{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T17:01:54Z","timestamp":1761238914052,"version":"3.41.2"},"reference-count":28,"publisher":"Wiley","issue":"1","license":[{"start":{"date-parts":[[2021,9,8]],"date-time":"2021-09-08T00:00:00Z","timestamp":1631059200000},"content-version":"vor","delay-in-days":250,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100004610","name":"Science and Technology Support Program of Jiangsu Province","doi-asserted-by":"publisher","award":["BE2018646"],"award-info":[{"award-number":["BE2018646"]}],"id":[{"id":"10.13039\/501100004610","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Wireless Communications and Mobile Computing"],"published-print":{"date-parts":[[2021,1]]},"abstract":"<jats:p>To determine the feasibility of using a deep learning (DL) approach to identify benign and malignant BI\u2010RADS 4 lesions with preoperative breast DCE\u2010MRI images and compare two 3D segmentation methods. The patients admitted from January 2014 to October 2020 were retrospectively analyzed. Breast MRI examination was performed before surgical resection or biopsy, and the masses were classified as BI\u2010RADS 4. The first postcontrast images of DCE\u2010MRI T1WI sequence were selected. There were two 3D segmentation methods for the lesions, one was manual segmentation along the edge of the lesion slice by slice, and the other was the minimum bounding cube of the lesion. Then, DL feature extraction was carried out; the pixel values of the image data are normalized to 0\u20101 range. The model was established based on the blueprint of the classic residual network ResNet50, retaining its residual module and improved 2D convolution module to 3D. At the same time, an attention mechanism was added to transform the attention mechanism module, which only fit the 2D image convolution module, into a 3D\u2010Convolutional Block Attention Module (CBAM) to adapt to 3D\u2010MRI. After the last CBAM, the algorithm stretches the output high\u2010dimensional features into a one\u2010dimensional vector and connects 2 fully connected slices, before finally setting two output results (P1, P2), which, respectively, represent the probability of benign and malignant lesions. Accuracy, sensitivity, specificity, negative predictive value, positive predictive value, the recall rate and area under the ROC curve (AUC) were used as evaluation indicators. A total of 203 patients were enrolled, with 207 mass lesions including 101 benign lesions and 106 malignant lesions. The data set was divided into the training set (<jats:italic>n<\/jats:italic> = 145), the validation set (<jats:italic>n<\/jats:italic> = 22), and the test set (<jats:italic>n<\/jats:italic> = 40) at the ratio of 7\u2009:\u20091\u2009:\u20092; fivefold cross\u2010validation was performed. The mean AUC based on the minimum bounding cube of lesion and the 3D\u2010ROI of lesion itself were 0.827 and 0.799, the accuracy was 78.54% and 74.63%, the sensitivity was 78.85% and 83.65%, the specificity was 78.22% and 65.35%, the NPV was 78.85% and 71.31%, the PPV was 78.22% and 79.52%, the recall rate was 78.85% and 83.65%, respectively. There was no statistical difference in AUC based on the lesion itself model and the minimum bounding cube model (<jats:italic>Z<\/jats:italic> = 0.771, <jats:italic>p<\/jats:italic> = 0.4408). The minimum bounding cube based on the edge of the lesion showed higher accuracy, specificity, and lower recall rate in identifying benign and malignant lesions. Based on the lesion 3D\u2010ROI segmentation using a minimum bounding cube can more effectively reflect the information of the lesion itself and the surrounding tissues. Its DL model performs better than the lesion itself. Using the DL approach with a 3D attention mechanism based on ResNet50 to identify benign and malignant BI\u2010RADS 4 lesions was feasible.<\/jats:p>","DOI":"10.1155\/2021\/4430886","type":"journal-article","created":{"date-parts":[[2021,9,8]],"date-time":"2021-09-08T18:50:09Z","timestamp":1631127009000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":7,"title":["Feasibility of Using Improved Convolutional Neural Network to Classify BI\u2010RADS 4 Breast Lesions: Compare Deep Learning Features of the Lesion Itself and the Minimum Bounding Cube of Lesion"],"prefix":"10.1155","volume":"2021","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-7927-8886","authenticated-orcid":false,"given":"Meihong","family":"Sheng","sequence":"first","affiliation":[]},{"given":"Weixia","family":"Tang","sequence":"additional","affiliation":[]},{"given":"Jiahuan","family":"Tang","sequence":"additional","affiliation":[]},{"given":"Ming","family":"Zhang","sequence":"additional","affiliation":[]},{"given":"Shenchu","family":"Gong","sequence":"additional","affiliation":[]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2140-2685","authenticated-orcid":false,"given":"Wei","family":"Xing","sequence":"additional","affiliation":[]}],"member":"311","published-online":{"date-parts":[[2021,9,8]]},"reference":[{"doi-asserted-by":"publisher","key":"e_1_2_13_1_2","DOI":"10.3322\/caac.21660"},{"doi-asserted-by":"publisher","key":"e_1_2_13_2_2","DOI":"10.1136\/bmj.f2618"},{"doi-asserted-by":"publisher","key":"e_1_2_13_3_2","DOI":"10.1056\/NEJMsr1504363"},{"doi-asserted-by":"publisher","key":"e_1_2_13_4_2","DOI":"10.1016\/S0140-6736(12)61611-0"},{"doi-asserted-by":"publisher","key":"e_1_2_13_5_2","DOI":"10.1002\/jmri.26985"},{"volume-title":"ACR