{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T06:15:43Z","timestamp":1788156943000,"version":"build-2803163510"},"reference-count":33,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T00:00:00Z","timestamp":1788134400000},"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. Radiol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Molecular subtyping of breast cancer, particularly the discrimination between Luminal A and Basal-like tumours, is critical for guiding treatment decisions and establishing prognosis. Radiomics offers a promising complementary, non-invasive imaging biomarker to support biopsy-based subtyping; however, two methodological gaps limit the validity of existing studies: the absence of controlled 2D-vs.-3D feature comparisons, and the systematic underreporting of patient-level data leakage in multi-slice pipelines.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>\n                      To address both limitations, this study benchmarks slice-level (2D) and volumetric (3D) radiomic pipelines for the classification of Luminal A vs. Basal-like tumours on the TCGA-BRCA cohort, evaluating seven classifiers combined with two feature selection strategies under a Stratified Group\n                      <jats:italic>k<\/jats:italic>\n                      -Fold (\n                      <jats:italic>k<\/jats:italic>\n                      \u2009=\u20095) cross-validation scheme that enforces strict patient-level isolation.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      The 3D pipeline achieved a mean AUC of 0.682 vs. 0.631 for the 2D pipeline. In the 3D configuration, the Support Vector Machine with SelectKBest-MI feature selection attained the highest performance (AUC\u2009=\u20090.768\u2009\u00b1\u20090.066), while the 2D configuration was led by Logistic Regression (AUC\u2009=\u20090.769\u2009\u00b1\u20090.109), albeit with substantially higher fold-to-fold variability. Six out of seven classifiers yielded higher AUC scores under the 3D configuration, although a paired significance test on fold-level AUC values did not reach statistical significance (\n                      <jats:italic>p<\/jats:italic>\n                      \u2009=\u20090.313).\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These results establish a reproducible, leakage-free baseline for non-invasive breast cancer subtype classification through radiomics, with 3D radiomics offering improved robustness rather than superior peak performance. Results are derived from a single cohort (TCGA-BRCA) and external validation is required before clinical generalizability can be claimed.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fradi.2026.1915426","type":"journal-article","created":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T05:40:24Z","timestamp":1788154824000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Slice-level vs. volumetric radiomics for molecular subtyping of breast cancer: impact of patient-level data partitioning on model performance"],"prefix":"10.3389","volume":"6","author":[{"given":"Laila","family":"El Jiani","sequence":"first","affiliation":[{"name":"Laboratory of Artificial Intelligence and Systems, Faculty of Science Ben M\u2019sick, Hassan II University","place":["Casablanca, Morocco"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zouheir","family":"Banou","sequence":"additional","affiliation":[{"name":"Laboratory of Artificial Intelligence and Systems, Faculty of Science Ben M\u2019sick, Hassan II University","place":["Casablanca, Morocco"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fatima Zahra","family":"Alaoui","sequence":"additional","affiliation":[{"name":"Laboratory of Artificial Intelligence and Systems, Faculty of Science Ben M\u2019sick, Hassan II University","place":["Casablanca, Morocco"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hasnae","family":"Sakhi","sequence":"additional","affiliation":[{"name":"Laboratory of Artificial Intelligence and Systems, Faculty of Science Ben M\u2019sick, Hassan II University","place":["Casablanca, Morocco"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"El Habib","family":"Benlahmar","sequence":"additional","affiliation":[{"name":"Laboratory of Information Processing and Modeling, Faculty of Science Ben M\u2019sick, Hassan II University","place":["Casablanca, Morocco"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Sanaa","family":"El Filali","sequence":"additional","affiliation":[{"name":"Laboratory of Artificial Intelligence and Systems, Faculty of Science Ben M\u2019sick, Hassan II University","place":["Casablanca, Morocco"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,8,31]]},"reference":[{"key":"B1","year":""},{"key":"B2","doi-asserted-by":"publisher","first-page":"8","DOI":"10.1038\/s41523-020-0151-5","article-title":"The