{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T04:57:34Z","timestamp":1781672254781,"version":"3.54.5"},"reference-count":30,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T00:00:00Z","timestamp":1781654400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Med."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Class imbalance is a fundamental challenge in medical image analysis, where certain disease categories occur far less frequently than others. This uneven data distribution often causes learning algorithms to favor common conditions while underperforming on rare but clinically significant cases. In chest X-ray analysis, the deployment of artificial intelligence is particularly hindered by the long-tailed distribution of thoracic diseases, as conventional deep learning models exhibit optimization bias toward majority classes, leading to reduced sensitivity for rare yet critical pathologies.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      To address this challenge, this paper presents Dynamic Adaptive Weighting with Hybrid Networks (DAWN-Net), a unified framework that synergizes data-level and algorithm-level interventions. Unlike conventional approaches that treat augmentation and re-weighting in isolation, DAWN-Net introduces a Hybrid Synergy mechanism. At the data level, we propose\n                      <jats:italic>Augmentation-Aware Manifold Smoothing<\/jats:italic>\n                      , which generates synthetic variations in the local tangent space of minority samples to densify their feature representation. Architecturally, the model employs a dual-stream design comprising a\n                      <jats:italic>Hierarchical Feature Propagation Network (HFPN)<\/jats:italic>\n                      to capture high-frequency local textural details, and a\n                      <jats:italic>Semantic Context Modeling Network (SCMN)<\/jats:italic>\n                      to enforce global anatomical consistency. These components are jointly optimized using a novel\n                      <jats:italic>Momentum-Adjusted Gradient Harmonization<\/jats:italic>\n                      loss, which dynamically recalibrates gradient contributions based on batch-wise class statistics and augmentation intensity.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Validation on three large-scale benchmarks\u2014NIH ChestXray14, CheXpert, and PadChest\u2014demonstrates that DAWN-Net consistently outperforms state-of-the-art baselines, particularly in the detection of rare diseases such as Hernia and Fibrosis.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>By mitigating the optimization bias and improving sensitivity for rare yet critical pathologies, DAWN-Net overcomes the limitations of conventional deep learning models, thereby offering a more reliable solution for safety-critical radiological diagnosis.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1795508","type":"journal-article","created":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T04:47:03Z","timestamp":1781671623000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Integrating augmentation-aware manifold smoothing and momentum-adjusted loss for handling class imbalance in thoracic disease detection"],"prefix":"10.3389","volume":"13","author":[{"given":"Sneha","family":"H. R","sequence":"first","affiliation":[{"name":"Department of Information Science and Engineering, Nitte Meenakshi Institute of Technology (NMIT)","place":["Nitte (Deemed to be University), Bangalore, India"]},{"name":"Depatment of Computer Science and Engineering, National Institute of Technology Karnataka (NITK)","place":["Surathkal, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Annappa","family":"B","sequence":"additional","affiliation":[{"name":"Depatment of Computer Science and Engineering, National Institute of Technology Karnataka (NITK)","place":["Surathkal, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,6,17]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"463","DOI":"10.1109\/TSMCC.2011.2161285","article-title":"A review on ensembles for the class imbalance problem: bagging-, boosting-, and hybrid-based approaches","volume":"42","author":"Galar","year":"2012","journal-title":"IEEE Trans Syst Man, Cybernet Part C"},{"key":"B2","doi-asserted-by":"publisher","first-page":"790","DOI":"10.1109\/ICIP.2017.8296389","article-title":"DenseNet