{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T05:02:20Z","timestamp":1784869340496,"version":"3.55.0"},"reference-count":57,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T00:00:00Z","timestamp":1784851200000},"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>Objective<\/jats:title>\n                    <jats:p>\n                      To investigate the performance and clinical application potential of an improved YOLOv5-based deep learning model for automated detection and classification of pulmonary adenocarcinoma\n                      <jats:italic>in situ<\/jats:italic>\n                      (AIS) and minimally invasive adenocarcinoma (MIA), with particular emphasis on attention mechanisms and multiscale feature fusion for small pulmonary nodule detection.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>CT images of 225 pathologically confirmed AIS or MIA lesions treated at the Affiliated People's Hospital of Fujian University of Traditional Chinese Medicine between December 2019 and June 2025 were retrospectively collected. In the deep learning workflow, the 225 lesions were first divided at the lesion level: 27 lesions (11 MIA and 16 AIS) were set aside, and the 218 original PNG images derived from these lesions were reserved as an independent test set that was excluded from all data augmentation and model training. The remaining lesions were converted into PNG format and further augmented using horizontal flipping, yielding a final dataset of 3,000 images for training and validation. Subsequently, the dataset was partitioned into training and validation sets at an 8:2 ratio. During this process, the origins of the augmented lesion images were manually verified to ensure that an original image and its corresponding augmented variants were not assigned to different subsets simultaneously. YOLOv5s was used as the baseline model. Local contrast-spatial attention modules (LCSA and LCSAv2), BiFPN, and ASFF were incorporated for attention enhancement and multiscale feature fusion. Five-fold cross-validation was performed. Mean average precision (mAP), precision, and recall were evaluated across IoU thresholds from 0.6 to 0.9, and ablation experiments were conducted to quantify the contribution of each module.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In the baseline model selection stage, comparison under identical conditions showed that YOLOv5s achieved detection performance comparable to YOLOv11s while reducing GFLOPs by approximately 25%; therefore, YOLOv5s was selected as the improvement baseline. In five-fold cross-validation at IoU = 0.6, baseline YOLOv5s achieved an mAP@50 (AP at the matching IoU of 0.6) of 0.947, precision of 0.849, and recall of 0.932. After LCSA x ASFF was introduced, recall increased to 0.941 and mAP@50 was maintained at 0.946. Under the stringent localization criterion of IoU = 0.9, LCSA \u00d7 ASFF achieved an mAP@50 of 0.878, a 3.8-percentage-point higher value than the baseline (0.846); precision (0.879) and recall (0.773) were also the highest among all models, with the smallest performance attenuation in the high-IoU range. Ablation experiments showed that the combined use of LCSA and ASFF produced stronger synergistic effects in the high-IoU range than either module alone, whereas LCSAv2 combined with ASFF did not show the expected synergistic gain. On the independent test set, baseline YOLOv5s achieved precision, recall, and mAP@50 values of 0.747, 0.715, and 0.753, respectively; the improved LCSA \u00d7 ASFF model achieved corresponding values of 0.693, 0.753, and 0.788. Recall and mAP@50 increased simultaneously, indicating better overall detection performance than the baseline. The improved YOLOv5s-based deep learning model performed better in automated detection, and the LCSA \u00d7 ASFF combination improved localization accuracy while maintaining recall through the synergy between local contrast enhancement and adaptive scale fusion.