{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T13:00:48Z","timestamp":1791464448302,"version":"4.3.4"},"reference-count":43,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2026,1,7]],"date-time":"2026-01-07T00:00:00Z","timestamp":1767744000000},"content-version":"vor","delay-in-days":6,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["12405382"],"award-info":[{"award-number":["12405382"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["aapm.onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Medical Physics"],"published-print":{"date-parts":[[2026,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Medical image analysis has witnessed substantial advancements through recent deep learning (DL) algorithms development. Vision Transformers (ViTs) have emerged as a powerful alternative solution by leveraging self\u2010attention to model both local and global interactions. Despite their promise, ViTs are data\u2010intensive and lack inductive biases, limiting their utility in medical imaging. Conversely, radiomics offers domain\u2010specific, interpretable descriptors of image heterogeneity but lacks scalability and integration with deep learning. This study proposes a unified Radiomics\u2010Embedded Vision Transformer (RE\u2010ViT) framework that combines handcrafted radiomic features and data\u2010driven visual embeddings within a ViT architecture.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Purpose<\/jats:title>\n                    <jats:p>To develop and evaluate a RE\u2010ViT framework that integrates radiomics and patch\u2010wise ViT embeddings to improve feature representation for medical image classification across heterogeneous datasets.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Following the classic ViT design, the input image was first resampled into multiple image patches. For each image patch, handcrafted radiomic features, including intensity, texture, and spatial heterogeneity descriptors, were extracted. Simultaneously, standard patch embeddings were obtained via linear projection of pixel intensities. The two embeddings were averaged, normalized, and combined with positional encodings before being tokenized and processed by a ViT encoder. A learnable token aggregates patch\u2010level information for final classification. The model was evaluated on three publicly available datasets, BUSI (lesion malignancy diagnosis through breast ultrasound), ChestXray2017 (lung pneumonitis diagnosis through chest x\u2010ray), and Retinal OCT (retina disease diagnosis through retinal OCT), using 10\u2010fold cross\u2010validation. Performance metrics included accuracy, macro area under the ROC curve (AUC), sensitivity, and specificity. Ablation studies were implemented to assess the contribution of RE\u2010ViT architectural components on these three clinical problems. Comparative analyses were also conducted against CNN (VGG\u201016, ResNet) and hybrid (TransMed) models.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The proposed RE\u2010ViT model demonstrated consistently robust classification performance across all three medical imaging datasets. In BUSI, RE\u2010ViT achieved an accuracy of 0.848\u00a0\u00b1\u00a00.027, AUC of 0.950\u00a0\u00b1\u00a00.011, sensitivity of 0.796\u00a0\u00b1\u00a00.042, and specificity of 0.905\u00a0\u00b1\u00a00.020. In ChestXray2017, it yielded an accuracy of 0.950\u00a0\u00b1\u00a00.012, AUC of 0.989\u00a0\u00b1\u00a00.004, sensitivity of 0.953\u00a0\u00b1\u00a00.010, and specificity of 0.975\u00a0\u00b1\u00a00.005. In Retinal OCT, RE\u2010ViT achieved an accuracy of 0.938\u00a0\u00b1\u00a00.001, AUC of 0.986\u00a0\u00b1\u00a00.001, sensitivity of 0.914\u00a0\u00b1\u00a00.023, and specificity of 0.969\u00a0\u00b1\u00a00.024. In the comparison studies, the RE\u2010ViT matches or outperforms alternatives. Ablation revealed significant performance drops when removing either radiomics or projection\u2010based embeddings. Attention map visualizations demonstrated imaging modality\u2010specific utilization of radiomics and learned features, with improved localization of clinically relevant regions.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>The proposed radiomics\u2010embedded vision transformer was developed for multiple image classification tasks. Current results underscore the potential of our approach to advance other transformer\u2010based medical image classification tasks.