{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T06:03:55Z","timestamp":1790229835301,"version":"4.1.0"},"reference-count":34,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T00:00:00Z","timestamp":1790208000000},"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. Mol. Biosci."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>To develop and externally validate a preoperative CT imaging model integrating primary tumour and visceral adipose tissue (VAT) features for predicting recurrence in colorectal cancer (CRC).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>This retrospective multicentre study included 1,369 patients with CRC undergoing upfront curative-intent resection, comprising a training cohort (n = 705) and two external test cohorts (n = 425 and 239). A 3D convolutional neural network extracted tumour-centred volumetric features, while an attention-based multiple instance learning branch aggregated whole-abdominal VAT slices into patient-level representations. The two CT-derived scores constituted the preoperative imaging model. A clinicopathological component was combined with the imaging scores to construct the postoperative multimodal model; robustness to variable selection was evaluated using an all-candidate ridge-penalised Cox sensitivity model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Both tumour and VAT DL scores remained independently associated with recurrence-free survival after multivariable adjustment. In external test cohorts 1 and 2, the preoperative Tumor + VAT imaging model achieved C-indices of 0.734 and 0.772, whereas the postoperative multimodal model achieved C-indices of 0.772 and 0.810 and 2-year time-dependent AUCs of 0.757 and 0.836, respectively. Training-derived tertile cut-offs from the postoperative model consistently stratified patients into low-, intermediate- and high-risk groups across all cohorts (all log-rank P &amp;lt; 0.001). In the two external cohorts, the clinicopathological-only model achieved C-indices of 0.709 and 0.715, compared with 0.772 and 0.810 for the postoperative multimodal model.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The preoperative CT model non-invasively stratified recurrence risk using tumour and whole-abdominal VAT phenotypes, while postoperative clinicopathological features further improved prognostic performance.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fmolb.2026.1937172","type":"journal-article","created":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T05:44:18Z","timestamp":1790228658000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Preoperative CT deep learning of tumor and visceral adipose tissue combined with postoperative pathology predicts colorectal cancer recurrence: a multicenter study"],"prefix":"10.3389","volume":"13","author":[{"given":"Pengfei","family":"Wei","sequence":"first","affiliation":[{"name":"Department of Radiology, Jiangxi Cancer Hospital & Institute, Jiangxi Clinical Research Center for Cancer, The Second Affiliated Hospital of Nanchang Medical College","place":["Nanchang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Linhua","family":"Zhong","sequence":"additional","affiliation":[{"name":"Department of Radiology, Jiangxi Cancer Hospital & Institute, Jiangxi Clinical Research Center for Cancer, The Second Affiliated Hospital of Nanchang Medical College","place":["Nanchang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongming","family":"Tan","sequence":"additional","affiliation":[{"name":"Department of Radiology, The First Affiliated Hospital, Jiangxi Medical College, Nanchang University","place":["Nanchang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinhong","family":"Zhao","sequence":"additional","affiliation":[{"name":"Department of Radiology, The Second Affiliated Hospital, Jiangxi Medical College, Nanchang University","place":["Nanchang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongjie","family":"Zhou","sequence":"additional","affiliation":[{"name":"Department of Radiology, Jiangxi Cancer Hospital & Institute, Jiangxi Clinical Research Center for Cancer, The Second Affiliated Hospital of Nanchang Medical College","place":["Nanchang, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,24]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1007\/s00330-021-08104-4","article-title":"Prediction of recurrence after surgery in colorectal cancer patients using radiomics from diagnostic contrast-enhanced computed tomography: a two-center study","volume":"32","author":"Badic","year":"2022","journal-title":"Eur. Radiol."},{"key":"B2","doi-asserted-by":"publisher","first-page":"e70115","DOI":"10.1111\/1759-7714.70115","article-title":"Radiomic analysis and liquid biopsy in preoperative CT of NSCLC: an explorative experience","volume":"16","author":"Belfiore","year":"2025","journal-title":"Thorac. Cancer"},{"key":"B3","doi-asserted-by":"publisher","first-page":"e240029","DOI":"10.6004\/jnccn.2024.0029","article-title":"Colon cancer, version 3.2024, NCCN clinical practice guidelines in oncology","volume":"22","author":"Benson","year":"2024","journal-title":"J. Natl. Compr. Canc Netw."