for dense flow","author":"Zhu","year":"2017","journal-title":"2017 IEEE International Conference on Image Processing (ICIP)"},{"key":"B3","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40537-019-0192-5","article-title":"Survey on deep learning with class imbalance","volume":"6","author":"Johnson","year":"2019","journal-title":"J Big Data"},{"key":"B4","doi-asserted-by":"publisher","first-page":"224","DOI":"10.7763\/IJMLC.2013.V3.307","article-title":"Addressing the class imbalance problem in medical datasets","volume":"3","author":"Rahman","year":"2013","journal-title":"Int J Mach Learn Comput"},{"key":"B5","doi-asserted-by":"publisher","first-page":"101935","DOI":"10.1016\/j.artmed.2020.101935","article-title":"Handling imbalanced medical image data: a deep-learning-based one-class classification approach","volume":"108","author":"Gao","year":"2020","journal-title":"Artif Intell Med"},{"key":"B6","doi-asserted-by":"publisher","first-page":"7895","DOI":"10.1007\/s00330-024-10834-0","article-title":"Class imbalance on medical image classification: towards better evaluation practices for discrimination and calibration performance","volume":"34","author":"Mosquera","year":"2024","journal-title":"Eur Radiol"},{"key":"B7","doi-asserted-by":"publisher","first-page":"69","DOI":"10.1007\/s10044-024-01285-w","article-title":"Few-shot learning for COVID-19 chest X-ray classification with imbalanced data: an inter vs. intra domain study","volume":"27","author":"Gal\u00e1n-Cuenca","year":"2024","journal-title":"Patt Analy Applic"},{"key":"B8","doi-asserted-by":"publisher","first-page":"108518","DOI":"10.1016\/j.cmpb.2024.108518","article-title":"Advancing chest X-ray diagnostics: a novel CycleGAN-based preprocessing approach for enhanced lung disease classification in ChestX-Ray14","volume":"259","author":"Chehade","year":"2025","journal-title":"Comput Methods Progr Biomed"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/INMIC64792.2024.11004380","article-title":"Pneumonia detection in chest X-ray images: handling class imbalance","author":"Ali","year":"2024","journal-title":"2024 26th International Multi-Topic Conference (INMIC)"},{"key":"B10","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/ICMI60790.2024.10586129","article-title":"Advancing medical image diagnostics through multi-modal fusion: insights from mimic chest x-ray dataset analysis","author":"Upadhya","year":"2024","journal-title":"2024 IEEE 3rd International Conference on Computing and Machine Intelligence (ICMI)"},{"key":"B11","article-title":"Resnet in resnet: generalizing residual architectures","author":"Targ","year":"2016","journal-title":"arXiv preprint arXiv:1603.08029"},{"key":"B12","doi-asserted-by":"publisher","first-page":"9872","DOI":"10.1109\/ACCESS.2018.2890127","article-title":"Multiple feature reweight densenet for image classification","volume":"7","author":"Zhang","year":"2019","journal-title":"IEEE Access"},{"key":"B13","doi-asserted-by":"publisher","first-page":"31117","DOI":"10.1109\/ACCESS.2024.3369900","article-title":"Fast and efficient lung abnormality identification with explainable AI: a comprehensive framework for chest CT scan and X-ray images","volume":"12","author":"Hasan","year":"2024","journal-title":"IEEE Access"},{"key":"B14","doi-asserted-by":"publisher","first-page":"1","DOI":"10.5121\/ijdkp.2015.5201","article-title":"A review on evaluation metrics for data classification evaluations","volume":"5","author":"Hossin","year":"2015","journal-title":"Int J Data Min Knowl Manage Proc"},{"key":"B15","doi-asserted-by":"publisher","first-page":"6614016","DOI":"10.1155\/pm\/6614016","article-title":"A deep convolutional neural network model for lung disease detection using chest X-ray imaging","volume":"2025","author":"Dardouri","year":"2025","journal-title":"Pulm Med"},{"key":"B16","doi-asserted-by":"publisher","first-page":"10173","DOI":"10.1109\/TPAMI.2023.3250241","article-title":"Normalization techniques in training DNNS: methodology, analysis and application","volume":"45","author":"Huang","year":"2023","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"B17","doi-asserted-by":"publisher","first-page":"3853","DOI":"10.1007\/s13369-021-05347-7","article-title":"An improved hybrid approach for handling class imbalance problem","volume":"46","author":"Desuky","year":"2021","journal-title":"Arabian J Sci