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The improved YOLOv5s model achieved automated detection and classification of small pulmonary nodules through end-to-end training, showing clear advantages in clinical workflow automation without manual intervention. The synergistic effect of local contrast enhancement and adaptive scale fusion effectively improved AIS\/MIA localization precision and classification accuracy, supporting its potential as an assistive tool for early lung adenocarcinoma screening, pending external multi-centre validation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmed.2026.1885660","type":"journal-article","created":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T04:24:31Z","timestamp":1784867071000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Application of an improved YOLOv5s-based deep learning model for automated detection of pulmonary adenocarcinoma in situ and minimally invasive adenocarcinoma"],"prefix":"10.3389","volume":"13","author":[{"given":"Zhipeng","family":"Sun","sequence":"first","affiliation":[{"name":"The First Clinical Medical College, The Affiliated People's Hospital of Fujian University of Traditional Chinese Medicine","place":["Fuzhou, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinghui","family":"Chen","sequence":"additional","affiliation":[{"name":"The First Clinical Medical College, The Affiliated People's Hospital of Fujian University of Traditional Chinese Medicine","place":["Fuzhou, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lianxin","family":"Xie","sequence":"additional","affiliation":[{"name":"The First Clinical Medical College, The Affiliated People's Hospital of Fujian University of Traditional Chinese Medicine","place":["Fuzhou, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tao","family":"Yang","sequence":"additional","affiliation":[{"name":"The First Clinical Medical College, The Affiliated People's Hospital of Fujian University of Traditional Chinese Medicine","place":["Fuzhou, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lanlan","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Medical Imaging, Chongqing Eye and Optometry Ophthalmology Hospital","place":["Chongqing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chengbin","family":"Ye","sequence":"additional","affiliation":[{"name":"Department of Medical Imaging, The Affiliated People's Hospital of Fujian University of Traditional Chinese Medicine","place":["Fuzhou, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongjia","family":"Zhao","sequence":"additional","affiliation":[{"name":"The Affiliated People's Hospital of Fujian University of Traditional Chinese Medicine","place":["Fuzhou, Fujian, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,24]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"624","DOI":"10.1038\/s41571-023-00798-3","article-title":"The global burden of lung cancer: current status and future trends","volume":"20","author":"Leiter","year":"2023","journal-title":"Nat Rev Clin Oncol."},{"key":"B2","doi-asserted-by":"publisher","first-page":"362","DOI":"10.1016\/j.jtho.2021.11.003","article-title":"The 2021 WHO Classification of Lung Tumors: impact of advances since 2015","volume":"17","author":"Nicholson","year":"2022","journal-title":"J Thorac Oncol."},{"key":"B3","doi-asserted-by":"publisher","first-page":"142","DOI":"10.1016\/j.canlet.2019.02.047","article-title":"Lung cancer survival among never smokers","volume":"451","author":"Casal-Mouri\u00f1o","year":"2019","journal-title":"Cancer Lett."},{"key":"B4","doi-asserted-by":"publisher","first-page":"113","DOI":"10.1007\/s11604-017-0711-2","article-title":"HRCT texture analysis for pure or part-solid ground-glass nodules: distinguishability of adenocarcinoma in situ or minimally invasive adenocarcinoma from invasive adenocarcinoma","volume":"36","author":"Yagi","year":"2018","journal-title":"Jpn J Radiol."},{"key":"B5","doi-asserted-by":"publisher","first-page":"407","DOI":"10.1016\/j.athoracsur.2016.02.074","article-title":"Neither maximum tumor size nor solid component size is prognostic in part-solid lung cancer: impact of tumor size should be applied exclusively to solid lung cancer","volume":"102","author":"Hattori","year":"2016","journal-title":"Ann Thorac Surg."},{"key":"B6","doi-asserted-by":"publisher","first-page":"236","DOI":"10.1016\/j.lungcan.2014.02.008","article-title":"High resolution CT in differentiating minimally invasive component in early lung adenocarcinoma","volume":"84","author":"Zhang","year":"2014","journal-title":"Lung Cancer."},{"key":"B7","doi-asserted-by":"publisher","first-page":"20210768","DOI":"10.1259\/bjr.20210768","article-title":"Radiomics for differentiating minimally invasive adenocarcinoma from precursor lesions in pure ground-glass opacities on chest computed tomography","volume":"95","author":"Zhu","year":"2022","journal-title":"Br J Radiol."},{"key":"B8","doi-asserted-by":"publisher","first-page":"1015","DOI":"10.1111\/1759-7714.13838","article-title":"Can computed tomography differentiate adenocarcinoma in situ from minimally invasive adenocarcinoma?","volume":"12","author":"Yoshida","year":"2021","journal-title":"Thorac Cancer."