{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,3]],"date-time":"2026-11-03T14:50:04Z","timestamp":1793717404607,"version":"4.6.1"},"reference-count":37,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,9,15]],"date-time":"2025-09-15T00:00:00Z","timestamp":1757894400000},"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. Immunol."],"abstract":"<jats:sec><jats:title>Background<\/jats:title><jats:p>In women with ductal carcinoma <jats:italic>in situ<\/jats:italic> (DCIS) undergoing breast-conserving surgery, still part will progress to invasive breast cancer (IBC) in the future. Mammograms offer rich tumor data for patient stratification, but current prediction methods focus on clinicopathological factors, overlooking imaging insights.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods<\/jats:title><jats:p>We retrospectively analyzed 140 DCIS patients from Harbin Medical University Cancer Hospital (2011-2020, followed up to 2025). Preoperative digital mammograms and clinicopathological data were collected, with mammographic features extracted using pyradiomics and supervised by a senior radiologist. Feature selection employed 10-fold cross-validated LASSO regression. The dataset was split into training (n=100) and validation (n=40) sets (10:4 ratio). Sixteen machine learning algorithms combining mammographic deep learning features and clinicopathological variables were developed and compared for predicting DCIS recurrence. Model performance was assessed using ROC, sensitivity, specificity, PPV, NPV, and SHAP values for interpretation.<\/jats:p><\/jats:sec><jats:sec><jats:title>Results<\/jats:title><jats:p>The Gradient Boosting Machine (GBM) algorithm had the best predictive performance, with an AUC of 0.918 (95% CI 0.873-0.963) in the test set. SHAP values indicated that the mammographic signature (MS) was the most significant predictor, followed by Ki-67 index and histological grade. Patients not receiving radiotherapy had higher recurrence rates than those who did. Decision curve analysis validated the model\u2019s clinical utility across various risk thresholds.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusion<\/jats:title><jats:p>Our study developed an interpretable GBM model incorporating mammographic and clinical data to predict DCIS recurrence (AUC\u00a0=\u00a00.918). Key predictors were mammographic signature, Ki-67, and tumor grade, offering clinicians a practical tool for personalized postoperative management.<\/jats:p><\/jats:sec>","DOI":"10.3389\/fimmu.2025.1681072","type":"journal-article","created":{"date-parts":[[2025,9,15]],"date-time":"2025-09-15T05:29:29Z","timestamp":1757914169000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":19,"title":["Integrating deep learning features from mammography with SHAP values for a machine learning model predicting over 5-year recurrence of breast ductal carcinoma In Situ post-lumpectomy"],"prefix":"10.3389","volume":"16","author":[{"given":"Yupeng","family":"Sha","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Quan","family":"Yuan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yi","family":"Du","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuqi","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming","family":"Niu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaoshuan","family":"Liang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shanshan","family":"Sun","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tong","family":"Li","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shu","family":"Gong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiguang","family":"Han","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2025,9,15]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-025-03502-3","article-title":"Global patterns and trends in breast cancer incidence and mortality across 185 countries","volume":"31","author":"Kim","year":"2025","journal-title":"Nat Med"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejso.2021.10.030","article-title":"Advances and controversies in management of breast ductal carcinoma in situ (DCIS)","volume":"48","author":"Farante","year":"2022","journal-title":"Eur J Surg Oncol"},{"key":"B3","doi-asserted-by":"publisher","DOI":"10.1200\/JCO.2014.56.2595","article-title":"Effect of radiotherapy after breast-Conserving surgery for ductal carcinoma in situ: 20 years follow-Up in the randomized sweDCIS trial","volume":"32","author":"W\u00e4rnberg","year":"2014","journal-title":"J Clin Oncol"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.1016\/j.breast.2022.01.001","article-title":"Unresected screen-detected ductal carcinoma in situ: Outcomes of 311 women in the Forget-Me-Not 2 study","volume":"61","author":"Maxwell","year":"2022","journal-title":"Breast"},{"key":"B5","doi-asserted-by":"publisher","first-page":"405","DOI":"10.1186\/s12885-015-1419-2","article-title":"Opportunistic mammography screening provides effective detection rates in a limited resource healthcare