},{"key":"B4","doi-asserted-by":"publisher","first-page":"229","DOI":"10.3322\/caac.21834","article-title":"Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries","volume":"74","author":"Bray","year":"2024","journal-title":"CA Cancer J. Clin."},{"key":"B5","doi-asserted-by":"publisher","first-page":"13128","DOI":"10.3390\/ijms232113128","article-title":"Insights behind the relationship between colorectal cancer and obesity: is visceral adipose tissue the missing link?","volume":"23","author":"Chaplin","year":"2022","journal-title":"Int. J. Mol. Sci."},{"key":"B6","doi-asserted-by":"publisher","first-page":"102105","DOI":"10.1016\/j.media.2021.102105","article-title":"Dual attention multiple instance learning with unsupervised complementary loss for COVID-19 screening","volume":"72","author":"Chikontwe","year":"2021","journal-title":"Med. Image Anal."},{"key":"B7","doi-asserted-by":"publisher","first-page":"e078378","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"Collins","year":"2024","journal-title":"BMJ"},{"key":"B8","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1111\/j.2517-6161.1972.tb00899.x","article-title":"Regression models and life-tables","volume":"34","author":"Cox","year":"1972","journal-title":"J. R. Stat. Soc. Ser. B Stat. Methodol."},{"key":"B9","doi-asserted-by":"publisher","first-page":"1467","DOI":"10.1016\/S0140-6736(19)32319-0","article-title":"Colorectal cancer","volume":"394","author":"Dekker","year":"2019","journal-title":"Lancet"},{"key":"B10","first-page":"2127","article-title":"Attention-based deep multiple instance learning","author":"Ilse","year":"2018","journal-title":"Proceedings of the 35th International Conference on Machine Learning"},{"key":"B11","doi-asserted-by":"publisher","first-page":"e33","DOI":"10.1016\/S2589-7500(23)00208-X","article-title":"End-to-end prognostication in colorectal cancer by deep learning: a retrospective, multicentre study","volume":"6","author":"Jiang","year":"2024","journal-title":"Lancet Digit. Health"},{"key":"B12","doi-asserted-by":"publisher","first-page":"886","DOI":"10.1038\/nrc3174","article-title":"Molecular mechanisms of cancer development in obesity","volume":"11","author":"Khandekar","year":"2011","journal-title":"Nat. Rev. Cancer"},{"key":"B13","doi-asserted-by":"publisher","first-page":"103","DOI":"10.1097\/PAS.0000000000000518","article-title":"Perineural invasion is a strong prognostic factor in colorectal cancer: a systematic review","volume":"40","author":"Knijn","year":"2016","journal-title":"Am. J. Surg. Pathol."},{"key":"B14","doi-asserted-by":"publisher","first-page":"2527","DOI":"10.1002\/ijc.26031","article-title":"The seed and soil hypothesis revisited-the role of tumor-stroma interactions in metastasis to different organs","volume":"128","author":"Langley","year":"2011","journal-title":"Int. J. Cancer"},{"key":"B15","doi-asserted-by":"publisher","first-page":"436","DOI":"10.1038\/nature14539","article-title":"Deep learning","volume":"521","author":"LeCun","year":"2015","journal-title":"Nature"},{"key":"B16","doi-asserted-by":"publisher","first-page":"103442","DOI":"10.1016\/j.ebiom.2021.103442","article-title":"Deep learning radiomics-based prediction of distant metastasis in patients with locally advanced rectal cancer after neoadjuvant chemoradiotherapy: a multicentre study","volume":"69","author":"Liu","year":"2021","journal-title":"EBioMedicine"},{"key":"B17","doi-asserted-by":"publisher","first-page":"6654","DOI":"10.1038\/s41467-021-26990-6","article-title":"Deep learning for the prediction of early on-treatment response in metastatic colorectal cancer from serial medical imaging","volume":"12","author":"Lu","year":"2021","journal-title":"Nat. Commun."},{"key":"B18","doi-asserted-by":"publisher","first-page":"2914","DOI":"10.3390\/diagnostics15222914","article-title":"Delta-radiomics biomarker in colorectal cancer liver metastases treated with cetuximab plus avelumab (CAVE trial)","volume":"15","author":"Nardone","year":"2025","journal-title":"Diagn. (Basel)"},{"key":"B19","doi-asserted-by":"publisher","first-page":"571","DOI":"10.1016\/S0140-6736(00)49915-0","article-title":"The distribution of secondary growths in cancer of the breast","volume":"133","author":"Paget","year":"1889","journal-title":"Lancet"},{"key":"B20","doi-asserted-by":"publisher","first-page":"53","DOI":"10.5009\/gnl20266","article-title":"Impact of visceral fat on survival and metastasis of stage III colorectal cancer","volume":"16","author":"Park","year":"2022","journal-title":"Gut Liver"},{"key":"B21","doi-asserted-by":"publisher","first-page":"174","DOI":"10.1007\/s00384-025-04969-9","article-title":"MRI-based radiomics for preoperative T-staging of rectal cancer: a retrospective analysis","volume":"40","author":"Patane","year":"2025","journal-title":"Int. J. Colorectal Dis."