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1002\/mp.70246","type":"journal-article","created":{"date-parts":[[2026,1,7]],"date-time":"2026-01-07T18:13:46Z","timestamp":1767809626000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":4,"title":["An exploratory study on integrating radiomics with vision transformers for enhancing medical imaging classification accuracy"],"prefix":"10.1002","volume":"53","author":[{"given":"Zhenyu","family":"Yang","sequence":"first","affiliation":[{"name":"Medical Physics Graduate Program Duke Kunshan University  Kunshan China"},{"name":"Jiangsu Provincial University Key (Construction) Laboratory for Smart Diagnosis and Treatment of Lung Cancer Duke Kunshan University Kunshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Rihui","family":"Zhang","sequence":"additional","affiliation":[{"name":"Medical Physics Graduate Program Duke Kunshan University  Kunshan China"},{"name":"Jiangsu Provincial University Key (Construction) Laboratory for Smart Diagnosis and Treatment of Lung Cancer Duke Kunshan University Kunshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haiming","family":"Zhu","sequence":"additional","affiliation":[{"name":"Medical Physics Graduate Program Duke Kunshan University  Kunshan China"},{"name":"Jiangsu Provincial University Key (Construction) Laboratory for Smart Diagnosis and Treatment of Lung Cancer Duke Kunshan University Kunshan China"},{"name":"Department of Radiotherapy and Oncology The First People's Hospital of Kunshan  Kunshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Haipeng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Medical Physics Graduate Program Duke Kunshan University  Kunshan China"},{"name":"Jiangsu Provincial University Key (Construction) Laboratory for Smart Diagnosis and Treatment of Lung Cancer Duke Kunshan University Kunshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jianliang","family":"Wang","sequence":"additional","affiliation":[{"name":"Jiangsu Provincial University Key (Construction) Laboratory for Smart Diagnosis and Treatment of Lung Cancer Duke Kunshan University Kunshan China"},{"name":"Department of Radiology The First People's Hospital of Kunshan  Kunshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Minbin","family":"Chen","sequence":"additional","affiliation":[{"name":"Jiangsu Provincial University Key (Construction) Laboratory for Smart Diagnosis and Treatment of Lung Cancer Duke Kunshan University Kunshan China"},{"name":"Department of Radiotherapy and Oncology The First People's Hospital of Kunshan  Kunshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fang\u2010Fang","family":"Yin","sequence":"additional","affiliation":[{"name":"Medical Physics Graduate Program Duke Kunshan University  Kunshan China"},{"name":"Jiangsu Provincial University Key (Construction) Laboratory for Smart Diagnosis and Treatment of Lung Cancer Duke Kunshan University Kunshan China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chunhao","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Radiation Oncology Duke University  Durham USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2026,1,7]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2020.04.157"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.1007\/s12194\u2010017\u20100406\u20105"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.eswa.2020.114095"},{"key":"e_1_2_9_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2016.2528162"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2018.02.010"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10916\u2010024\u201002105\u20108"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.media.2023.102802"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/3505244"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.1186\/s13244\u2010020\u201000887\u20102"},{"key":"e_1_2_9_11_1","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.2015151169"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.semcancer.2020.04.002"},{"key":"e_1_2_9_13_1","doi-asserted-by":"publisher","DOI":"10.1007\/s10549\u2010018\u20104675\u20104"},{"key":"e_1_2_9_14_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.lungcan.2017.10.015"},{"key":"e_1_2_9_15_1","doi-asserted-by":"publisher","DOI":"10.21037\/tcr.2016.06.18"},{"key":"e_1_2_9_16_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00330\u2010024\u201010911\u20104"},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.3390\/cancers17050768"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2023.1185771"},{"key":"e_1_2_9_19_1","doi-asserted-by":"publisher","DOI":"10.3389\/fonc.2024.1419621"},{"key":"e_1_2_9_20_1","first-page":"144","volume-title":"Image Processing","author":"Qiu J","year":"2024"},{"key":"e_1_2_9_21_1","doi-asserted-by":"publisher","DOI":"10.3390\/jcm11113013"},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2022.3217218"},{"key":"e_1_2_9_23_1","doi-asserted-by":"publisher","DOI":"10.21037\/qims\u201022\u2010531"},{"key":"e_1_2_9_24_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijrobp.2021.02.004"},{"key":"e_1_2_9_25_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.dib.2019.104863"},{"key":"e_1_2_9_26_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jksuci.2023.01.014"},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41597\u2010022\u201001721\u20108"},{"key":"e_1_2_9_28_1","unstructured":"DosovitskiyA et\u00a0al. \u201cAn image is worth 16\u00d716 words: transformers for image recognition at scale \u201d ArXiv Prepr. ArXiv201011929 2020."