{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,14]],"date-time":"2026-07-14T05:04:13Z","timestamp":1784005453791,"version":"3.55.0"},"reference-count":35,"publisher":"MDPI AG","issue":"14","license":[{"start":{"date-parts":[[2026,7,11]],"date-time":"2026-07-11T00:00:00Z","timestamp":1783728000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Cancers"],"abstract":"<jats:p>Background\/Objectives: Accurate visual categorization of small enhancing lesions on contrast-enhanced breast MRI remains challenging. The purpose of this study was to develop a deep learning (DL) model for detecting small (\u22642 cm) invasive breast cancers using multi-institutional breast MRI data. Methods: This retrospective study included a total of 1721 women (mean age: 54 years; range, 21\u201387 years) with T1-stage invasive breast cancer across five hospitals. All patients underwent dynamic contrast-enhanced breast MRI before surgery between 2010 and 2020. Enhancing breast masses on early postcontrast T1-weighted images (14,917 labeled cancer images and 1443 labeled benign images) were used as the ground truth. In this multicenter cancer-enriched cohort, a DL model was developed to distinguish cancerous lesions from noncancerous lesions. MR images were independently reviewed by three breast imaging radiologists to detect subcentimeter (\u22641 cm) cancers, and the readers were informed that each examination contained a single cancer. The performance of the model was evaluated using the area under the precision\u2013recall curve (AUPRC), sensitivity, precision, and F1 score. Results: The DL model achieved an AUPRC of 0.42 in detecting small (\u22642 cm) invasive breast cancers presenting as enhancing masses on MRI (sensitivity, 83.2%; precision, 33.2%; F1 score, 0.47). For subcentimeter (\u22641 cm) cancers, the detection performance of the model was lower (sensitivity, 75.3%; precision, 20.8%; F1 score, 0.33) than that of the radiologists (mean sensitivity, 89.5%; mean precision, 72.9%; mean F1 score, 0.80). When the radiologists used the DL model, their mean precision in detecting subcentimeter invasive breast cancers improved from 72.9% to 83.2%, with no significant change in sensitivity (89.5% vs. 86.8%, p &gt; 0.05). Conclusions: The DL model demonstrated potential as an assistive tool by improving radiologists\u2019 precision for detecting small invasive breast cancers on contrast-enhanced breast MRI in this enriched reader-study setting, although it did not significantly improve their sensitivity.<\/jats:p>","DOI":"10.3390\/cancers18142229","type":"journal-article","created":{"date-parts":[[2026,7,13]],"date-time":"2026-07-13T07:28:27Z","timestamp":1783927707000},"page":"2229","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Deep Learning for Early Breast Cancer Detection on Contrast-Enhanced Breast MRI: A Multicenter Study"],"prefix":"10.3390","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6128-5669","authenticated-orcid":false,"given":"Na Young","family":"Jung","sequence":"first","affiliation":[{"name":"Department of Radiology, Eunpyeong St. Mary\u2019s Hospital, College of Medicine, The Catholic University of Korea, Seoul 03312, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9658-831X","authenticated-orcid":false,"given":"Jihe","family":"Lim","sequence":"additional","affiliation":[{"name":"Department of Radiology, Hallym University Dongtan Sacred Heart Hospital, Hwaseong 18450, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hyug-Gi","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Radiology, Kyung Hee University Hospital, Seoul 02447, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4478-9720","authenticated-orcid":false,"given":"Sung Hun","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Radiology, Seoul St. Mary\u2019s Hospital, College of Medicine, The Catholic University of Korea, Seoul 06591, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bobae","family":"Choi","sequence":"additional","affiliation":[{"name":"Department of Radiology, Chungnam National University Hospital, Daejeon 35015, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bo La","family":"Yun","sequence":"additional","affiliation":[{"name":"Department of Radiology, Seoul National University Bundang Hospital, Seongnam 13620, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ga Eun","family":"Park","sequence":"additional","affiliation":[{"name":"Department of Radiology, Seoul St. Mary\u2019s Hospital, College of Medicine, The Catholic University of Korea, Seoul 06591, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4251-5470","authenticated-orcid":false,"given":"Jang-Hoon","family":"Oh","sequence":"additional","affiliation":[{"name":"Department of Radiology, Kyung Hee University Hospital, Seoul 02447, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9512-5361","authenticated-orcid":false,"given":"Bo Kyoung","family":"Seo","sequence":"additional","affiliation":[{"name":"Department of Radiology, Korea University Ansan Hospital, Ansan 15355, Republic of Korea"},{"name":"Department of Radiology, Korea University College of Medicine, Seoul 02841, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1971-3428","authenticated-orcid":false,"given":"Min Sun","family":"Bae","sequence":"additional","affiliation":[{"name":"Department of Radiology, Korea University Ansan Hospital, Ansan 15355, Republic of Korea"},{"name":"Department of Radiology, Korea University College of Medicine, Seoul 02841, Republic of Korea"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,7,11]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"520","DOI":"10.1148\/radiol.2019182947","article-title":"Breast MRI: State of the art","volume":"292","author":"Mann","year":"2019","journal-title":"Radiology"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"2091","DOI":"10.1056\/NEJMoa1903986","article-title":"Supplemental MRI screening for women with extremely dense breast tissue","volume":"381","author":"Bakker","year":"2019","journal-title":"N. Engl. J. Med."