},{"key":"B9","doi-asserted-by":"publisher","first-page":"e7140","DOI":"10.1002\/cam4.7140","article-title":"Artificial intelligence in lung cancer screening: detection, classification, prediction, and prognosis","volume":"13","author":"Quanyang","year":"2024","journal-title":"Cancer Med."},{"key":"B10","doi-asserted-by":"publisher","first-page":"759010","DOI":"10.1042\/BSR20212416","article-title":"Identification of pathological subtypes of early lung adenocarcinoma based on artificial intelligence parameters and CT signs","volume":"12","author":"Fang","year":"2022","journal-title":"Front Oncol."},{"key":"B11","doi-asserted-by":"publisher","first-page":"298","DOI":"10.3390\/diagnostics12020298","article-title":"Deep learning applications in computed tomography images for pulmonary nodule detection and diagnosis: a review","volume":"12","author":"Li","year":"2022","journal-title":"Diagnostics."},{"key":"B12","doi-asserted-by":"publisher","first-page":"30965","DOI":"10.1007\/s11042-023-16864-y","article-title":"Early detection and diagnosis of lung cancer using YOLO v7, and transfer learning","volume":"83","author":"Mammeri","year":"2024","journal-title":"Multimed Tools Appl."},{"key":"B13","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.media.2017.06.015","article-title":"Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge","volume":"42","author":"Setio","year":"2017","journal-title":"Med Image Anal."},{"key":"B14","doi-asserted-by":"publisher","first-page":"1013409","DOI":"10.1117\/12.2255795","article-title":"3D convolutional neural network for automatic detection of lung nodules in chest CT","volume":"10134","author":"Hamidian","year":"2017","journal-title":"Proc SPIE Int Soc Opt Eng"},{"key":"B15","doi-asserted-by":"publisher","first-page":"6881","DOI":"10.1158\/0008-5472.CAN-18-0696","article-title":"3D deep learning from CT scans predicts tumor invasiveness of subcentimeter pulmonary adenocarcinomas","volume":"78","author":"Zhao","year":"2018","journal-title":"Cancer Res."},{"key":"B16","doi-asserted-by":"publisher","first-page":"101814","DOI":"10.1016\/j.compmedimag.2020.101814","article-title":"3D deep learning based classification of pulmonary ground-glass opacity nodules with automatic segmentation","volume":"88","author":"Wang","year":"2021","journal-title":"Comput Med Imaging Graph."},{"key":"B17","doi-asserted-by":"publisher","first-page":"e2329674","DOI":"10.2214\/AJR.23.29674","article-title":"Lung-PNet: an automated deep learning model for the diagnosis of invasive adenocarcinoma in pure ground-glass nodules on chest CT","volume":"222","author":"Qi","year":"2024","journal-title":"AJR Am J Roentgenol."},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1007\/s44196-025-01069-y","article-title":"Optimizing deep learning models for clinical lung cancer detection: comparative analysis of segmentation-enhanced CNN-ensemble approaches in chest CT screening","author":"Hosseini","year":"2025","journal-title":"Int J Comput Intell Syst."},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2507.09898","article-title":"Advanced U-Net architectures with CNN backbones for automated lung cancer detection and segmentation in chest CT images","author":"Golkarieh","year":"2025","journal-title":"arXiv."},{"key":"B20","first-page":"779","article-title":"You only look once: unified, real-time object detection","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition.","author":"Redmon","year":"2016"},{"key":"B21","first-page":"7263","article-title":"YOLO9000: better, faster, stronger","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition.","author":"Redmon","year":"2017"},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1804.02767","article-title":"YOLOv3: an incremental improvement","author":"Redmon","year":"2018","journal-title":"arXiv"},{"key":"B23","doi-asserted-by":"crossref","DOI":"10.1109\/CVPRW50498.2020.00203","article-title":"CSPNet: a new backbone that can enhance learning capability of CNN","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition workshops.","author":"Wang","year":"2020"},{"key":"B24","first-page":"8759","article-title":"Path aggregation network for instance segmentation","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition.","author":"Liu","year":"2018"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2004.10934","article-title":"YOLOv4: optimal speed and accuracy of object detection","author":"Bochkovskiy","year":"2020","journal-title":"arXiv"},{"key":"B26","first-page":"101620","article-title":"Self-guided, multi-scale feature fusion improves small lung nodule detection on CT images","volume":"60","author":"Sun","year":"2020","journal-title":"Med Image Anal."