{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T04:58:46Z","timestamp":1790657926178,"version":"4.1.0"},"reference-count":44,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T00:00:00Z","timestamp":1790640000000},"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>Tongue diagnosis is a vital component of Traditional Chinese Medicine (TCM), with cracked tongue serving as a key diagnostic indicator. Although deep learning techniques have advanced TCM tongue image recognition, existing methods for cracked tongue detection still face challenges of low accuracy and slow processing speed.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>To address these issues, we propose an improved YOLOv5 framework integrating Switchable Atrous Convolution (SAC) and a Transformer encoder. First, given the varied morphology, size, and distribution of tongue cracks, SAC is introduced into the backbone network to adaptively adjust receptive fields via different atrous rates, enabling multi-scale feature extraction and better preservation of fine details. Second, a Transformer encoder module is incorporated to capture global information and long-range dependencies, thereby enhancing the network's overall feature representation capability. Experiments are conducted on a self-built tongue image dataset to evaluate the proposed modifications.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The improved YOLOv5 demonstrates superior performance over the original YOLOv5 in tongue crack detection, achieving an increase of 2.2% in AP@50 and 8.43% in AP@50:95 metrics on the self-built dataset.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These results confirm that the combination of SAC and Transformer effectively boosts detection accuracy and generalizability, addressing the limitations of prior approaches. The proposed method offers a more reliable and efficient solution for automated tongue crack analysis, with potential implications for clinical TCM diagnostic support.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1901533","type":"journal-article","created":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T04:40:19Z","timestamp":1790656819000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Cracks-YOLO: an improved YOLOv5 tongue crack detection method based on switchable atrous convolution and transformer"],"prefix":"10.3389","volume":"13","author":[{"given":"Miaomiao","family":"Ding","sequence":"first","affiliation":[{"name":"School of Medical Information Engineering, Guangzhou University of Chinese Medicine","place":["Guangzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Meiyi","family":"Wu","sequence":"additional","affiliation":[{"name":"School of Business, City University of Hong Kong","place":["Kowloon, Hong Kong SAR, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bochen","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Medical Information Engineering, Guangzhou University of Chinese Medicine","place":["Guangzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Huangbo","family":"Lin","sequence":"additional","affiliation":[{"name":"School of Medical Information Engineering, Guangzhou University of Chinese Medicine","place":["Guangzhou, China"]},{"name":"School of Biomedical Engineering, Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shaoyang","family":"Men","sequence":"additional","affiliation":[{"name":"School of Medical Information Engineering, Guangzhou University of Chinese Medicine","place":["Guangzhou, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,29]]},"reference":[{"key":"B1","volume-title":"Advanced Diagnostics of Chinese Medicine Chinese Medicine Books","author":"Zhu","year":"2004"},{"key":"B2","doi-asserted-by":"publisher","first-page":"130526","DOI":"10.1016\/j.eswa.2025.130526","article-title":"Metalink: an imbalanced multi-label classification approach for automatic tongue diagnosis","volume":"302","author":"Lin","year":"2025","journal-title":"Expert Syst Appl"},{"key":"B3","first-page":"126","article-title":"Epidemic survey on TCM syndrome typing and preliminary discussion on TCM etiology and pathogenesis of fatty liver","volume":"25","author":"Wang","year":"2005","journal-title":"Chin J Integr Med"},{"key":"B4","first-page":"13","article-title":"Distribution of TCM syndromes in 213 cases of diabetic nephropathy","volume":"28","author":"Wu","year":"2009","journal-title":"Beijing J Tradit Chin Med"},{"key":"B5","first-page":"9","article-title":"Correlation between objectified tongue characteristics and yin deficiency pattern in hypertension patients","volume":"28","author":"Duan","year":"2021","journal-title":"Mod Chin Clin Med"},{"key":"B6","first-page":"381","article-title":"Analysis of chinese medical characteristics of 35 patients with novel coronavirus pneumonia","volume":"29","author":"Huang","year":"2020","journal-title":"J Emerg Tradit Chin Med"},{"key":"B7","first-page":"1316","article-title":"Analysis of tongue diagnostic parameters in patients with coronary heart disease and its different complications","volume":"37","author":"Wang","year":"2022","journal-title":"China J Tradit Chin Med Pharm"},{"key":"B8","doi-asserted-by":"publisher","first-page":"112386","DOI":"10.1016\/j.patcog.2025.112386","article-title":"A boundary-enhanced and target-driven deformable convolutional network for abdominal multi-organ segmentation","volume":"172","author":"Ju","year":"2026","journal-title":"Pattern