Eng"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1088\/1757-899X\/428\/1\/012022","article-title":"Weighted focal loss: an effective loss function to overcome unbalance problem of chest X-ray14","author":"Qin","year":"2018","journal-title":"IOP Conference Series: Materials Science and Engineering"},{"key":"B19","doi-asserted-by":"publisher","first-page":"2187","DOI":"10.1109\/JBHI.2024.3362243","article-title":"Tri-net with cross-balanced pseudo supervision for class imbalanced medical image classification","volume":"28","author":"Qu","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"B20","doi-asserted-by":"publisher","first-page":"2657","DOI":"10.1109\/36.975000","article-title":"Effect of spatial resolution on classification errors of pure and mixed pixels in remote sensing","volume":"39","author":"Hsieh","year":"2001","journal-title":"IEEE Trans Geosci Rem Sens"},{"key":"B21","doi-asserted-by":"publisher","first-page":"100499","DOI":"10.1016\/j.dajour.2024.100499","article-title":"An explainable artificial intelligence model for multiple lung diseases classification from chest X-ray images using fine-tuned transfer learning","volume":"12","author":"Mahamud","year":"2024","journal-title":"Dec Analyt J"},{"key":"B22","doi-asserted-by":"publisher","first-page":"534","DOI":"10.1038\/s41598-023-47038-3","article-title":"Auto-detection of the coronavirus disease by using deep convolutional neural networks and X-ray photographs","volume":"14","author":"Hussein","year":"2024","journal-title":"Sci Rep"},{"key":"B23","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1186\/s12880-023-01177-1","article-title":"Attention-enhanced architecture for improved pneumonia detection in chest X-ray images","volume":"24","author":"Li","year":"2024","journal-title":"BMC Med Imaging"},{"key":"B24","doi-asserted-by":"publisher","first-page":"111806","DOI":"10.1016\/j.asoc.2024.111806","article-title":"Automated detection of COVID-19 and pneumonia diseases using data mining and transfer learning algorithms with focal loss from chest X-ray images","volume":"162","author":"Khattab","year":"2024","journal-title":"Appl Soft Comput"},{"key":"B25","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/IT64745.2025.10930261","article-title":"Tuberculosis detection from chest X-rays using convolutional neural network","author":"Rai\u010devi\u0107","year":"2025","journal-title":"2025 29th International Conference on Information Technology (IT)"},{"key":"B26","doi-asserted-by":"publisher","first-page":"1109","DOI":"10.1049\/ipr2.13010","article-title":"Optimizing chest tuberculosis image classification with oversampling and transfer learning","volume":"18","author":"Alqahtani","year":"2024","journal-title":"IET Image Proc"},{"key":"B27","doi-asserted-by":"publisher","first-page":"2097","DOI":"10.1109\/CVPR.2017.369","article-title":"Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases","author":"Wang","year":"2017","journal-title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition"},{"key":"B28","doi-asserted-by":"publisher","first-page":"590","DOI":"10.1609\/aaai.v33i01.3301590","article-title":"CheXpert: a large chest radiograph dataset with uncertainty labels and expert comparison","author":"Irvin","year":"2019","journal-title":"Proceedings of the AAAI Conference on Artificial Intelligence"},{"key":"B29","doi-asserted-by":"publisher","first-page":"101797","DOI":"10.1016\/j.media.2020.101797","article-title":"PadChest: a large chest x-ray image dataset with multi-label annotated reports","volume":"66","author":"Bustos","year":"2020","journal-title":"Med Image Anal"},{"key":"B30","doi-asserted-by":"publisher","first-page":"317","DOI":"10.1038\/s41597-019-0322-0","article-title":"MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reports","volume":"6","author":"Johnson","year":"2019","journal-title":"Sci Data"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1795508\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,17]],"date-time":"2026-06-17T04:47:09Z","timestamp":1781671629000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1795508\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,17]]},"references-count":30,"alternative-id":["10.3389\/fmed.2026.1795508"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1795508","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,17]]},"article-number":"1795508"}}