},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.1002\/mp.15837"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1088\/1361-6560\/acf10d"},{"key":"e_1_2_9_31_1","unstructured":"HuZ YangZ LafataKJ YinF\u2010F WangC \u201cA radiomics\u2010boosted deep\u2010learning model for COVID\u201019 and non\u2010COVID\u201019 pneumonia classification using chest x\u2010ray image\u201d."},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvcir.2022.103664"},{"key":"e_1_2_9_33_1","doi-asserted-by":"crossref","unstructured":"YuanL et\u00a0al. \u201cTokens\u2010to\u2010token vit: training vision transformers from scratch on imagenet \u201d presented at theProceedings of the IEEE\/CVF international conference on computer vision 2021 pp.558\u2010567.","DOI":"10.1109\/ICCV48922.2021.00060"},{"key":"e_1_2_9_34_1","unstructured":"RidnikT Ben\u2010BaruchE NoyA Zelnik\u2010ManorL \u201cImagenet\u201021k pretraining for the masses \u201d ArXiv Prepr. ArXiv210410972 2021."},{"key":"e_1_2_9_35_1","unstructured":"AbnarS ZuidemaW \u201cQuantifying attention flow in transformers \u201d ArXiv Prepr. ArXiv200500928 2020. doi:10.18653\/v1\/2020.acl\u2010main.385"},{"key":"e_1_2_9_36_1","doi-asserted-by":"crossref","unstructured":"CheferH GurS WolfL \u201cTransformer interpretability beyond attention visualization \u201d presented at theProceedings of the IEEE\/CVF conference on computer vision and pattern recognition 2021 pp.782\u2010791.","DOI":"10.1109\/CVPR46437.2021.00084"},{"key":"e_1_2_9_37_1","doi-asserted-by":"crossref","unstructured":"MascarenhasS AgarwalM \u201cA comparison between VGG16 VGG19 and ResNet50 architecture frameworks for Image Classification \u201d presented at the2021 International conference on disruptive technologies for multi\u2010disciplinary research and applications (CENTCON) IEEE 2021 pp.96\u201099. doi:10.1109\/CENTCON52345.2021.9687944","DOI":"10.1109\/CENTCON52345.2021.9687944"},{"key":"e_1_2_9_38_1","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics11081384"},{"key":"e_1_2_9_39_1","first-page":"272","volume-title":"International MICCAI brainlesion workshop","author":"Hatamizadeh A","year":"2021"},{"key":"e_1_2_9_40_1","first-page":"36314","article-title":"Medformer: a multi\u2010granularity patching transformer for medical time\u2010series classification","volume":"37","author":"Wang Y","year":"2024","journal-title":"Adv Neural Inf Process Syst"},{"key":"e_1_2_9_41_1","unstructured":"RyooMS PiergiovanniA ArnabA DehghaniM AngelovaA \u201cTokenlearner: what can 8 learned tokens do for images and videos? \u201d ArXiv Prepr. ArXiv210611297 2021."},{"key":"e_1_2_9_42_1","doi-asserted-by":"crossref","unstructured":"ZhangY LiuH HuQ \u201cTransfuse: fusing transformers and cnns for medical image segmentation \u201d presented at theInternational conference on medical image computing and computer\u2010assisted intervention Springer 2021 pp.14\u201024.","DOI":"10.1007\/978-3-030-87193-2_2"},{"key":"e_1_2_9_43_1","first-page":"2025","article-title":"The shapley value","volume":"3","author":"Winter E","year":"2002","journal-title":"Handb Game Theory Econ Appl"},{"key":"e_1_2_9_44_1","unstructured":"KimB WattenbergM GilmerJ CaiC WexlerJ ViegasF.Interpretability beyond feature attribution: quantitative testing with concept activation vectors (tcav). presented at theInternational conference on machine learning PMLR 2018 pp.2668\u20102677."}],"container-title":["Medical Physics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70246","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/full-xml\/10.1002\/mp.70246","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/pdf\/10.1002\/mp.70246","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,19]],"date-time":"2026-03-19T01:57:04Z","timestamp":1773885424000},"score":1,"resource":{"primary":{"URL":"https:\/\/aapm.onlinelibrary.wiley.com\/doi\/10.1002\/mp.70246"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1]]},"references-count":43,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1]]}},"alternative-id":["10.1002\/mp.70246"],"URL":"https:\/\/doi.org\/10.1002\/mp.70246","archive":["Portico"],"relation":{},"ISSN":["0094-2405","2473-4209"],"issn-type":[{"value":"0094-2405","type":"print"},{"value":"2473-4209","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1]]},"assertion":[{"value":"2025-04-22","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-12-15","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-01-07","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70246"}}