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"746","DOI":"10.1001\/jama.2020.0572","article-title":"Comparison of abbreviated breast MRI vs. digital breast tomosynthesis for breast cancer detection among women with dense breasts undergoing screening","volume":"323","author":"Comstock","year":"2020","journal-title":"JAMA"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1394","DOI":"10.1001\/jama.2012.388","article-title":"Detection of breast cancer with addition of annual screening ultrasound or a single screening MRI to mammography in women with elevated breast cancer risk","volume":"307","author":"Berg","year":"2012","journal-title":"JAMA"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"e221822","DOI":"10.1148\/radiol.221822","article-title":"Abbreviated breast MRI: State of the art","volume":"310","author":"Kuhl","year":"2024","journal-title":"Radiology"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"e221785","DOI":"10.1148\/radiol.221785","article-title":"Supplemental breast cancer screening in women with dense breasts and negative mammography: A systematic review and meta-analysis","volume":"306","author":"Hussein","year":"2023","journal-title":"Radiology"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"716","DOI":"10.1148\/radiol.2016151419","article-title":"Breast cancers detected at screening MR imaging and mammography in patients at high risk: Method of detection reflects tumor histopathologic results","volume":"280","author":"Sung","year":"2016","journal-title":"Radiology"},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"20170102","DOI":"10.1259\/bjr.20170102","article-title":"MRI appearance of invasive subcentimetre breast carcinoma: Benign characteristics are common","volume":"90","author":"Meissnitzer","year":"2017","journal-title":"Br. J. Radiol."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"e241629","DOI":"10.1148\/radiol.241629","article-title":"Cancer detection in breast MRI screening via explainable AI anomaly detection","volume":"316","author":"Oviedo","year":"2025","journal-title":"Radiology"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"38","DOI":"10.1148\/radiol.2020200292","article-title":"Artificial intelligence applied to breast MRI for improved diagnosis","volume":"298","author":"Jiang","year":"2021","journal-title":"Radiology"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"eabo4802","DOI":"10.1126\/scitranslmed.abo4802","article-title":"Improving breast cancer diagnostics with deep learning for MRI","volume":"14","author":"Witowski","year":"2022","journal-title":"Sci. Transl. Med."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"87","DOI":"10.1186\/s13058-023-01687-4","article-title":"Deep learning applications to breast cancer detection by magnetic resonance imaging: A literature review","volume":"25","author":"Adam","year":"2023","journal-title":"Breast Cancer Res."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"e222639","DOI":"10.1148\/radiol.222639","article-title":"Standalone AI for breast cancer detection at screening digital mammography and digital breast tomosynthesis: A systematic review and meta-analysis","volume":"307","author":"Yoon","year":"2023","journal-title":"Radiology"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1016\/j.rcl.2020.08.007","article-title":"Applications of Artificial Intelligence in Breast Imaging","volume":"59","author":"Morgan","year":"2021","journal-title":"Radiol. Clin. N. Am."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2290","DOI":"10.1002\/jmri.29358","article-title":"AI applications to breast MRI: Today and tomorrow","volume":"60","author":"Brunekreef","year":"2024","journal-title":"J. Magn. Reson. Imaging"},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"290","DOI":"10.1148\/radiol.2018181352","article-title":"Radiomic versus convolutional neural networks analysis for classification of contrast-enhancing lesions at multiparametric breast MRI","volume":"290","author":"Truhn","year":"2019","journal-title":"Radiology"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"e200231","DOI":"10.1148\/ryai.200231","article-title":"Radiologist-level performance by using deep learning for segmentation of breast cancers on MRI scans","volume":"4","author":"Hirsch","year":"2021","journal-title":"Radiol. Artif. Intell."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"29","DOI":"10.1148\/radiol.2021203960","article-title":"Deep learning for automated triaging of 4581 breast MRI examinations from the DENSE trial","volume":"302","author":"Verburg","year":"2022","journal-title":"Radiology"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"4361","DOI":"10.2217\/fon-2022-0593","article-title":"Detection and classification of breast lesions with You Only Look Once version 5","volume":"18","author":"Meng","year":"2022","journal-title":"Future Oncol."