system","volume":"15","author":"Teh","year":"2015","journal-title":"BMC Cancer"},{"key":"B6","doi-asserted-by":"publisher","DOI":"10.1148\/radiology.170.2.2536185","article-title":"Ductal carcinoma in situ: mammographic findings and clinical implications","volume":"170","author":"Dershaw","year":"1989","journal-title":"Radiology"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1148\/radiol.211839","article-title":"Ductal carcinoma in situ: state-of-the-art review","volume":"302","author":"Grimm","year":"2022","journal-title":"Radiology"},{"key":"B8","doi-asserted-by":"publisher","first-page":"109","DOI":"10.1186\/s13244-025-01983-x","article-title":"Mammography-based artificial intelligence for breast cancer detection, diagnosis, and BI-RADS categorization using multi-view and multi-level convolutional neural networks","volume":"16","author":"Tan","year":"2025","journal-title":"Insights into Imaging"},{"key":"B9","doi-asserted-by":"publisher","first-page":"3113","DOI":"10.3390\/diagnostics13193113","article-title":"Breast cancer detection and prevention using machine learning","volume":"13","author":"Khalid","year":"2023","journal-title":"Diagnostics"},{"key":"B10","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1186\/s40644-024-00691-x","article-title":"Application of deep learning on mammographies to discriminate between low and high-risk DCIS for patient participation in active surveillance trials","volume":"24","author":"Alaeikhanehshir","year":"2024","journal-title":"Cancer Imaging"},{"key":"B11","doi-asserted-by":"publisher","DOI":"10.1111\/cts.70056","article-title":"Practical guide to SHAP analysis: Explaining supervised machine learning model predictions in drug development","volume":"17","author":"Ponce-Bobadilla","year":"2024","journal-title":"Clin Trans Sci"},{"key":"B12","doi-asserted-by":"publisher","first-page":"107","DOI":"10.1186\/s12911-023-02171-x","article-title":"Construction of the XGBoost model for early lung cancer prediction based on metabolic indices","volume":"23","author":"Guan","year":"2023","journal-title":"BMC Med Inf Decision Making"},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.18383\/j.tom.2019.00026","article-title":"A fully automated deep learning network for brain tumor segmentation","volume":"6","author":"Yogananda","year":"2020","journal-title":"Tomography"},{"key":"B14","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1186\/s12859-021-03968-1","article-title":"Penalized partial least squares for pleiotropy","volume":"22","author":"Broc","year":"2021","journal-title":"BMC Bioinf"},{"key":"B15","doi-asserted-by":"publisher","DOI":"10.1016\/j.gpb.2018.08.004","article-title":"Integration of A deep learning classifier with A random forest approach for predicting malonylation sites","volume":"16","author":"Chen","year":"2019","journal-title":"Genomics Proteomics Bioinf"},{"key":"B16","doi-asserted-by":"publisher","DOI":"10.1016\/j.csbj.2022.03.035","article-title":"Machine learning application in personalised lung cancer recurrence and survivability prediction","volume":"20","author":"Yang","year":"2022","journal-title":"Comput Struct Biotechnol J"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.1213\/ANE.0000000000003424","article-title":"Regression: the apple does not fall far from the tree","volume":"127","author":"Vetter","year":"2018","journal-title":"Anesth Analgesia"},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.1109\/TCYB.2020.3031610","article-title":"Fuzzy KNN method with adaptive nearest neighbors","volume":"52","author":"Bian","year":"2022","journal-title":"IEEE Trans Cybernetics"},{"key":"B19","doi-asserted-by":"publisher","first-page":"103761","DOI":"10.1016\/j.compbiomed.2020.103761","article-title":"Diagnostic classification of cancers using extreme gradient boosting algorithm and multi-omics data","volume":"121","author":"Ma","year":"2020","journal-title":"Comput Biol Med"},{"key":"B20","doi-asserted-by":"publisher","first-page":"2394","DOI":"10.3390\/antiox11122394","article-title":"Gradient boosting machine identified predictive variables for breast cancer patients pre- and post-Radiotherapy: preliminary results of an 8-Year follow-Up study","volume":"11","author":"Rodr\u00edguez-Tom\u00e0s","year":"2022","journal-title":"Antioxidants"},{"key":"B21","doi-asserted-by":"publisher","DOI":"10.21037\/qims-24-428","article-title":"Preoperatively predicting human epidermal growth factor receptor 2-low expression in breast cancer using neural network model based on multiparameter magnetic resonance imaging","volume":"14","author":"Zhao","year":"2024","journal-title":"Quantitative Imaging Med Surg"},{"key":"B22","doi-asserted-by":"publisher","first-page":"7505","DOI":"10.1038\/s41598-025-92080-y","article-title":"Integrating radiomics into predictive models for low nuclear grade DCIS using machine learning","volume":"15","author":"Wu","year":"2025","journal-title":"Sci Rep"},{"key":"B23","doi-asserted-by":"publisher","first-page":"316","DOI":"10.1186\/s12911-022-02062-7","article-title":"Facial