{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T12:32:48Z","timestamp":1789129968708,"version":"build-2803163510"},"reference-count":28,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T00:00:00Z","timestamp":1789084800000},"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. Oncol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Accurate histopathological discrimination of lung adenocarcinoma (ACA) and squamous cell carcinoma (SCC) from benign tissue is a clinically demanding task in which the two malignant subtypes are frequently confused even by trained pathologists.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We present a dual-branch deep learning framework that couples a hierarchical shifted-window vision transformer (SwinV2-CR-Tiny) with a fully-convolutional masked-autoencoder-pretrained convolutional network (ConvNeXtV2-Tiny) to capture complementary global-contextual and local-textural representations of haematoxylin-and-eosin (H&amp;amp;E) stained tissue. The two branches are integrated through a learnable cross-feature gating module, refined by feature-space gated-modulation heads, and compressed by an Electric-Eel-Foraging-Optimizer (EEFO)-driven feature-selection stage. A Crested-Porcupine-Optimizer (CPO) calibrates class weighting and a Fennec-Fox-Optimizer (FFO) tunes per-class decision thresholds, while training combines Macenko stain normalization, CLAHE, MixUp\/CutMix augmentation, a focal loss with label smoothing, Sharpness-Aware Minimization (SAM) and Stochastic Weight Averaging (SWA). Predictive uncertainty is quantified by Monte-Carlo (MC) dropout and model decisions are explained with Grad-CAM++ and Integrated Gradients. The model was evaluated on a stratified 70\/15\/15 partition of the LC25000 lung cohort (15,000 images; 5,000 per class).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The proposed model attains a test accuracy of 94.04%, macro-F1 of 0.9404, macro one-vs-rest AUC of 0.9886 and Cohen's \u03ba of 0.9107, with near-perfect benign separation (sensitivity 0.999) and the residual error concentrated in the clinically expected ACA\u2194SCC confusion. Contrary to our prior expectation, MC-dropout predictive uncertainty did not separate misclassified from correctly classified cases on this cohort (mean across-pass standard deviation 0.0261 for errors versus 0.0269 for correct predictions). We additionally report a negative calibration finding: the model is systematically under-confident and naive temperature scaling increased the expected calibration error.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>The negative uncertainty result is attributed to the same confidence-suppressing training objective and is reported rather than omitted. We further discuss the data-provenance characteristics of LC25000 that constrain the interpretation of absolute accuracy. The framework offers an interpret-able, transparently-evaluated decision-support pipeline whose error profile aligns with established lung-cancer histomorphology.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fonc.2026.1907746","type":"journal-article","created":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T12:03:38Z","timestamp":1789128218000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["An explainable dual-branch transformer\u2013ConvNeXt framework for robust lung histopathology classification"],"prefix":"10.3389","volume":"16","author":[{"given":"G.","family":"Sunitha","sequence":"first","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J.","family":"Vellingiri","sequence":"additional","affiliation":[{"name":"School of Computer Science Engineering and Information Systems, Vellore Institute of Technology","place":["Vellore, India"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,11]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1912.12142","article-title":"Lung and colon cancer histopathological image dataset (LC25000)","author":"Borkowski","year":"2019"},{"key":"B2","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1016\/j.media.2017.07.005","article-title":"A survey on deep learning in medical image analysis","volume":"42","author":"Litjens","year":"2017","journal-title":"Med Img Anal"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2020.101813","article-title":"Deep neural network models for computational histopathology: A survey","volume":"67","author":"Srinidhi","year":"2021","journal-title":"Med Img Anal"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1101\/197574","article-title":"Classification and mutation prediction from non-small cell lung cancer histopathology images using deep learning","volume":"24","author":"Coudray","year":"2018","journal-title":"Nat Med"},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.1038\/nature21056","article-title":"Dermatologist-level classification of skin cancer with deep neural