{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T17:19:21Z","timestamp":1788542361712,"version":"build-2803163510"},"reference-count":35,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T00:00:00Z","timestamp":1788480000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T00:00:00Z","timestamp":1788480000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"name":"Chongqing Natural Science Foundation Project","award":["CSTB2022NSCQ-MSX0773"],"award-info":[{"award-number":["CSTB2022NSCQ-MSX0773"]}]},{"DOI":"10.13039\/100017501","name":"Science-Health Joint Medical Scientific Research Project of Chongqing","doi-asserted-by":"publisher","award":["2023MSXM040"],"award-info":[{"award-number":["2023MSXM040"]}],"id":[{"id":"10.13039\/100017501","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Clinical Research Incubation Project of the First Affiliated Hospital of Army Medical University","award":["2024IITZDB11"],"award-info":[{"award-number":["2024IITZDB11"]}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Insights Imaging"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>This study aimed to develop a deep learning model (MST-Net) for the segmentation-free automated and precise differentiation between T2 and T3 staging of rectal cancer based on preoperative T2-weighted magnetic resonance imaging (T2WI).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>Preoperative T2WI images and postoperative pathologically confirmed T2\/T3 staging labels from rectal cancer patients were retrospectively collected from two centers between January 2020 and January 2025. The core of MST-Net includes: a pyramid cross-stage feature fusion module to integrate shallow texture details and deep semantic information, mitigating the loss of spatial details in deep networks; and a lightweight spatial attention mechanism to adaptively focus on key radiomics features within the tumor region. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC) and accuracy (ACC).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      A total of 309 patients were included in the analysis, comprising a training set (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u2009216) and an internal test set (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u200954) from Center 1, and an external independent validation set (\n                      <jats:italic>n<\/jats:italic>\n                      \u2009=\u200939) from Center 2. MST-Net achieved an AUC of 0.95 (ACC\u2009=\u200990%) on the internal test set, outperforming junior radiologists (ACC\u2009=\u200969%) and performing comparably to senior radiologists (ACC\u2009=\u200985%). On the external validation set, it achieved an AUC of 0.85 (ACC\u2009=\u200979%), also surpassing junior radiologists (ACC\u2009=\u200965%) and reaching the level of senior radiologists (ACC\u2009=\u200972%).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>MST-Net enables segmentation-free automated and high-accuracy prediction of T2\/T3 staging in rectal cancer based on routine T2WI images. This model provides a reproducible auxiliary tool for clinical personalized treatment decision-making, potentially reducing the dependence of staging results on physician experience.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Key Points<\/jats:title>\n                    <jats:p>\n                      <jats:list list-type=\"bullet\">\n                        <jats:list-item>\n                          <jats:p>\n                            <jats:bold>Question<\/jats:bold>\n                            Accurate differentiation between T2 and T3 rectal cancer on preoperative T2-weighted MRI remains clinically challenging but is essential for treatment planning.\n                          <\/jats:p>\n                        <\/jats:list-item>\n                        <jats:list-item>\n                          <jats:p>\n                            <jats:bold>Findings<\/jats:bold>\n                            MST-Net achieved high diagnostic accuracy and robust discrimination for T2 versus T3 staging in both internal and external validation cohorts.