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"2084","DOI":"10.1007\/s00330-023-10184-3","article-title":"Localization of contrast-enhanced breast lesions in ultrafast screening MRI using deep convolutional neural networks","volume":"34","author":"Jing","year":"2024","journal-title":"Eur. Radiol."},{"key":"ref_21","unstructured":"Morris, E.A., Comstock, C.E., and Lee, C.H. (2013). ACR BI-RADS magnetic resonance imaging. ACR BI-RADS Atlas, Breast Imaging Reporting and Data System, American College of Radiology."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"99","DOI":"10.1016\/j.rcl.2020.09.001","article-title":"Abbreviated MR imaging for breast cancer","volume":"59","author":"Heacock","year":"2021","journal-title":"Radiol. Clin. N. Am."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"85","DOI":"10.1016\/j.rcl.2020.09.004","article-title":"Magnetic resonance imaging in screening of breast cancer","volume":"59","author":"Gao","year":"2021","journal-title":"Radiol. Clin. N. Am."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"687","DOI":"10.1016\/j.rcl.2023.12.010","article-title":"Auditing abbreviated breast MR imaging: Clinical considerations and implications","volume":"62","author":"Seely","year":"2024","journal-title":"Radiol. Clin. N. Am."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"73","DOI":"10.1148\/radiol.2021202927","article-title":"Breast cancer screening with abbreviated breast MRI: 3-year outcome analysis","volume":"299","author":"Kwon","year":"2021","journal-title":"Radiology"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"412","DOI":"10.1016\/j.acra.2011.12.014","article-title":"Small masses on breast MR: Is biopsy necessary?","volume":"19","author":"Raza","year":"2012","journal-title":"Acad. Radiol."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"6721","DOI":"10.1007\/s00330-020-06991-7","article-title":"Improved characterization of sub-centimeter enhancing breast masses on MRI with radiomics and machine learning in BRCA mutation carriers","volume":"30","author":"Daimiel","year":"2020","journal-title":"Eur. Radiol."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"5897","DOI":"10.1002\/mp.15156","article-title":"Automatic breast lesion detection in ultrafast DCE-MRI using deep learning","volume":"48","author":"Ayatollahi","year":"2021","journal-title":"Med. Phys."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"219","DOI":"10.1016\/j.diii.2019.02.008","article-title":"Detection and characterization of MRI breast lesions using deep learning","volume":"100","author":"Herent","year":"2019","journal-title":"Diagn. Interv. Imaging"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"8706","DOI":"10.1007\/s00330-022-08863-8","article-title":"Using deep learning to safely exclude lesions with only ultrafast breast MRI to shorten acquisition and reading time","volume":"32","author":"Jing","year":"2022","journal-title":"Eur. Radiol."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"S161","DOI":"10.1016\/j.acra.2022.12.038","article-title":"Deep learning-based automatic diagnosis of breast cancer on MRI using Mask R-CNN for detection followed by ResNet50 for classification","volume":"30","author":"Zhang","year":"2023","journal-title":"Acad. Radiol."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"10536","DOI":"10.1038\/s41598-020-67441-4","article-title":"A deep learning methodology for improved breast cancer diagnosis using multiparametric MRI","volume":"10","author":"Hu","year":"2020","journal-title":"Sci. Rep."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1158","DOI":"10.1159\/000510001","article-title":"Metastases from occult breast cancer: A case report of carcinoma of unknown primary syndrome","volume":"13","author":"Barbieri","year":"2020","journal-title":"Case Rep. Oncol."},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Wu, T., Huang, Q., Liu, Z., Wang, Y., and Lin, D. (2020). Distribution-balanced loss for multi-label classification in long-tailed datasets. Computer Vision, Proceedings of the European Conference on Computer Vision (ECCV) 2020: Proceedings, Part IV, Springer.","DOI":"10.1007\/978-3-030-58548-8_10"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"249","DOI":"10.1016\/j.neunet.2018.07.011","article-title":"A systematic study of the class imbalance problem in convolutional neural networks","volume":"106","author":"Buda","year":"2018","journal-title":"Neural Netw."}],"container-title":["Cancers"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-6694\/18\/14\/2229\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,14]],"date-time":"2026-07-14T04:13:13Z","timestamp":1784002393000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-6694\/18\/14\/2229"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,11]]},"references-count":35,"journal-issue":{"issue":"14","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["cancers18142229"],"URL":"https:\/\/doi.org\/10.3390\/cancers18142229","relation":{},"ISSN":["2072-6694"],"issn-type":[{"value":"2072-6694","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,11]]}}}