{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T13:24:13Z","timestamp":1783085053228,"version":"3.54.6"},"reference-count":32,"publisher":"BMJ","issue":"4","license":[{"start":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T00:00:00Z","timestamp":1783036800000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"name":"\"Pioneer\" and \"Leading Goose\u201d R&D Program of Zhejiang Province","award":["2024C03181"],"award-info":[{"award-number":["2024C03181"]}]},{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2023YFC2706400"],"award-info":[{"award-number":["2023YFC2706400"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Zhejiang Provincial Medical and Health Science and Technology Program","award":["2022KY860"],"award-info":[{"award-number":["2022KY860"]}]},{"name":"Medical Health Science and Technology Project of Zhejiang Provincial Health Commission","award":["2023KY832"],"award-info":[{"award-number":["2023KY832"]}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["World Jnl Ped Surgery"],"accepted":{"date-parts":[[2026,5,19]]},"published-print":{"date-parts":[[2026,8]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>As the most common extracranial solid tumor in children, accurate and consistent pathological diagnosis of neuroblastoma (NB) is critical for clinical decision-making. However, traditional methods relying on manual interpretation are constrained by tumor heterogeneity and inter-observer variability, necessitating the development of objective and quantitative intelligent auxiliary diagnostic technologies.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study introduces a two-stage deep learning framework with hybrid supervision, achieving precise classification of NB through collaborative optimization of cell-level segmentation and multiscale classification. In the first stage, the enhanced Medical Segment Anything Model (MedSAM) uses a cross-attention mechanism to achieve pixel-level tumor cell segmentation, attaining a Dice coefficient of 0.94. In the second stage, an optimized Swin Transformer (ST) is employed to construct the classification network, enabling full slice analysis via a confidence voting strategy.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>On an independent dataset comprising 185 whole-slide images from 185 patients, the model attained an overall area under the receiver operating characteristic curve (AUC) of 0.864 (95% confidence interval (CI): 0.747 to 0.952), significantly outperforming existing methods (8.7% higher than the second-best model, ResNeXt). Among the three NB subtypes (undifferentiated, poorly differentiated, and differentiating), the recognition accuracy is the highest for the poorly differentiated subtype, which accounts for the largest proportion in clinical practice.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>This approach effectively addresses issues related to the tumor microenvironment interference and small-sample generalization through a cascaded feature extraction and decision-making mechanism, providing a robust intelligent auxiliary tool for NB pathological diagnosis.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/wjps-2025-001149","type":"journal-article","created":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T13:11:36Z","timestamp":1783084296000},"page":"e001149","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Dual-stage deep learning framework for neuroblastoma differentiation by integrating cell segmentation and multiscale modeling"],"prefix":"10.1136","volume":"9","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1603-1728","authenticated-orcid":false,"given":"Jieni","family":"Xiong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8868-2525","authenticated-orcid":false,"given":"Zhu","family":"Zhu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Weizhong","family":"Gu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yawen","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Manli","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiabin","family":"Cai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chensheng","family":"Sun","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7749-2475