therapeutic response of ER+\/HER2\u2212 breast cancers differs according to the molecular Basal or Luminal subtype","volume":"6","author":"Bertucci","year":"2020","journal-title":"npj Breast Cancer"},{"key":"B3","doi-asserted-by":"publisher","first-page":"1281","DOI":"10.7150\/jca.13141","article-title":"Cancer hallmarks, biomarkers and breast cancer molecular subtypes","volume":"7","author":"Dai","year":"2016","journal-title":"J Cancer"},{"key":"B4","doi-asserted-by":"publisher","first-page":"686","DOI":"10.1309\/AJCPT41TCBUEVDQC","article-title":"Advantages and disadvantages of technologies for HER2 testing in breast cancer specimens","volume":"144","author":"Furrer","year":"2015","journal-title":"Am J Clin Pathol"},{"key":"B5","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1148\/radiol.11102368","article-title":"Ductal carcinoma in situ at core-needle biopsy: meta-analysis of underestimation and predictors of invasive breast cancer","volume":"260","author":"Brennan","year":"2011","journal-title":"Radiology"},{"key":"B6","doi-asserted-by":"publisher","first-page":"12","DOI":"10.1186\/s13244-019-0803-x","article-title":"Image-guided breast biopsy and localisation: recommendations for information to women and referring physicians by the European society of breast imaging","volume":"11","author":"Bick","year":"2020","journal-title":"Insights Imaging"},{"key":"B7","doi-asserted-by":"publisher","first-page":"155","DOI":"10.1038\/s41523-021-00363-0","article-title":"Subclonal heterogeneity and evolution in breast cancer","volume":"7","author":"Mavrommati","year":"2021","journal-title":"npj Breast Cancer"},{"key":"B8","doi-asserted-by":"publisher","first-page":"883","DOI":"10.1056\/NEJMoa1113205","article-title":"Intratumor heterogeneity and branched evolution revealed by multiregion sequencing","volume":"366","author":"Gerlinger","year":"2012","journal-title":"N Engl J Med"},{"key":"B9","doi-asserted-by":"publisher","first-page":"1296","DOI":"10.1007\/s11547-021-01389-x","article-title":"A deep look into radiomics","volume":"126","author":"Scapicchio","year":"2021","journal-title":"Radiol Med"},{"key":"B10","doi-asserted-by":"publisher","first-page":"217","DOI":"10.1007\/s10549-018-4675-4","article-title":"Rapid review: radiomics and breast cancer","volume":"169","author":"Valdora","year":"2018","journal-title":"Breast Cancer Res Treat"},{"key":"B11","doi-asserted-by":"publisher","first-page":"172","DOI":"10.1186\/s13244-021-01115-1","article-title":"Measuring the bias of incorrect application of feature selection when using cross-validation in radiomics","volume":"12","author":"Demircio\u011flu","year":"2021","journal-title":"Insights Imaging"},{"key":"B12","doi-asserted-by":"publisher","first-page":"303","DOI":"10.3390\/jimaging8110303","article-title":"Harmonization strategies in multicenter MRI-based radiomics","volume":"8","author":"Stamoulou","year":"2022","journal-title":"J Imaging"},{"key":"B13","doi-asserted-by":"publisher","first-page":"984626","DOI":"10.3389\/fonc.2022.984626","article-title":"Deep learning-based automatic segmentation for size and volumetric measurement of breast cancer on magnetic resonance imaging","volume":"12","author":"Yue","year":"2022","journal-title":"Front Oncol"},{"key":"B14","author":"Wang","year":""},{"key":"B15","doi-asserted-by":"publisher","first-page":"365","DOI":"10.1148\/radiol.14132641","article-title":"Radiogenomic analysis of breast cancer: luminal B molecular subtype is associated with enhancement dynamics at MR imaging","volume":"273","author":"Mazurowski","year":"2014","journal-title":"Radiology"},{"key":"B16","doi-asserted-by":"publisher","first-page":"902","DOI":"10.1002\/jmri.24879","article-title":"Computational approach to radiogenomics of breast cancer: luminal A and luminal B molecular subtypes are associated with imaging features on routine breast MRI extracted using computer vision algorithms","volume":"42","author":"Grimm","year":"2015","journal-title":"J Magn Reson Imaging"},{"key":"B17","doi-asserted-by":"publisher","first-page":"508","DOI":"10.1038\/s41416-018-0185-8","article-title":"A machine learning approach to radiogenomics of breast cancer: a study of 922 subjects and 529 DCE-MRI features","volume":"119","author":"Saha","year":"2018","journal-title":"Br J Cancer"},{"key":"B18","doi-asserted-by":"publisher","first-page":"3944","DOI":"10.3390\/cancers14163944","article-title":"A simultaneous multiparametric 18F-FDG PET\/MRI radiomics model for the diagnosis of triple negative breast cancer","volume":"14","author":"Romeo","year":"2022","journal-title":"Cancers"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1853","DOI":"10.3390\/jcm9061853","article-title":"Non-invasive assessment of breast cancer molecular subtypes with multiparametric MRI radiomics","volume":"9","author":"Leithner","year":"2020","journal-title":"J