},{"key":"B22","doi-asserted-by":"publisher","first-page":"5177","DOI":"10.3748\/wjg.v20.i18.5177","article-title":"Obesity and colorectal cancer: role of adipokines in tumor initiation and progression","volume":"20","author":"Riondino","year":"2014","journal-title":"World J. Gastroenterol."},{"key":"B23","doi-asserted-by":"crossref","first-page":"618","DOI":"10.1109\/ICCV.2017.74","article-title":"Grad-CAM: visual explanations from deep networks via gradient-based localization","volume-title":"2017 IEEE International Conference on Computer Vision (ICCV)","author":"Selvaraju","year":"2017"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1075434","DOI":"10.1155\/2019\/1075434","article-title":"Automated segmentation of colorectal tumor in 3D MRI using 3D multiscale densely connected convolutional neural network","volume":"2019","author":"Soomro","year":"2019","journal-title":"J. Healthc. Eng."},{"key":"B25","doi-asserted-by":"publisher","first-page":"e240300","DOI":"10.1148\/ryai.240300","article-title":"Checklist for artificial intelligence in medical imaging (CLAIM): 2024 update","volume":"6","author":"Tejani","year":"2024","journal-title":"Radiol. Artif. Intell."},{"key":"B26","doi-asserted-by":"publisher","first-page":"2261","DOI":"10.1056\/NEJMoa2200075","article-title":"Circulating tumor DNA analysis guiding adjuvant therapy in stage II colon cancer","volume":"386","author":"Tie","year":"2022","journal-title":"N. Engl. J. Med."},{"key":"B27","doi-asserted-by":"publisher","first-page":"297","DOI":"10.1038\/nrclinonc.2016.199","article-title":"Surveillance after curative treatment for colorectal cancer","volume":"14","author":"van der Stok","year":"2017","journal-title":"Nat. Rev. Clin. Oncol."},{"key":"B28","doi-asserted-by":"publisher","first-page":"581","DOI":"10.1038\/s41556-026-01885-0","article-title":"Peritumoural adipose tissue drives immune evasion in colorectal cancer via adipose-mesenchymal transformation","volume":"28","author":"Wang","year":"2026","journal-title":"Nat. Cell Biol."},{"key":"B29","doi-asserted-by":"publisher","first-page":"e230024","DOI":"10.1148\/ryai.230024","article-title":"TotalSegmentator: robust segmentation of 104 anatomic structures in CT images","volume":"5","author":"Wasserthal","year":"2023","journal-title":"Radiol. Artif. Intell."},{"key":"B30","doi-asserted-by":"publisher","first-page":"1113","DOI":"10.1007\/s11605-015-2761-z","article-title":"Prognostic value of perineural invasion in colorectal cancer: a meta-analysis","volume":"19","author":"Yang","year":"2015","journal-title":"J. Gastrointest. Surg."},{"key":"B31","doi-asserted-by":"publisher","first-page":"101789","DOI":"10.1016\/j.media.2020.101789","article-title":"Whole slide images based cancer survival prediction using attention guided deep multiple instance learning networks","volume":"65","author":"Yao","year":"2020","journal-title":"Med. Image Anal."},{"key":"B32","doi-asserted-by":"publisher","first-page":"108874","DOI":"10.1016\/j.cmpb.2025.108874","article-title":"Interpretable machine learning models based on body composition and inflammatory nutritional index (BCINI) to predict early postoperative recurrence of colorectal cancer: multi-center study","volume":"269","author":"Zhou","year":"","journal-title":"Comput. Methods Programs Biomed."},{"key":"B33","doi-asserted-by":"publisher","first-page":"110482","DOI":"10.1016\/j.ejso.2025.110482","article-title":"CT-based radiomics deep learning signatures for noninvasive prediction of early recurrence after radical surgery in locally advanced colorectal cancer: a multicenter study","volume":"51","author":"Zhou","year":"","journal-title":"Eur. J. Surg. Oncol."},{"key":"B34","doi-asserted-by":"publisher","first-page":"1288","DOI":"10.1245\/s10434-025-18815-3","article-title":"Development and validation of time-to-event machine learning models for predicting disease-free survival in patients with locally advanced colorectal cancer: a multicenter cohort study","volume":"33","author":"Zhou","year":"2026","journal-title":"Ann. Surg. Oncol."}],"container-title":["Frontiers in Molecular Biosciences"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmolb.2026.1937172\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T05:44:20Z","timestamp":1790228660000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fmolb.2026.1937172\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,24]]},"references-count":34,"alternative-id":["10.3389\/fmolb.2026.1937172"],"URL":"https:\/\/doi.org\/10.3389\/fmolb.2026.1937172","relation":{},"ISSN":["2296-889X"],"issn-type":[{"value":"2296-889X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,24]]},"article-number":"1937172"}}