","authenticated-orcid":false,"given":"Jinhu","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9935-9969","authenticated-orcid":false,"given":"Gang","family":"Yu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,7,3]]},"reference":[{"key":"2026070306100979000_9.4.e001149.1","doi-asserted-by":"publisher","DOI":"10.1002\/med.21750"},{"key":"2026070306100979000_9.4.e001149.2","doi-asserted-by":"publisher","DOI":"10.1002\/pbc.24433"},{"key":"2026070306100979000_9.4.e001149.3","doi-asserted-by":"publisher","DOI":"10.1038\/s41571-022-00643-z"},{"key":"2026070306100979000_9.4.e001149.4","doi-asserted-by":"publisher","DOI":"10.1200\/JCO.2008.16.6785"},{"key":"2026070306100979000_9.4.e001149.5","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2008.08.027"},{"key":"2026070306100979000_9.4.e001149.6","doi-asserted-by":"crossref","DOI":"10.4103\/jpi.jpi_73_17","article-title":"Convolutional Deep Belief Network with Feature Encoding for Classification of Neuroblastoma Histological Images","volume":"9","author":"Gheisari","year":"2018","journal-title":"J Pathol Inform"},{"key":"2026070306100979000_9.4.e001149.7","doi-asserted-by":"crossref","first-page":"395","DOI":"10.1109\/TMI.2020.3027547","article-title":"DetexNet: Accurately Diagnosing Frequent and Challenging Pediatric Malignant Tumors","volume":"40","author":"Liu","year":"2021","journal-title":"IEEE Trans Med Imaging"},{"key":"2026070306100979000_9.4.e001149.8","doi-asserted-by":"crossref","DOI":"10.1038\/s41698-024-00745-0","article-title":"Artificial intelligence-based morphologic classification and molecular characterization of neuroblastic tumors from digital histopathology","volume":"8","author":"Ramesh","year":"2024","journal-title":"NPJ Precis Oncol"},{"key":"2026070306100979000_9.4.e001149.9","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1097-0142(19990715)86:2<364::AID-CNCR21>3.0.CO;2-7"},{"key":"2026070306100979000_9.4.e001149.10","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2008.10.035"},{"key":"2026070306100979000_9.4.e001149.11","doi-asserted-by":"crossref","unstructured":"Tafavogh S , Meng Q , Catchpoole DR , et al . Automated quantitative and qualitative analysis of whole neuroblastoma tumour images for prognosis. The 10th lASTED international conference on biomedical engineering; 2014:244\u201351.","DOI":"10.2316\/P.2014.818-079"},{"key":"2026070306100979000_9.4.e001149.12","doi-asserted-by":"crossref","DOI":"10.3390\/diagnostics8030056","article-title":"Computer Aided Classification of Neuroblastoma Histological Images Using Scale Invariant Feature Transform with Feature Encoding","volume":"8","author":"Gheisari","year":"2018","journal-title":"Diagnostics (Basel)"},{"key":"2026070306100979000_9.4.e001149.13","doi-asserted-by":"crossref","DOI":"10.3389\/fonc.2020.586292","article-title":"SuperHistopath: A Deep Learning Pipeline for Mapping Tumor Heterogeneity on Low-Resolution Whole-Slide Digital Histopathology Images","volume":"10","author":"Zormpas-Petridis","year":"2020","journal-title":"Front Oncol"},{"key":"2026070306100979000_9.4.e001149.14","first-page":"5998","article-title":"Attention is all you need","volume":"30","author":"Vaswani","year":"2017","journal-title":"Neural Information Processing Systems"},{"key":"2026070306100979000_9.4.e001149.15","doi-asserted-by":"crossref","unstructured":"Liu Z , Lin Y , Cao Y , et al . Swin transformer: hierarchical vision transformer using shifted windows. 2021 IEEE\/CVF International Conference on Computer Vision (ICCV); 2021:9992\u201310002.","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"2026070306100979000_9.4.e001149.16","doi-asserted-by":"crossref","first-page":"125009","DOI":"10.1016\/j.eswa.2024.125009","article-title":"Swin transformer-based fork architecture for automated breast tumor classification","volume":"256","author":"\u00dczen","year":"2024","journal-title":"Expert Syst Appl"},{"key":"2026070306100979000_9.4.e001149.17","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-024-02857-3"},{"key":"2026070306100979000_9.4.e001149.18","unstructured":"Ilse M , Tomczak J , Welling M . Attention-based deep multiple instance learning. International conference on machine learning; 2018:2127\u201336."