\n                          <\/jats:p>\n                        <\/jats:list-item>\n                        <jats:list-item>\n                          <jats:p>\n                            <jats:bold>Critical Relevance<\/jats:bold>\n                            This study demonstrates that a segmentation-free deep learning model using routine T2-weighted MRI can enable accurate preoperative T-staging of rectal cancer, supporting standardized and efficient radiological assessment and facilitating more consistent clinical decision-making in rectal cancer management.\n                          <\/jats:p>\n                        <\/jats:list-item>\n                      <\/jats:list>\n                    <\/jats:p>\n                  <\/jats:sec>","DOI":"10.1186\/s13244-026-02377-3","type":"journal-article","created":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T17:08:11Z","timestamp":1788541691000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["MST-Net: a multi-scale spatial transformation network for automated T2\/T3 staging of rectal cancer on preoperative T2-weighted MRI"],"prefix":"10.1186","volume":"17","author":[{"given":"Wei","family":"Liu","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ling","family":"Yin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiong","family":"Ran","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tian-Xiang","family":"Xu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yun","family":"Du","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiang-Yi","family":"He","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shao-Quan","family":"Zhou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4402-5472","authenticated-orcid":false,"given":"Jian-Jun","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,9,4]]},"reference":[{"key":"2377_CR1","first-page":"229","volume":"74","author":"F Bray","year":"2024","unstructured":"Bray F, Laversanne M, Sung H et al (2024) Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 74:229\u2013263","journal-title":"CA Cancer J Clin"},{"key":"2377_CR2","doi-asserted-by":"publisher","first-page":"483","DOI":"10.21147\/j.issn.1000-9604.2022.05.08","volume":"34","author":"Q Li","year":"2022","unstructured":"Li Q, Wu H, Cao M et al (2022) Colorectal cancer burden, trends and risk factors in China: a review and comparison with the United States. Chin J Cancer Res 34:483\u2013495","journal-title":"Chin J Cancer Res"},{"key":"2377_CR3","first-page":"47","volume":"4","author":"B Han","year":"2024","unstructured":"Han B, Zheng R, Zeng H et al (2024) Cancer incidence and mortality in China, 2022. J Natl Cancer Cent 4:47\u201353","journal-title":"J Natl Cancer Cent"},{"key":"2377_CR4","doi-asserted-by":"publisher","first-page":"1139","DOI":"10.6004\/jnccn.2022.0051","volume":"20","author":"AB Benson","year":"2022","unstructured":"Benson AB, Venook AP, Al-Hawary MM et al (2022) Rectal cancer, version 2.2022, NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw 20:1139\u20131167","journal-title":"J Natl Compr Canc Netw"},{"key":"2377_CR5","doi-asserted-by":"publisher","first-page":"1104","DOI":"10.1007\/s00330-018-5719-1","volume":"29","author":"MJ Gollub","year":"2019","unstructured":"Gollub MJ, Blazic I, Felder S et al (2019) Value of adding dynamic contrast-enhanced MRI visual assessment to conventional MRI and clinical assessment in the diagnosis of complete tumour response to chemoradiotherapy for rectal cancer. Eur Radiol 29:1104\u20131113","journal-title":"Eur Radiol"},{"key":"2377_CR6","doi-asserted-by":"publisher","DOI":"10.2214\/AJR.24.31523","volume":"224","author":"H Khasawneh","year":"2025","unstructured":"Khasawneh H, Khatri G, Sheedy SP et al (2025) MRI for rectal cancer: updates and controversies\u2014AJR Expert Panel Narrative Review. AJR Am J Roentgenol 224:e2431523","journal-title":"AJR Am J