},{"key":"B27","doi-asserted-by":"publisher","first-page":"588","DOI":"10.1007\/978-3-319-19992-4_46","article-title":"Multi-scale CNNs for lung nodule classification","volume":"10265","author":"Shen","year":"2017","journal-title":"Inf Process Med Imaging."},{"key":"B28","first-page":"10781","article-title":"EfficientDet: scalable and efficient object detection","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition.","author":"Tan","year":"2020"},{"key":"B29","doi-asserted-by":"publisher","first-page":"101772","DOI":"10.1016\/j.media.2020.101772","article-title":"MSCS-DeepLN: evaluating lung nodule malignancy using multi-scale cost-sensitive neural networks","volume":"65","author":"Xu","year":"2020","journal-title":"Med Image Anal."},{"key":"B30","doi-asserted-by":"publisher","first-page":"331","DOI":"10.1007\/s41095-022-0271-y","article-title":"Attention mechanisms in computer vision: a survey","volume":"8","author":"Guo","year":"2022","journal-title":"Comput Vis Media."},{"key":"B31","doi-asserted-by":"publisher","first-page":"2011","DOI":"10.1109\/TPAMI.2019.2913372","article-title":"Squeeze-and-excitation networks","volume":"42","author":"Hu","year":"2020","journal-title":"IEEE Trans Pattern Anal Mach Intell."},{"key":"B32","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/978-3-030-00928-1_48","article-title":"Concurrent spatial and channel squeeze and excitation in fully convolutional networks","volume":"48","author":"Roy","year":"2018","journal-title":"Med Image Anal."},{"key":"B33","doi-asserted-by":"publisher","first-page":"5517","DOI":"10.1038\/s41598-023-32312-1","article-title":"Multi-scale detection of pulmonary nodules by integrating attention mechanism","volume":"13","author":"Cao","year":"2023","journal-title":"Sci Rep."},{"key":"B34","doi-asserted-by":"publisher","first-page":"97187","DOI":"10.1038\/s41598-025-97187-w","article-title":"Lung nodule detection using a multi-scale convolutional neural network and global channel spatial attention mechanisms","volume":"15","author":"Li","year":"2025","journal-title":"Sci Rep."},{"key":"B35","doi-asserted-by":"publisher","first-page":"166","DOI":"10.1007\/s10006-025-01463-y","article-title":"Multi-label diagnosis of dental conditions from panoramic X-rays using attention-enhanced deep learning","volume":"29","author":"Raeisi","year":"2025","journal-title":"Oral Maxillofac Surg."},{"key":"B36","doi-asserted-by":"publisher","DOI":"10.1186\/s12903-026-07727-7","article-title":"Advanced deep learning techniques for classifying dental conditions using panoramic X-ray images","author":"Golkarieh","year":"2026","journal-title":"BMC Oral Health."},{"key":"B37","doi-asserted-by":"publisher","first-page":"535","DOI":"10.1186\/s12903-025-05704-0","article-title":"Optimized classification of dental implants using convolutional neural networks and pre-trained models","volume":"25","author":"Lashaki","year":"2025","journal-title":"BMC Oral Health."},{"key":"B38","doi-asserted-by":"publisher","first-page":"1680","DOI":"10.3390\/make5040083","article-title":"A comprehensive review of YOLO architectures in computer vision: YOLOv1 to YOLOv8 and YOLO-NAS","volume":"5","author":"Terven","year":"2023","journal-title":"Mach Learn Knowl Extr."},{"key":"B39","doi-asserted-by":"publisher","first-page":"2255","DOI":"10.32604\/cmc.2025.067994","article-title":"A systematic review of YOLO-based object detection in medical imaging: advances, challenges, and future directions","volume":"85","author":"Cai","year":"2025","journal-title":"Comput Mater Contin."},{"key":"B40","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1911.09516","article-title":"Learning spatial fusion for single-shot object detection","author":"Liu","year":"2019","journal-title":"arXiv."},{"key":"B41","first-page":"13713","article-title":"Coordinate attention for efficient mobile network design","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition.","author":"Hou","year":"2021"},{"key":"B42","first-page":"3","article-title":"CBAM: convolutional block attention module","volume-title":"Proceedings of the European Conference on Computer Vision (ECCV).","author":"Woo","year":"2018"},{"key":"B43","article-title":"BAM: bottleneck attention module","volume-title":"Proceedings of the British Machine Vision Conference (BMVC).","author":"Park","year":"2018"},{"key":"B44","first-page":"11534","article-title":"ECA-Net: efficient channel attention for deep convolutional neural networks","volume-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition","author":"Wang","year":"2020"},{"key":"B45","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2410.17725","article-title":"YOLOv11: an overview of the key architectural enhancements","author":"Khanam","year":"2024","journal-title":"arXiv."