{"key":"pone.0357260.ref017","doi-asserted-by":"crossref","first-page":"863","DOI":"10.1613\/jair.1.11192","article-title":"SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary","volume":"61","author":"A Fernandez","year":"2018","journal-title":"jair"},{"key":"pone.0357260.ref018","article-title":"Understanding disentangling in Beta-VAE","author":"CP Burgess","year":"2018","journal-title":"arXiv preprint"},{"key":"pone.0357260.ref019","article-title":"Infovae: Information maximizing variational autoencoders","author":"S Zhao","year":"2017","journal-title":"arXiv preprint"},{"key":"pone.0357260.ref020","article-title":"Modeling tabular data using conditional gan","volume":"32","author":"L Xu","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"pone.0357260.ref021","unstructured":"Zhao Z, Kunar A, Birke R, Chen LY. Ctab-gan: Effective table data synthesizing. In: PMLR, 2021."},{"key":"pone.0357260.ref022","doi-asserted-by":"crossref","first-page":"3428","DOI":"10.1109\/ACCESS.2023.3348810","article-title":"Advancing precision medicine: VAE enhanced predictions of pancreatic cancer patient survival in local hospital","volume":"12","author":"Y Wang","year":"2024","journal-title":"IEEE Access"},{"key":"pone.0357260.ref023","doi-asserted-by":"crossref","first-page":"112223","DOI":"10.1016\/j.asoc.2024.112223","article-title":"Challenges and opportunities of generative models on tabular data","volume":"166","author":"AX Wang","year":"2024","journal-title":"Applied Soft Computing"},{"issue":"11","key":"pone.0357260.ref024","doi-asserted-by":"crossref","first-page":"3964","DOI":"10.1109\/TPAMI.2020.2992934","article-title":"Normalizing Flows: An Introduction and Review of Current Methods","volume":"43","author":"I Kobyzev","year":"2021","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"12","key":"pone.0357260.ref025","doi-asserted-by":"crossref","first-page":"123007","DOI":"10.1103\/PhysRevD.110.123007","article-title":"Bayesian evidence estimation from posterior samples with normalizing flows","volume":"110","author":"R Srinivasan","year":"2024","journal-title":"Physical Review D"},{"key":"pone.0357260.ref026","first-page":"5443","article-title":"Posterior collapse and latent variable non-identifiability","volume":"34","author":"Y Wang","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"pone.0357260.ref027","doi-asserted-by":"crossref","first-page":"30655","DOI":"10.1109\/ACCESS.2022.3158977","article-title":"SMOTified-GAN for Class Imbalanced Pattern Classification Problems","volume":"10","author":"A Sharma","year":"2022","journal-title":"IEEE Access"},{"issue":"3","key":"pone.0357260.ref028","doi-asserted-by":"crossref","first-page":"110","DOI":"10.3390\/ijfs11030110","article-title":"Enhancing Financial Fraud Detection through Addressing Class Imbalance Using Hybrid SMOTE-GAN Techniques","volume":"11","author":"PCY Cheah","year":"2023","journal-title":"IJFS"},{"issue":"4","key":"pone.0357260.ref029","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3704437","article-title":"Systematic Review of Generative Modelling Tools and Utility Metrics for Fully Synthetic Tabular Data","volume":"57","author":"AD Lautrup","year":"2024","journal-title":"ACM Comput Surv"},{"key":"pone.0357260.ref030","article-title":"Self-consuming generative models go mad","author":"S Alemohammad","year":"2023","journal-title":"arXiv preprint"},{"key":"pone.0357260.ref031","doi-asserted-by":"crossref","unstructured":"Lin S, Wang K, Zeng X, Zhao R. Explore the power of synthetic data on few-shot object detection. In: 2023.","DOI":"10.1109\/CVPRW59228.2023.00071"},{"key":"pone.0357260.ref032","unstructured":"Bansal H, Grover A. Leaving reality to imagination: robust classification via generated datasets. arXiv preprint. 2023. https:\/\/doi.org\/10.48550\/arXiv.230202503"},{"key":"pone.0357260.ref033","unstructured":"Kotelnikov A, Baranchuk D, Rubachev I, Babenko A. Tabddpm: Modelling tabular data with diffusion models [Conference Paper]. PMLR; 2023."},{"key":"pone.0357260.ref034","unstructured":"Alishahi M, Zheng Y, Wang J, Yeh CCM, Phillips JM. TabKDE: Simple and scalable tabular data generation with kernel density estimates. In: 2026. https:\/\/arxiv.org\/abs\/260517642"},{"key":"pone.0357260.ref035","doi-asserted-by":"crossref","first-page":"108","DOI":"10.1109\/OJEMB.2022.3181796","article-title":"Bayesian Inference-Based Gaussian Mixture Models With Optimal Components Estimation Towards Large-Scale Synthetic Data Generation for In Silico Clinical Trials","volume":"3","author":"VC Pezoulas","year":"2022","journal-title":"IEEE Open J Eng Med Biol"},{"key":"pone.0357260.ref036","volume-title":"Foundations of machine learning","author":"M Mohri","year":"2018"},{"key":"pone.0357260.ref037","unstructured":"Rezende D, Mohamed S. In: 2015. 1530\u20138."},{"key":"pone.0357260.ref038","unstructured":"Jolicoeur-Martineau A, Fatras K, Kachman T. Generating and imputing tabular data via diffusion and flow-based gradient-boosted trees. In: International conference on artificial intelligence and statistics, 2024. 1288\u201396."}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0357260","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T17:53:18Z","timestamp":1788457998000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0357260"}},"subtitle":[],"editor":[{"given":"Zeheng","family":"Wang","sequence":"first","affiliation":[{"name":"Commonwealth Scientific and Industrial Research Organisation","place":["AUSTRALIA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,3]]},"references-count":38,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,3]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0357260","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,3]]}}}