{"authors":[{"id":null,"fullName":"Guzman Requena, Alejandro","name":"Alejandro","surname":"Guzman Requena","rank":1,"pid":null},{"id":null,"fullName":"Márquez Vara, Noah","name":"Noah","surname":"Márquez Vara","rank":2,"pid":null},{"id":null,"fullName":"Díaz, Oliver","name":"Oliver","surname":"Díaz","rank":3,"pid":null}],"openAccessColor":null,"publiclyFunded":false,"eoscIfGuidelines":null,"type":"publication","language":{"code":"und","label":"Undetermined"},"countries":[{"code":"ES","label":"Spain","provenance":null}],"subjects":[{"subject":{"scheme":"keyword","value":"Breast cancer"},"provenance":null},{"subject":{"scheme":"keyword","value":"Xarxes neuronals convolucionals"},"provenance":null},{"subject":{"scheme":"keyword","value":"Aprenentatge automàtic"},"provenance":null},{"subject":{"scheme":"keyword","value":"Machine learning"},"provenance":null},{"subject":{"scheme":"FOS","value":"0202 electrical engineering, electronic engineering, information engineering"},"provenance":null},{"subject":{"scheme":"keyword","value":"Convolutional neural networks"},"provenance":null},{"subject":{"scheme":"FOS","value":"02 engineering and technology"},"provenance":null},{"subject":{"scheme":"keyword","value":"Mamografia"},"provenance":null},{"subject":{"scheme":"keyword","value":"Càncer de mama"},"provenance":null},{"subject":{"scheme":"keyword","value":"Mammography"},"provenance":null}],"mainTitle":"From hand-crafted radiomics to deep learning: evaluating breast cancer classification methods in mammograms","subTitle":null,"descriptions":["This study evaluates the performance of some machine learning (ML) and deep learning (DL) models for breast cancer tumor classification in mammography (MG) images, by training them on the BCDR dataset. The study compares the use of radiomics-based features in ML models, including Random Forest, Support Vector Machines, and XGBoost, with two deep learning approaches using Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). Radiomics features were extracted from segmented regions of interest (ROIs) and used to train the ML models, performing hyperparameter tuning and cross-validation to optimize the results. CNN and ViT models were trained using the MG location present in the ROI segmentations, to explore the impact of tumor region localization assistance on classification performance. To examine and verify the performance of the ML models and the ViT, the area under the receiver operating characteristic curve (AUC-ROC) and the training execution time of all experiments (performed on the same device) were used. The results indicate that, while all methods achieve good performance on the training dataset (mean AUCROC scores around 0.9), they exhibit substantial performance drops when tested on external data. Among the evaluated models, ViT achieves the highest overall AUC-ROC in both internal (0.93) and external (0.68) validation, surpassing CNNs and radiomics-based ML models. However, ViT also incurs the highest computational cost, highlighting a trade-off between accuracy and training time. These findings underscore the need for multicenter, multi-vendor data to improve model generalization and reliability, as well as for continued refinement of advanced architectures, such as transformers, to optimize breast cancer lesion classification in clinical settings."],"publicationDate":"2025-04-10","publisher":"SPIE","embargoEndDate":null,"sources":["Crossref","Comunicacions a congressos (Matemàtiques i Informàtica)","reponame:Dipòsit Digital de la UB","instname:"],"formats":["application/pdf","10 p."],"contributors":null,"coverages":null,"bestAccessRight":{"code":"c_abf2","label":"OPEN","scheme":"http://vocabularies.coar-repositories.org/documentation/access_rights/"},"container":{"name":"Medical Imaging 2025: Imaging