},{"key":"2026070306100979000_9.4.e001149.19","doi-asserted-by":"crossref","first-page":"3950","DOI":"10.1109\/TCYB.2019.2935141","article-title":"Weakly Supervised Deep Learning for Whole Slide Lung Cancer Image Analysis","volume":"50","author":"Wang","year":"2020","journal-title":"IEEE Trans Cybern"},{"key":"2026070306100979000_9.4.e001149.20","doi-asserted-by":"crossref","DOI":"10.3390\/cancers12020507","article-title":"Successful Identification of Nasopharyngeal Carcinoma in Nasopharyngeal Biopsies Using Deep Learning","volume":"12","author":"Chuang","year":"2020","journal-title":"Cancers (Basel)"},{"key":"2026070306100979000_9.4.e001149.21","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-019-0508-1"},{"key":"2026070306100979000_9.4.e001149.22","doi-asserted-by":"crossref","first-page":"106607","DOI":"10.1016\/j.bspc.2024.106607","article-title":"Classify breast cancer pathological tissue images using multi-scale bar convolution pooling structure with patch attention","volume":"96","author":"Guo","year":"2024","journal-title":"Biomed Signal Process Control"},{"key":"2026070306100979000_9.4.e001149.23","doi-asserted-by":"publisher","DOI":"10.1002\/jemt.24716"},{"key":"2026070306100979000_9.4.e001149.24","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-44824-z"},{"key":"2026070306100979000_9.4.e001149.25","doi-asserted-by":"crossref","unstructured":"He K , Zhang X , Ren S , et al . Deep residual learning for image recognition. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR); 2016:770\u20138.","DOI":"10.1109\/CVPR.2016.90"},{"key":"2026070306100979000_9.4.e001149.26","doi-asserted-by":"crossref","unstructured":"Huang G , Liu Z , Van Der Maaten L , et al . Densely connected convolutional networks. 2017 IEEE conference on Computer Vision and Pattern Recognition (CVPR); 2017:2261\u20139.","DOI":"10.1109\/CVPR.2017.243"},{"key":"2026070306100979000_9.4.e001149.27","doi-asserted-by":"crossref","unstructured":"Xie S , Girshick R , Dollar P , et al . Aggregated residual transformations for deep neural networks. 2017 IEEE conference on Computer Vision and Pattern Recognition (CVPR); 2017:5987\u201395.","DOI":"10.1109\/CVPR.2017.634"},{"key":"2026070306100979000_9.4.e001149.28","doi-asserted-by":"crossref","unstructured":"Woo S , Park J , Lee JY , et al . CBAM: convolutional block attention module. European Conference on Computer Vision (ECCV); 2018:3\u201319.","DOI":"10.1007\/978-3-030-01234-2_1"},{"key":"2026070306100979000_9.4.e001149.29","doi-asserted-by":"crossref","unstructured":"Zhang X , Zhou X , Lin M , et al . ShuffleNet: an extremely efficient convolutional neural network for mobile devices. 2018 IEEE\/CVF conference on computer vision and pattern recognition; 2017:6848\u201356.","DOI":"10.1109\/CVPR.2018.00716"},{"key":"2026070306100979000_9.4.e001149.30","doi-asserted-by":"publisher","DOI":"10.1109\/TMI.2022.3176598"},{"key":"2026070306100979000_9.4.e001149.31","doi-asserted-by":"crossref","first-page":"555","DOI":"10.1038\/s41551-020-00682-w","article-title":"Data-efficient and weakly supervised computational pathology on whole-slide images","volume":"5","author":"Lu","year":"2021","journal-title":"Nat Biomed Eng"},{"key":"2026070306100979000_9.4.e001149.32","first-page":"641","article-title":"Distributed and Time-Delayed k-Winner-Take-All Network for Competitive Coordination of Multiple Robots","volume":"PP","author":"Jin","year":"2022","journal-title":"IEEE Trans Cybern"}],"container-title":["World Journal of Pediatric Surgery"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/wjps-2025-001149","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T13:11:53Z","timestamp":1783084313000},"score":1,"resource":{"primary":{"URL":"https:\/\/wjps.bmj.com\/lookup\/doi\/10.1136\/wjps-2025-001149"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,3]]},"references-count":32,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,7,3]]},"published-print":{"date-parts":[[2026,8]]}},"alternative-id":["10.1136\/wjps-2025-001149"],"URL":"https:\/\/doi.org\/10.1136\/wjps-2025-001149","relation":{},"ISSN":["2516-5410"],"issn-type":[{"value":"2516-5410","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,3]]}}}