Roentgenol"},{"key":"2377_CR7","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.232838","volume":"314","author":"S Nougaret","year":"2025","unstructured":"Nougaret S, Gormly K, Lambregts DMJ et al (2025) MRI of the rectum: a decade into DISTANCE, moving to DISTANCED. Radiology 314:e232838","journal-title":"Radiology"},{"key":"2377_CR8","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.222223","volume":"307","author":"X Jiang","year":"2023","unstructured":"Jiang X, Zhao H, Saldanha OL et al (2023) An MRI deep learning model predicts outcome in rectal cancer. Radiology 307:e222223","journal-title":"Radiology"},{"key":"2377_CR9","doi-asserted-by":"publisher","first-page":"6594","DOI":"10.7150\/jca.32685","volume":"10","author":"L Yu","year":"2019","unstructured":"Yu L, Wang L, Tan Y et al (2019) Accuracy of magnetic resonance imaging in staging rectal cancer with multidisciplinary team: a single-center experience. J Cancer 10:6594\u20136598","journal-title":"J Cancer"},{"key":"2377_CR10","doi-asserted-by":"publisher","first-page":"334","DOI":"10.1097\/SLA.0000000000004499","volume":"276","author":"AC Lord","year":"2022","unstructured":"Lord AC, D\u2019Souza N, Shaw A et al (2022) MRI-diagnosed tumor deposits and EMVI status have superior prognostic accuracy to current clinical TNM staging in rectal cancer. Ann Surg 276:334\u2013344","journal-title":"Ann Surg"},{"key":"2377_CR11","doi-asserted-by":"publisher","first-page":"367","DOI":"10.1148\/rg.2019180114","volume":"39","author":"N Horvat","year":"2019","unstructured":"Horvat N, Rocha CCT, Oliveira BC et al (2019) MRI of rectal cancer: tumor staging, imaging techniques, and management. Radiographics 39:367\u2013387","journal-title":"Radiographics"},{"key":"2377_CR12","doi-asserted-by":"publisher","first-page":"4525","DOI":"10.1007\/s00261-021-03137-1","volume":"46","author":"X Lin","year":"2021","unstructured":"Lin X, Zhao S, Jiang HJ et al (2021) A radiomics-based nomogram for preoperative T staging prediction of rectal cancer. Abdom Radiol 46:4525\u20134535","journal-title":"Abdom Radiol"},{"key":"2377_CR13","doi-asserted-by":"publisher","first-page":"1476","DOI":"10.3389\/fonc.2020.01476","volume":"10","author":"HC Lu","year":"2020","unstructured":"Lu HC, Wang F, Yin JD (2020) Texture analysis based on sagittal fat-suppression and transverse T2-weighted magnetic resonance imaging for determining local invasion of rectal cancer. Front Oncol 10:1476","journal-title":"Front Oncol"},{"key":"2377_CR14","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1007\/s00384-025-04969-9","volume":"40","author":"V Patan\u00e8","year":"2025","unstructured":"Patan\u00e8 V, Atripaldi U, Sansone M et al (2025) MRI-based radiomics for preoperative T-staging of rectal cancer: a retrospective analysis. Int J Colorectal Dis 40:174","journal-title":"Int J Colorectal Dis"},{"key":"2377_CR15","doi-asserted-by":"publisher","first-page":"821","DOI":"10.1097\/CM9.0000000000001401","volume":"134","author":"QY Wu","year":"2021","unstructured":"Wu QY, Liu SL, Sun P et al (2021) Establishment and clinical application value of an automatic diagnosis platform for rectal cancer T-staging based on a deep neural network. Chin Med J 134:821\u2013828","journal-title":"Chin Med J"},{"key":"2377_CR16","doi-asserted-by":"publisher","first-page":"247","DOI":"10.1186\/s12876-024-03316-6","volume":"24","author":"L Fan","year":"2024","unstructured":"Fan L, Wu H, Wu Y et al (2024) Preoperative prediction of rectal cancer staging combining MRI deep transfer learning, radiomics features, and clinical factors: accurate differentiation from stage T2 to T3. BMC Gastroenterol 24:247","journal-title":"BMC Gastroenterol"},{"key":"2377_CR17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1007\/s00330-022-08952-8","volume":"33","author":"M Hou","year":"2023","unstructured":"Hou M, Zhou L, Sun J (2023) Deep-learning-based 3D super-resolution MRI radiomics model: superior predictive performance in preoperative T-staging of