Informatics","issnPrinted":null,"issnOnline":null,"issnLinking":null,"ep":null,"iss":null,"sp":"31","vol":null,"edition":null,"conferencePlace":null,"conferenceDate":null},"documentationUrls":null,"codeRepositoryUrl":null,"programmingLanguage":null,"contactPeople":null,"contactGroups":null,"tools":null,"size":null,"version":null,"geoLocations":null,"id":"doi_dedup___::e58cba1a0c989cd7ed2acc1baa3fb771","originalIds":["10.1117/12.3046672","50|doiboost____|e58cba1a0c989cd7ed2acc1baa3fb771","oai:dnet:ubarcelona__::f10adb9118931cba45b6f95a9c39b53d","50|RECOLECTA___::e123925b8fcf390fd91a25ac31bd9798","oai:recercat.cat:2445/220790","50|od_______272::0baad9cc9642b40c2adbd555f8e4c5f3","50|od_______963::f10adb9118931cba45b6f95a9c39b53d","oai:diposit.ub.edu:2445/220790"],"pids":[{"scheme":"doi","value":"10.1117/12.3046672"},{"scheme":"handle","value":"2445/220790"}],"dateOfCollection":null,"lastUpdateTimeStamp":null,"indicators":{"citationImpact":{"citationCount":0.0,"influence":2.1746283E-9,"popularity":2.2497804E-9,"impulse":0.0,"citationClass":"C5","influenceClass":"C5","impulseClass":"C5","popularityClass":"C5"}},"projects":[{"id":"corda_____he::b99dd52a74cdca7480c43e03b02ef58d","code":"101057699","acronym":"RadioVal","title":"International Clinical Validation of Radiomics Artificial Intelligence for Breast Cancer Treatment Planning","funder":"European Commission","pids":[{"scheme":"doi","value":"10.3030/101057699"}]},{"id":"corda__h2020::7d386b683cadfaf9aec343d449010b74","code":"952103","acronym":"EuCanImage","title":"A European Cancer Image Platform Linked to Biological and Health Data for Next-Generation Artificial Intelligence and Precision Medicine in Oncology","funder":"European Commission","pids":[{"scheme":"doi","value":"10.3030/952103"}]}],"organizations":[{"legalName":"University of Barcelona","acronym":"UB","id":"openorgs____::6c3442fa797102c1c5c1a4319fce4119","pids":[{"scheme":"fundref","value":"501100005774"},{"scheme":"Wikidata","value":"Q219615"},{"scheme":"GRID","value":"grid.5841.8"},{"scheme":"ISNI","value":"0000000419370247"},{"scheme":"ROR","value":"https://ror.org/021018s57"},{"scheme":"FundRef","value":"501100006203"},{"scheme":"OrgRef","value":"61406"},{"scheme":"mag_id","value":"71999127"},{"scheme":"wikidata","value":"Q219615"},{"scheme":"RRID","value":"RRID:SCR_002575"},{"scheme":"fundref","value":"501100006203"},{"scheme":"PIC","value":"999986387"},{"scheme":"OrgReg","value":"ES0017"},{"scheme":"FundRef","value":"501100005774"}],"countries":[{"code":"ES","label":"Spain"}],"websiteurl":"http://www.ub.edu/web/ub/en/"},{"legalName":"Computer Vision Center","acronym":"Computer Vision Center","id":"pending_org_::faa69bf86a693b1f22d8ede7739c889c","pids":[{"scheme":"wikidata","value":"Q47076707"},{"scheme":"ISNI","value":"0000000121705884"},{"scheme":"ROR","value":"https://ror.org/00s0nnj93"},{"scheme":"Wikidata","value":"Q47076707"}],"countries":[{"code":"ES","label":"Spain"}],"websiteurl":null},{"legalName":"University of Barcelona (UB)","acronym":"University of Barcelona (UB)","id":"pending_org_::41b446de3a30033977674529b1dbf761","pids":null,"countries":[{"code":"ES","label":"Spain"}],"websiteurl":null}],"communities":[{"code":"spain","label":"Spanish Research Portal","provenance":null},{"code":"cancer-research","label":"Cancer Research","provenance":null}],"collectedFrom":[{"key":"openaire____::4cb2a3eb94033446c37331f60fad0847","value":"Recolector de Ciencia Abierta, RECOLECTA"},{"key":"opendoar____::1ce927f875864094e3906a4a0b5ece68","value":"Diposit Digital de la Universitat de Barcelona"},{"key":"openaire____::081b82f96300b6a6e3d282bad31cb6e2","value":"Crossref"},{"key":"opendoar____::7a614fd06c325499f1680b9896beedeb","value":"Research Repository of