biotype classification for orthodontic treatment planning using an alternative learning algorithm for tree augmented Naive Bayes","volume":"22","author":"Ruz","year":"2022","journal-title":"BMC Med Inform Decis Mak"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1080\/1062936X.2022.2160011","article-title":"iACP-GE: accurate identification of anticancer peptides by using gradient boosting decision tree and extra tree","volume":"34","author":"Liang","year":"2023","journal-title":"SAR QSAR Environ Res"},{"key":"B25","doi-asserted-by":"publisher","first-page":"4275","DOI":"10.1038\/s41598-023-31288-2","article-title":"IntelliSleepScorer, a software package with a graphic user interface for automated sleep stage scoring in mice based on a light gradient boosting machine algorithm","volume":"13","author":"Wang","year":"2023","journal-title":"Sci Rep"},{"key":"B26","doi-asserted-by":"publisher","first-page":"345","DOI":"10.1186\/s12911-022-02087-y","article-title":"Application of machine learning techniques for predicting survival in ovarian cancer","volume":"22","author":"Sorayaie Azar","year":"2022","journal-title":"BMC Med Inform Decis Mak"},{"key":"B27","doi-asserted-by":"publisher","DOI":"10.1097\/JS9.0000000000000026","article-title":"The use of multilayer perceptron and radial basis function: an artificial intelligence model to predict progression of oral cancer","volume":"109","author":"Jayaram","year":"2023","journal-title":"Int J Surg"},{"key":"B28","doi-asserted-by":"publisher","DOI":"10.21037\/qims-21-39","article-title":"Application of machine learning with multiparametric dual-energy computed tomography of the breast to differentiate between benign and Malignant lesions","volume":"12","author":"Lan","year":"2021","journal-title":"Quantitative Imaging Med Surg"},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.4258\/hir.2023.29.3.228","article-title":"Understanding arteriosclerotic heart disease patients using electronic health records: A machine learning and shapley additive exPlanations approach","volume":"29","author":"Miranda","year":"2023","journal-title":"Healthc Inform Res"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.3389\/fcvm.2021.736491","article-title":"A machine learning-based prediction model for cardiovascular risk in women with preeclampsia","volume":"8","author":"Wang","year":"2021","journal-title":"Front Cardiovasc Med"},{"key":"B31","doi-asserted-by":"publisher","DOI":"10.1007\/s10549-015-3464-6","article-title":"A population-based validation study of the DCIS Score predicting recurrence risk in individuals treated by breast-conserving surgery alone","volume":"152","author":"Rakovitch","year":"2015","journal-title":"Breast Cancer Res Treat"},{"key":"B32","doi-asserted-by":"publisher","first-page":"82","DOI":"10.1186\/s13058-024-01830-9","article-title":"Deep learning of mammogram images to reduce unnecessary breast biopsies: a preliminary study","volume":"26","author":"Liu","year":"2024","journal-title":"Breast Cancer Res"},{"key":"B33","doi-asserted-by":"publisher","DOI":"10.1038\/s41374-021-00540-6","article-title":"Deep learning-based grading of ductal carcinoma in situ in breast histopathology images","volume":"101","author":"Wetstein","year":"2021","journal-title":"Lab Invest"},{"key":"B34","doi-asserted-by":"publisher","first-page":"288","DOI":"10.1186\/s13054-021-03724-0","article-title":"Machine learning model for early prediction of acute kidney injury (AKI) in pediatric critical care","volume":"25","author":"Dong","year":"2021","journal-title":"Crit Care"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.1161\/STROKEAHA.118.024293","article-title":"Machine learning\u2013based model for prediction of outcomes in acute stroke","volume":"50","author":"Heo","year":"2019","journal-title":"Stroke"},{"key":"B36","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1186\/s13014-018-0964-7","article-title":"Role of postoperative radiotherapy in reducing ipsilateral recurrence in DCIS: an observational study of 1048 cases","volume":"13","author":"Corradini","year":"2018","journal-title":"Radiat Oncol"},{"key":"B37","doi-asserted-by":"publisher","DOI":"10.1038\/s41416-019-0478-6","article-title":"Ductal carcinoma in situ: to treat or not to treat, that is the question","volume":"121","author":"van Seijen","year":"2019","journal-title":"Br J Cancer"}],"container-title":["Frontiers in Immunology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fimmu.2025.1681072\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,15]],"date-time":"2025-09-15T05:29:30Z","timestamp":1757914170000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fimmu.2025.1681072\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,15]]},"references-count":37,"alternative-id":["10.3389\/fimmu.2025.1681072"],"URL":"https:\/\/doi.org\/10.3389\/fimmu.2025.1681072","relation":{},"ISSN":["1664-3224"],"issn-type":[{"value":"1664-3224","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,9,15]]},"article-number":"1681072"}}