rectal cancer. Eur Radiol 33:1\u201310","journal-title":"Eur Radiol"},{"key":"2377_CR18","doi-asserted-by":"publisher","first-page":"1850","DOI":"10.1007\/s00261-023-04164-w","volume":"49","author":"B Deng","year":"2024","unstructured":"Deng B, Wang Q, Liu Y et al (2024) A nomogram based on MRI radiomics features of mesorectal fat for diagnosing T2- and T3-stage rectal cancer. Abdom Radiol 49:1850\u20131860","journal-title":"Abdom Radiol"},{"key":"2377_CR19","doi-asserted-by":"publisher","DOI":"10.1186\/s12880-019-0392-7","volume":"19","author":"X Ma","year":"2019","unstructured":"Ma X, Shen F, Jia Y et al (2019) MRI-based radiomics of rectal cancer: preoperative assessment of the pathological features. BMC Med Imaging 19:86","journal-title":"BMC Med Imaging"},{"key":"2377_CR20","doi-asserted-by":"publisher","DOI":"10.1016\/j.bdr.2022.100346","volume":"30","author":"X Zhang","year":"2022","unstructured":"Zhang X, Zhang B, Wang B et al (2022) Automatic prediction of T2\/T3 staging of rectal cancer based on radiomics and machine learning. Big Data Res 30:100346","journal-title":"Big Data Res"},{"key":"2377_CR21","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2023.1326324","volume":"10","author":"C Tian","year":"2023","unstructured":"Tian C, Ma X, Lu H et al (2023) Deep learning models for preoperative T-stage assessment in rectal cancer using MRI: exploring the impact of rectal filling. Front Med 10:1326324","journal-title":"Front Med"},{"key":"2377_CR22","doi-asserted-by":"publisher","first-page":"1083","DOI":"10.1002\/jmri.28856","volume":"59","author":"Y Wei","year":"2024","unstructured":"Wei Y, Wang H, Chen Z et al (2024) Deep learning-based multiparametric MRI model for preoperative T-stage in rectal cancer. J Magn Reson Imaging 59:1083\u20131092","journal-title":"J Magn Reson Imaging"},{"key":"2377_CR23","doi-asserted-by":"crossref","unstructured":"He K, Zhang X, Ren S et al (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR). IEEE, Las Vegas, pp 770\u2013778","DOI":"10.1109\/CVPR.2016.90"},{"key":"2377_CR24","doi-asserted-by":"crossref","unstructured":"Sandler M, Howard A, Zhu M et al (2018) MobileNetV2: inverted residuals and linear bottlenecks. In: Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR). IEEE, Salt Lake City, pp 4510\u20134520","DOI":"10.1109\/CVPR.2018.00474"},{"key":"2377_CR25","doi-asserted-by":"publisher","first-page":"4524","DOI":"10.1016\/j.acra.2025.03.048","volume":"32","author":"Y Lin","year":"2025","unstructured":"Lin Y, Liu Y, Zhang X et al (2025) A high-resolution T2WI-based deep learning model for preoperative discrimination between T2 and T3 rectal cancer: a multicenter study. Acad Radiol 32:4524\u20134531","journal-title":"Acad Radiol"},{"key":"2377_CR26","unstructured":"Tan M, Le QV (2019) EfficientNet: rethinking model scaling for convolutional neural networks. In: Proceedings of the 36th international conference on machine learning (ICML). PMLR, Long Beach, pp 6105\u20136114"},{"key":"2377_CR27","doi-asserted-by":"crossref","unstructured":"Liu Z, Mao H, Wu CY et al (2022) A ConvNet for the 2020s. In: Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR). IEEE, New Orleans, pp 11966\u201311976","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"2377_CR28","doi-asserted-by":"crossref","unstructured":"Liu Z, Lin Y, Cao Y et al (2021) Swin transformer: hierarchical vision transformer using shifted windows. In: Proceedings of the IEEE\/CVF international conference on computer vision (ICCV). IEEE, Montreal, pp 10012\u201310022","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"2377_CR29","doi-asserted-by":"crossref","unstructured":"Ma N, Zhang X, Zheng HT et al (2018) ShuffleNet V2: practical guidelines for efficient CNN architecture design. In: Proceedings of the European conference on computer vision (ECCV). Springer, Munich, pp 122\u2013138","DOI":"10.1007\/978-3-030-01264-9_8"},{"key":"2377_CR30","doi-asserted-by":"crossref","unstructured":"Chattopadhay A, Sarkar