Recognit"},{"key":"B9","doi-asserted-by":"publisher","first-page":"103509","DOI":"10.1016\/j.inffus.2025.103509","article-title":"Collaborative learning of dynamic and static information for medical image segmentation","volume":"125","author":"Ju","year":"2026","journal-title":"Inform Fusion"},{"key":"B10","first-page":"2849","article-title":"Progress in clinical application of tongue inspection objectivity based on digital image processing technique","volume":"30","author":"Qi","year":"2015","journal-title":"China J Tradit Chin Med Pharm"},{"key":"B11","first-page":"5","article-title":"Discussion on tongue diagnosis objectification using computer image recognition technology","volume":"5","author":"Sun","year":"1986","journal-title":"J Anhui Univ Chin Med"},{"key":"B12","first-page":"336","article-title":"Study on method of automatic diagnosis of tongue feature in traditional chinese medicine","volume":"13","author":"Yu","year":"1994","journal-title":"Chin J Biomed Eng"},{"key":"B13","first-page":"132","article-title":"Towards the standardization of tongue diagnosis: an image processing approach","volume":"20","author":"Zhu","year":"2001","journal-title":"Chin J Biomed Eng"},{"key":"B14","first-page":"42","article-title":"Identification of tongue body and fur based on color image region separation","volume":"23","author":"Xu","year":"2009","journal-title":"J Shanghai Univ Tradit Chin Med"},{"key":"B15","author":"Srividhya","year":""},{"key":"B16","doi-asserted-by":"publisher","first-page":"107231","DOI":"10.1016\/j.bspc.2024.107231","article-title":"Improving small intestinal stromal tumor detection using 3D context and auxiliary priori cues","volume":"102","author":"Ju","year":"2025","journal-title":"Biomed Signal Process Control"},{"key":"B17","first-page":"303","article-title":"Review of deep learning in classification of tongue image","volume":"17","author":"Wu","year":"2023","journal-title":"J Front Comput Sci Technol"},{"key":"B18","first-page":"148","article-title":"Tongue image analysis in traditional chinese medicine based on deep learning","volume":"47","author":"Li","year":"2020","journal-title":"Comput Sci"},{"key":"B19","author":"Xue","year":""},{"key":"B20","author":"Chen","year":""},{"key":"B21","author":"Tang","year":""},{"key":"B22","doi-asserted-by":"publisher","first-page":"e6262","DOI":"10.1002\/cpe.6262","article-title":"A weakly supervised tooth mark and crack detection method in tongue image","volume":"33","author":"Weng","year":"2021","journal-title":"Concurr Comput Pract Exp"},{"key":"B23","author":"Jocher","year":""},{"key":"B24","author":"Redmon","year":""},{"key":"B25","author":"Wang","year":""},{"key":"B26","doi-asserted-by":"publisher","first-page":"834","DOI":"10.1109\/TPAMI.2017.2699184","article-title":"DeepLab: Semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs","volume":"40","author":"Chen","year":"2018","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"B27","author":"Yu","year":""},{"key":"B28","first-page":"259","article-title":"Building deep neural networks with dilated convolutions to reconstruct high-resolution image","volume":"32","author":"Zhang","year":"2019","journal-title":"Concurr Comput Pract Exp"},{"key":"B29","first-page":"497","article-title":"Multi-scale convolution neural network model based on attention mechanism","volume":"34","author":"Wang","year":"2021","journal-title":"Concurr Comput Pract Exp"},{"key":"B30","author":"Qiao","year":""},{"key":"B31","author":"He","year":""},{"key":"B32","author":"Ioffe","year":""},{"key":"B33","doi-asserted-by":"publisher","first-page":"3","DOI":"10.1016\/j.neunet.2017.12.012","article-title":"Sigmoid-weighted linear units for neural network function approximation in reinforcement learning","volume":"107","author":"Elfwing","year":"2018","journal-title":"Neural Netw"},{"key":"B34","author":"Dosovitskiy","year":""},{"key":"B35","author":"Vaswani","year":""},{"key":"B36","author":"Liu","year":""},{"key":"B37","author":"Bochkovskiy","year":""},{"key":"B38","author":"Srinivas","year":""},{"key":"B39","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume":"139","author":"Touvron","year":"2021","journal-title":"Proc Int Conf Mach Learn"},{"key":"B40","author":"Lin","year":""},{"key":"B41","author":"Bolya","year":""},{"key":"B42","author":"Selvaraju","year":""},{"key":"B43","doi-asserted-by":"publisher","first-page":"5670","DOI":"10.1109\/jbhi.2024.3523219","article-title":"Learning from target-level incomplete annotation: a novel perspective for weakly-supervised multi-lesion segmentation","volume":"30","author":"Ju","year":"2024","journal-title":"IEEE J Biomed Health Inform"},{"key":"B44","doi-asserted-by":"publisher","first-page":"7469","DOI":"10.1109\/JBHI.2024.3440925","article-title":"CDI-NSTSEG: a clinical diagnosis-inspired effective and efficient framework for non-salient small tumor segmentation","volume":"28","author":"Ju","year":"2024","journal-title":"IEEE J Biomed Health Inform"}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1901533\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,29]],"date-time":"2026-09-29T04:40:39Z","timestamp":1790656839000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1901533\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,29]]},"references-count":44,"alternative-id":["10.3389\/fmed.2026.1901533"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1901533","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,29]]},"article-number":"1901533"}}