Clin Med"},{"key":"B20","doi-asserted-by":"publisher","first-page":"127","DOI":"10.1186\/s13244-024-01713-9","article-title":"Molecular subtypes of breast cancer identified by dynamically enhanced MRI radiomics: the delayed phase cannot be ignored","volume":"15","author":"Huang","year":"2024","journal-title":"Insights Imaging"},{"key":"B21","doi-asserted-by":"publisher","first-page":"328","DOI":"10.1148\/radiol.2020191145","article-title":"The image biomarker standardization initiative: standardized quantitative radiomics for high-throughput image-based phenotyping","volume":"295","author":"Zwanenburg","year":"2020","journal-title":"Radiology"},{"key":"B22","doi-asserted-by":"publisher","first-page":"4006","DOI":"10.1038\/ncomms5006","article-title":"Decoding tumour phenotype by noninvasive imaging using a quantitative radiomics approach","volume":"5","author":"Aerts","year":"2014","journal-title":"Nat Commun"},{"key":"B23","doi-asserted-by":"publisher","first-page":"6334","DOI":"10.1002\/mp.14538","article-title":"Machine learning prediction of axillary lymph node metastasis in breast cancer: 2D versus 3D radiomic features","volume":"47","author":"Arefan","year":"2020","journal-title":"Med Phys"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1245","DOI":"10.3390\/electronics15061245","article-title":"Foundation models for volumetric medical imaging: opportunities, challenges, and future directions","volume":"15","author":"Ghosh","year":"2026","journal-title":"Electronics"},{"key":"B25","doi-asserted-by":"publisher","first-page":"2228","DOI":"10.1016\/j.acra.2023.12.005","article-title":"Contrast-enhanced mammography radiomics analysis for preoperative prediction of breast cancer molecular subtypes","volume":"31","author":"Zhu","year":"2024","journal-title":"Acad Radiol"},{"key":"B26","doi-asserted-by":"publisher","first-page":"650","DOI":"10.1007\/s00330-021-08146-8","article-title":"Radiomic machine learning for predicting prognostic biomarkers and molecular subtypes of breast cancer using tumor heterogeneity and angiogenesis properties on MRI","volume":"32","author":"Lee","year":"2022","journal-title":"Eur Radiol"},{"key":"B27","doi-asserted-by":"publisher","first-page":"16012","DOI":"10.1038\/npjbcancer.2016.12","article-title":"Quantitative MRI radiomics in the prediction of molecular classifications of breast cancer subtypes in the TCGA\/TCIA data set","volume":"2","author":"Li","year":"2016","journal-title":"npj Breast Cancer"},{"key":"B28","doi-asserted-by":"publisher","first-page":"17244","DOI":"10.1038\/s41598-022-20931-z","article-title":"Investigation of radiomics based intra-patient inter-tumor heterogeneity and the impact of tumor subsampling strategies","volume":"12","author":"Henry","year":"2022","journal-title":"Sci Rep"},{"key":"B29","doi-asserted-by":"publisher","first-page":"114249","DOI":"10.1016\/j.asoc.2025.114249","article-title":"On the value of uncertainty quantification in deep learning based breast cancer molecular subtype classification","volume":"186","author":"Tchokponhoue","year":"2026","journal-title":"Appl Soft Comput"},{"key":"B30","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1038\/nature11412","article-title":"Comprehensive molecular portraits of human breast tumours","volume":"490","year":"2012","journal-title":"Nature"},{"key":"B31","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1007\/978-3-032-14964-0_4","article-title":"Segment anything model for breast imaging segmentation: state-of-the-art","volume":"2645","author":"El Jiani","year":"2026","journal-title":"Commun Comput Inf Sci"},{"key":"B32","doi-asserted-by":"publisher","first-page":"e104","DOI":"10.1158\/0008-5472.CAN-17-0339","article-title":"Computational radiomics system to decode the radiographic phenotype","volume":"77","author":"van Griethuysen","year":"2017","journal-title":"Cancer Res"},{"key":"B33","doi-asserted-by":"publisher","first-page":"1121","DOI":"10.23939\/mmc2025.04.1121","article-title":"Blood vessel segmentation in breast MRI: comprehensive review of techniques and challenges","volume":"12","author":"El Jiani","year":"2025","journal-title":"Math Model Comput"}],"container-title":["Frontiers in Radiology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2026.1915426\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,31]],"date-time":"2026-08-31T05:40:25Z","timestamp":1788154825000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fradi.2026.1915426\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,31]]},"references-count":33,"alternative-id":["10.3389\/fradi.2026.1915426"],"URL":"https:\/\/doi.org\/10.3389\/fradi.2026.1915426","relation":{},"ISSN":["2673-8740"],"issn-type":[{"value":"2673-8740","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,31]]},"article-number":"1915426"}}