},{"key":"B46","doi-asserted-by":"publisher","first-page":"124","DOI":"10.1007\/s12194-023-00756-9","article-title":"Hierarchical approach for pulmonary-nodule identification from CT images using YOLO model and a 3D neural network classifier","volume":"17","author":"Ahmadyar","year":"2023","journal-title":"Radiol Phys Technol."},{"key":"B47","doi-asserted-by":"publisher","first-page":"100301","DOI":"10.1016\/j.cosrev.2020.100301","article-title":"A comprehensive and systematic look up into deep learning based object detection techniques: a review","volume":"38","author":"Sharma","year":"2020","journal-title":"Comput Sci Rev."},{"key":"B48","doi-asserted-by":"publisher","first-page":"1080","DOI":"10.1186\/s12885-024-12823-4","article-title":"Evaluation of the invasiveness of pure ground-glass nodules based on dual-head ResNet technique","volume":"24","author":"Yang","year":"2024","journal-title":"BMC Cancer."},{"key":"B49","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TGRS.2022.3176028","article-title":"BSNet: dynamic hybrid gradient convolution based boundary-sensitive network for remote sensing image segmentation","volume":"60","author":"Hou","year":"2022","journal-title":"IEEE Trans Geosci Remote Sens."},{"key":"B50","doi-asserted-by":"publisher","first-page":"83873","DOI":"10.1109\/ACCESS.2019.2924944","article-title":"Spatially attentive visual tracking using multi-model adaptive response fusion","volume":"7","author":"Zhang","year":"2019","journal-title":"IEEE Access."},{"key":"B51","doi-asserted-by":"publisher","first-page":"18635","DOI":"10.1007\/s11042-022-14066-6","article-title":"Facial expression recognition based on improved depthwise separable convolutional network","volume":"82","author":"Huo","year":"2022","journal-title":"Multimed Tools Appl."},{"key":"B52","doi-asserted-by":"publisher","first-page":"368504251336695","DOI":"10.1177\/00368504251336695","article-title":"Omni-dimensional dynamic convolution with coordinate attention detection scheme","volume":"108","author":"Bai","year":"2025","journal-title":"Sci Prog."},{"key":"B53","doi-asserted-by":"publisher","first-page":"2193","DOI":"10.3390\/s24072193","article-title":"SEB-YOLO: an improved YOLOv5 model for remote sensing small target detection","volume":"24","author":"Hui","year":"2024","journal-title":"Sensors."},{"key":"B54","doi-asserted-by":"publisher","first-page":"4224","DOI":"10.3390\/rs15174224","article-title":"Adaptive adjacent layer feature fusion for object detection in remote sensing images","volume":"15","author":"Zhang","year":"2023","journal-title":"Remote Sens."},{"key":"B55","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/TIM.2023.3246536","article-title":"Adaptively attentional feature fusion oriented to multiscale object detection in remote sensing images","volume":"72","author":"Zhao","year":"2023","journal-title":"IEEE Trans Instrum Meas."},{"key":"B56","doi-asserted-by":"publisher","first-page":"9099","DOI":"10.1109\/TIP.2021.3118953","article-title":"Parallel residual bi-fusion feature pyramid network for accurate single-shot object detection","volume":"30","author":"Chen","year":"2020","journal-title":"IEEE Trans Image Process."},{"key":"B57","doi-asserted-by":"publisher","first-page":"3447","DOI":"10.3390\/s23073447","article-title":"PLG-ViT vision transformer with parallel local and global self-attention","volume":"23","author":"Ebert","year":"2023","journal-title":"Sensors."}],"container-title":["Frontiers in Medicine"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1885660\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,24]],"date-time":"2026-07-24T04:24:36Z","timestamp":1784867076000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmed.2026.1885660\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,24]]},"references-count":57,"alternative-id":["10.3389\/fmed.2026.1885660"],"URL":"https:\/\/doi.org\/10.3389\/fmed.2026.1885660","relation":{},"ISSN":["2296-858X"],"issn-type":[{"value":"2296-858X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,24]]},"article-number":"1885660"}}