{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T18:50:43Z","timestamp":1782845443898,"version":"3.54.5"},"reference-count":26,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Form Res"],"abstract":"\n Background<\/jats:title>\n Breast implants, including textured variants, have been widely used in aesthetic and reconstructive mammoplasty. However, the textured type, which is one of the shell texture types of breast implants, has been identified as a possible etiologic factor for lymphoma, specifically breast implant\u2013associated anaplastic large cell lymphoma (BIA-ALCL). Identifying the shell texture type of the implant is critical to diagnosing BIA-ALCL. However, distinguishing the shell texture type can be difficult due to the loss of human memory and medical history. An alternative approach is to use ultrasonography, but this method also has limitations in quantitative assessment.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aims to determine the feasibility of using a deep learning model to classify the shell texture type of breast implants and make robust predictions from ultrasonography images from heterogeneous sources.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n A total of 19,502 breast implant images were retrospectively collected from heterogeneous sources, including images captured from both Canon and GE devices, images of ruptured implants, and images without implants, as well as publicly available images. The Canon images were trained using ResNet-50. The model\u2019s performance on the Canon dataset was evaluated using stratified 5-fold cross-validation. Additionally, external validation was conducted using the GE and publicly available datasets. The area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (PRAUC) were calculated based on the contribution of the pixels with Gradient-weighted Class Activation Mapping (Grad-CAM). To identify the significant pixels for classification, we masked the pixels that contributed less than 10%, up to a maximum of 100%. To assess the model\u2019s robustness to uncertainty, Shannon entropy was calculated for 4 image groups: Canon, GE, ruptured implants, and without implants.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n The deep learning model achieved an average AUROC of 0.98 and a PRAUC of 0.88 in the Canon dataset. The model achieved an AUROC of 0.985 and a PRAUC of 0.748 for images captured with GE devices. Additionally, the model predicted an AUROC of 0.909 and a PRAUC of 0.958 for the publicly available dataset. This model maintained the PRAUC values for quantitative validation when masking up to 90% of the least-contributing pixels and the remnant pixels in breast shell layers. Furthermore, the prediction uncertainty increased in the following order: Canon (0.066), GE (0072), ruptured implants (0.371), and no implants (0.777).<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n We have demonstrated the feasibility of using deep learning to predict the shell texture type of breast implants. This approach quantifies the shell texture types of breast implants, supporting the first step in the diagnosis of BIA-ALCL.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/58776","type":"journal-article","created":{"date-parts":[[2024,8,6]],"date-time":"2024-08-06T05:55:10Z","timestamp":1722923710000},"page":"e58776","source":"Crossref","is-referenced-by-count":0,"title":["A Deep Learning Model to Predict Breast Implant Texture Types Using Ultrasonography Images: Feasibility Development Study"],"prefix":"10.2196","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-7260-7504","authenticated-orcid":false,"given":"Ho Heon","family":"Kim","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-1931-5957","authenticated-orcid":false,"given":"Won Chan","family":"Jeong","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-8797-8256","authenticated-orcid":false,"given":"Kyungran","family":"Pi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4040-1557","authenticated-orcid":false,"given":"Angela Soeun","family":"Lee","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3230-6309","authenticated-orcid":false,"given":"Min Soo","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0186-5674","authenticated-orcid":false,"given":"Hye Jin","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7162-9944","authenticated-orcid":false,"given":"Jae Hong","family":"Kim","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2024,11,5]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jmbbm.2018.08.035"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1097\/01.prs.0000218184.47372.d5"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1097\/00006534-199708000-00065"},{"key":"ref4","unstructured":"Medical device reports of breast implant-associated anaplastic large cell lymphomaUS Food and Drug Administration202312152024-03-06https:\/\/www.fda.gov\/medical-devices\/breast-implants\/medical-device-reports-breast-implant-associated-anaplastic-large-cell-lymphoma"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1097\/PRS.0000000000005570"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1093\/asj\/sjy331"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1136\/bcr-2013-201950"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1097\/PRS.0000000000001033"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1097\/PRS.0000000000003282"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.21037\/gs-20-853"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1016\/j.ctrv.2020.101963"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1093\/asj\/sjac312"},{"key":"ref13","unstructured":"SeongJKLoss of data from 13,000 patients receiving a breast implantHit News202010132024-10-04http:\/\/www.hitnews.co.kr\/news\/articleView.html?idxno=30265"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1097\/PRS.0000000000010201"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.3390\/medicina59061092"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr.2016.90"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/iccv48922.2021.00986"},{"key":"ref18","unstructured":"DosovitskiyABeyerLKolesnikovAWeissenbornDZhaiXUnterthinerTAn image is worth 16x16 words: transformers for image recognition at scale2021ICLR 2021 - 9th International Conference on Learning RepresentationsMay 3-7, 2021Virtual event"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2023.47692"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-019-01228-7"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1002\/j.1538-7305.1948.tb01338.x"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpra.2022.02.004"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.3390\/biomedinformatics2010001"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0276501"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1148\/ryai.210064"},{"key":"ref26","unstructured":"Score-CAM: score-weighted visual explanations for convolutional neural networksGitHub2024-09-05https:\/\/github.com\/haofanwang\/Score-CAM"}],"container-title":["JMIR Formative Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/formative.jmir.org\/2024\/1\/e58776","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:50:57Z","timestamp":1782841857000},"score":1,"resource":{"primary":{"URL":"https:\/\/formative.jmir.org\/2024\/1\/e58776"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,11,5]]},"references-count":26,"URL":"https:\/\/doi.org\/10.2196\/58776","relation":{"has-preprint":[{"id-type":"doi","id":"10.2196\/preprints.58776","asserted-by":"object"}]},"ISSN":["2561-326X"],"issn-type":[{"value":"2561-326X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,11,5]]}}}