BI-RADS Magnetic Resonance Imaging, in: American College of Radiology, BI-RADS Committee, Editor. ACR BI-RADS Atlas: Breast Imaging Reporting and Data System","year":"2013","author":"Morris E. A.","key":"e_1_2_13_6_2"},{"doi-asserted-by":"publisher","key":"e_1_2_13_7_2","DOI":"10.2214\/AJR.16.16730"},{"doi-asserted-by":"publisher","key":"e_1_2_13_8_2","DOI":"10.2214\/AJR.14.13834"},{"doi-asserted-by":"publisher","key":"e_1_2_13_9_2","DOI":"10.1002\/cncr.22435"},{"doi-asserted-by":"publisher","key":"e_1_2_13_10_2","DOI":"10.1016\/j.compmedimag.2007.02.002"},{"doi-asserted-by":"publisher","key":"e_1_2_13_11_2","DOI":"10.1016\/j.inffus.2020.09.002"},{"doi-asserted-by":"publisher","key":"e_1_2_13_12_2","DOI":"10.3389\/fnins.2020.00496"},{"doi-asserted-by":"publisher","key":"e_1_2_13_13_2","DOI":"10.1016\/j.knosys.2019.105330"},{"doi-asserted-by":"publisher","key":"e_1_2_13_14_2","DOI":"10.1038\/s41568-018-0016-5"},{"doi-asserted-by":"publisher","key":"e_1_2_13_15_2","DOI":"10.1007\/s10620-018-5316-9"},{"doi-asserted-by":"publisher","key":"e_1_2_13_16_2","DOI":"10.3390\/medicina57040395"},{"key":"e_1_2_13_17_2","first-page":"1097","article-title":"Imagenet classification with deep convolutional neural networks","volume":"25","author":"Krizhevsky A.","year":"2012","journal-title":"Advances in Neural Information Processing Systems"},{"doi-asserted-by":"crossref","unstructured":"WooS. ParkJ. LeeJ. Y. andKweonI. S. CBAM: convolutional block attention module European Conference on Computer Vision (ECCV) 2018 3\u201319.","key":"e_1_2_13_18_2","DOI":"10.1007\/978-3-030-01234-2_1"},{"doi-asserted-by":"publisher","key":"e_1_2_13_19_2","DOI":"10.1007\/s10916-018-1088-1"},{"doi-asserted-by":"publisher","key":"e_1_2_13_20_2","DOI":"10.1016\/j.diii.2019.02.008"},{"doi-asserted-by":"publisher","key":"e_1_2_13_21_2","DOI":"10.1186\/s12880-017-0239-z"},{"key":"e_1_2_13_22_2","first-page":"190","article-title":"Differentiation between benign and malignant non-mass enhancement lesions using volumetric quantitative dynamic contrast-enhanced MR imaging","volume":"35","author":"Song C.","year":"2020","journal-title":"Radiology Practice"},{"doi-asserted-by":"publisher","key":"e_1_2_13_23_2","DOI":"10.1002\/jmri.26981"},{"doi-asserted-by":"publisher","key":"e_1_2_13_24_2","DOI":"10.1016\/j.pbiomolbio.2011.06.006"},{"doi-asserted-by":"publisher","key":"e_1_2_13_25_2","DOI":"10.2217\/fon-2016-0501"},{"doi-asserted-by":"publisher","key":"e_1_2_13_26_2","DOI":"10.1186\/s13058-017-0846-1"},{"key":"e_1_2_13_27_2","first-page":"2411","article-title":"Diagnostic value of T2WI relative signal intensity ratio combined with peritumoral edema in benign and malignant breast lesions","volume":"39","author":"Tang W. X.","year":"2020","journal-title":"Journal of Clinical Radiology"},{"key":"e_1_2_13_28_2","first-page":"324","article-title":"Value of early-phase enhancement ratio combined with peripheral vascular diameter in the differential diagnosis of benign and malignant breast lesions under dynamic contrast enhanced MRI","volume":"50","author":"Sheng M. H.","year":"2016","journal-title":"Chinese Journal of Radiology"}],"container-title":["Wireless Communications and Mobile Computing"],"original-title":[],"language":"en","link":[{"URL":"http:\/\/downloads.hindawi.com\/journals\/wcmc\/2021\/4430886.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"http:\/\/downloads.hindawi.com\/journals\/wcmc\/2021\/4430886.xml","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1155\/2021\/4430886","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,7]],"date-time":"2024-08-07T14:02:35Z","timestamp":1723039355000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1155\/2021\/4430886"}},"subtitle":[],"editor":[{"given":"Khin wee","family":"Lai","sequence":"additional","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2021,1]]},"references-count":28,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2021,1]]}},"alternative-id":["10.1155\/2021\/4430886"],"URL":"https:\/\/doi.org\/10.1155\/2021\/4430886","archive":["Portico"],"relation":{},"ISSN":["1530-8669","1530-8677"],"issn-type":[{"type":"print","value":"1530-8669"},{"type":"electronic","value":"1530-8677"}],"subject":[],"published":{"date-parts":[[2021,1]]},"assertion":[{"value":"2021-07-08","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-08-06","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2021-09-08","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"4430886"}}