A, Howlader P et al (2018) Grad-CAM++: improved visual explanations for deep convolutional networks. In: Proceedings of the IEEE winter conference on applications of computer vision (WACV). IEEE, Lake Tahoe, pp 839\u2013847","DOI":"10.1109\/WACV.2018.00097"},{"key":"2377_CR31","doi-asserted-by":"publisher","DOI":"10.1016\/j.suronc.2023.101987","volume":"50","author":"K Arndt","year":"2023","unstructured":"Arndt K, Vigna C, Kaul S et al (2023) Magnetic resonance imaging accuracy in staging early and locally advanced rectal cancer. Surg Oncol 50:101987","journal-title":"Surg Oncol"},{"key":"2377_CR32","doi-asserted-by":"publisher","first-page":"1674","DOI":"10.1002\/jso.27852","volume":"130","author":"E Milanzi","year":"2024","unstructured":"Milanzi E, Pelly RM, Hayes IP et al (2024) Accuracy of baseline magnetic resonance imaging for staging rectal cancer patients proceeding directly to surgery. J Surg Oncol 130:1674\u20131682","journal-title":"J Surg Oncol"},{"key":"2377_CR33","doi-asserted-by":"publisher","DOI":"10.1259\/bjr.20230091","volume":"96","author":"N El Khababi","year":"2023","unstructured":"El Khababi N, Beets-Tan RG, Curvo-Semedo L et al (2023) Pearls and pitfalls of structured staging and reporting of rectal cancer on MRI: an international multireader study. Br J Radiol 96:20230091","journal-title":"Br J Radiol"},{"key":"2377_CR34","doi-asserted-by":"publisher","first-page":"364","DOI":"10.1148\/radiol.212905","volume":"305","author":"QY Lu","year":"2022","unstructured":"Lu QY, Guan Z, Zhang XY et al (2022) Contrast-enhanced MRI for T restaging of locally advanced rectal cancer following neoadjuvant chemotherapy and radiation therapy. Radiology 305:364\u2013372","journal-title":"Radiology"},{"key":"2377_CR35","doi-asserted-by":"publisher","first-page":"191","DOI":"10.1007\/s00384-020-03751-3","volume":"36","author":"S Fritz","year":"2021","unstructured":"Fritz S, Killguss H, Schaudt A et al (2021) Preoperative versus pathological staging of rectal cancer\u2014challenging the indication of neoadjuvant chemoradiotherapy. Int J Colorectal Dis 36:191\u2013194","journal-title":"Int J Colorectal Dis"}],"container-title":["Insights into Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13244-026-02377-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1186\/s13244-026-02377-3","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1186\/s13244-026-02377-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T17:08:13Z","timestamp":1788541693000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1186\/s13244-026-02377-3"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,4]]},"references-count":35,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["2377"],"URL":"https:\/\/doi.org\/10.1186\/s13244-026-02377-3","relation":{},"ISSN":["1869-4101"],"issn-type":[{"value":"1869-4101","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,4]]},"assertion":[{"value":"6 February 2026","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"1 July 2026","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"25 July 2026","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 September 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"This study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Ethics Committee of the First Affiliated Hospital of Army Medical University, PLA (Approval Number: (B) KY2025272), and all patient data included in this research were anonymized.","order":1,"name":"Ethics","label":"Ethics approval and consent to participate","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"No personal data or any identifiable statement beyond images are used in the manuscript.","order":2,"name":"Ethics","label":"Consent for publication","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare that no known financial interests or personal relationships exist that could influence the findings of this study.","order":3,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Declarations"}}],"article-number":"220"}}