networks","volume":"542","author":"Esteva","year":"2017","journal-title":"Nature"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.3390\/s21030748","article-title":"A machine learning approach to diagnosing lung and colon cancer using a deep learning-based classification framework","volume":"21","author":"Masud","year":"2021","journal-title":"Sensors"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1109\/access.2022.3150924","article-title":"Malignancy detection in lung and colon histopathology images using transfer learning with class selective image processing","volume":"10","author":"Mehmood","year":"2022","journal-title":"IEEE Access"},{"key":"B8","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52688.2022.01170","article-title":"Swin Transformer V2: Scaling up capacity and resolution","author":"Liu","year":"2022"},{"key":"B9","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52729.2023.01548","article-title":"ConvNeXt V2: Co-designing and scaling ConvNets with masked autoencoders","author":"Woo","year":"2023"},{"key":"B10","doi-asserted-by":"crossref","DOI":"10.1109\/ISBI.2009.5193250","article-title":"A method for normalizing histology slides for quantitative analysis","author":"Macenko","year":"2009"},{"key":"B11","article-title":"mixup: Beyond empirical risk minimization","author":"Zhang","year":"2018"},{"key":"B12","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV.2019.00612","article-title":"CutMix: Regularization strategy to train strong classifiers with localizable features","author":"Yun","year":"2019"},{"key":"B13","article-title":"When does label smoothing help","author":"M\u00fcller","year":"2019"},{"key":"B14","article-title":"Sharpness-aware minimization for efficiently improving generalization","author":"Foret","year":"2021"},{"key":"B15","article-title":"Averaging weights leads to wider optima and better generalization","author":"Izmailov","year":"2018"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2018.2858826","article-title":"Focal loss for dense object detection","volume":"42","author":"Lin","year":"2020","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1016\/j.knosys.2023.111257","article-title":"Crested porcupine optimizer: A new nature-inspired metaheuristic","volume":"284","author":"Abdel-Basset","year":"2024","journal-title":"Knowledge-Based Syst"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2023.122200","article-title":"Electric eel foraging optimization: A new bio-inspired optimizer for engineering applications","volume":"238","author":"Zhao","year":"2024","journal-title":"Expert Syst Appl"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.1109\/access.2022.3197745","article-title":"Fennec fox optimization: A new nature-inspired optimization algorithm","volume":"10","author":"Trojovska","year":"2022","journal-title":"IEEE Access"},{"key":"B20","article-title":"Dropout as a Bayesian approximation: Representing model uncertainty in deep learning","author":"Gal","year":"2016"},{"key":"B21","article-title":"On calibration of modern neural networks","author":"Guo","year":"2017"},{"key":"B22","doi-asserted-by":"crossref","DOI":"10.1109\/WACV.2018.00097","article-title":"Grad-CAM++: Generalized gradient-based visual explanations for deep convolutional networks","author":"Chattopadhay","year":"2018"},{"key":"B23","article-title":"Axiomatic attribution for deep networks","author":"Sundararajan","year":"2017"},{"key":"B24","article-title":"A unified approach to interpreting model predictions","author":"Lundberg","year":"2017"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.1093\/oso\/9780199546329.003.0006","article-title":"Visualizing data using t-SNE","volume":"9","author":"van der Maaten","year":"2008","journal-title":"J Mach Learn Res"},{"key":"B26","doi-asserted-by":"publisher","DOI":"10.21105\/joss.00861","article-title":"UMAP: Uniform manifold approximation and projection for dimension reduction","volume":"3","author":"McInnes","year":"2018","journal-title":"J Open Source Software"},{"key":"B27","article-title":"An Image Is Worth 16 \u00d7 16 Words: Transformers for Image Recognition at Scale","author":"Dosovitskiy","year":"2021"},{"key":"B28","doi-asserted-by":"crossref","DOI":"10.1016\/B978-0-12-336156-1.50061-6","article-title":"Contrast limited adaptive histogram equalization","volume-title":"Graphics Gems IV","author":"Zuiderveld","year":"1994"}],"container-title":["Frontiers in Oncology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1907746\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,11]],"date-time":"2026-09-11T12:03:39Z","timestamp":1789128219000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fonc.2026.1907746\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,11]]},"references-count":28,"alternative-id":["10.3389\/fonc.2026.1907746"],"URL":"https:\/\/doi.org\/10.3389\/fonc.2026.1907746","relation":{},"ISSN":["2234-943X"],"issn-type":[{"value":"2234-943X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,11]]},"article-number":"1907746"}}