Catalonia"}],"instances":[{"pids":[{"scheme":"doi","value":"10.1117/12.3046672"}],"type":"Article","urls":["https://doi.org/10.1117/12.3046672"],"publicationDate":"2025-04-10","refereed":"peerReviewed","hostedBy":{"key":"openaire____::55045bd2a65019fd8e6741a755395c8c","value":"Unknown Repository"},"collectedFrom":{"key":"openaire____::081b82f96300b6a6e3d282bad31cb6e2","value":"Crossref"}},{"pids":[{"scheme":"handle","value":"2445/220790"}],"alternateIdentifiers":[{"scheme":"doi","value":"10.1117/12.3046672"}],"type":"Conference object","urls":["https://hdl.handle.net/2445/220790"],"publicationDate":"2025-01-01","refereed":"nonPeerReviewed","hostedBy":{"key":"openaire____::4cb2a3eb94033446c37331f60fad0847","value":"Recolector de Ciencia Abierta, RECOLECTA"},"collectedFrom":{"key":"openaire____::4cb2a3eb94033446c37331f60fad0847","value":"Recolector de Ciencia Abierta, RECOLECTA"}},{"pids":[{"scheme":"handle","value":"2445/220790"}],"accessRight":{"code":"c_16ec","label":"RESTRICTED","scheme":"http://vocabularies.coar-repositories.org/documentation/access_rights/","openAccessRoute":null},"type":"Conference object","urls":["https://hdl.handle.net/2445/220790"],"publicationDate":"2025-05-05","refereed":"nonPeerReviewed","hostedBy":{"key":"opendoar____::7a614fd06c325499f1680b9896beedeb","value":"Research Repository of Catalonia"},"collectedFrom":{"key":"opendoar____::7a614fd06c325499f1680b9896beedeb","value":"Research Repository of Catalonia"}},{"pids":[{"scheme":"handle","value":"2445/220790"}],"accessRight":{"code":"c_abf2","label":"OPEN","scheme":"http://vocabularies.coar-repositories.org/documentation/access_rights/","openAccessRoute":null},"type":"Conference object","urls":["https://hdl.handle.net/2445/220790"],"publicationDate":"2025-05-05","refereed":"nonPeerReviewed","hostedBy":{"key":"opendoar____::1ce927f875864094e3906a4a0b5ece68","value":"Diposit Digital de la Universitat de Barcelona"},"collectedFrom":{"key":"opendoar____::1ce927f875864094e3906a4a0b5ece68","value":"Diposit Digital de la Universitat de Barcelona"}}],"links":[{"header":{"relationType":"resultProject","relationClass":"isProducedBy","relatedIdentifier":"corda_____he::b99dd52a74cdca7480c43e03b02ef58d","relatedRecordType":"project","relationProvenance":"iis","trust":"0.897"},"collectedfrom":[{"dsId":"openaire____::3f264f93cf3b0cfc4ede188a6300455c","dsName":"CORDA - COmmon Research DAta Warehouse - Horizon Europe"}],"projectTitle":"International Clinical Validation of Radiomics Artificial Intelligence for Breast Cancer Treatment Planning","code":"101057699","funding":{"funder":{"id":"ec__________::EC","shortname":"EC","name":"European Commission","jurisdiction":{"code":"EU","label":"European Union"},"pid":null},"level0":{"id":"ec__________::EC::HE","description":"Horizon Europe Framework Programme","name":"HE"},"level1":{"id":"ec__________::EC::HE::HORIZON-RIA","description":"HORIZON  Research and Innovation Actions","name":"HORIZON-RIA"},"level2":{"id":null,"description":null,"name":null}},"startDate":"2022-09-01","endDate":"2027-02-28"},{"header":{"relationType":"resultProject","relationClass":"isProducedBy","relatedIdentifier":"corda__h2020::7d386b683cadfaf9aec343d449010b74","relatedRecordType":"project","relationProvenance":"iis","trust":"0.897"},"collectedfrom":[{"dsId":"openaire____::a55eb91348674d853191f4f4fd73d078","dsName":"CORDA - COmmon Research DAta Warehouse - Horizon 2020"}],"projectTitle":"A European Cancer Image Platform Linked to Biological and Health Data for Next-Generation Artificial Intelligence and Precision Medicine in Oncology","code":"952103","funding":{"funder":{"id":"ec__________::EC","shortname":"EC","name":"European Commission","jurisdiction":{"code":"EU","label":"European Union"},"pid":null},"level0":{"id":"ec__________::EC::H2020","description":"Horizon 2020 Framework Programme","name":"H2020"},"level1":{"id":"ec__________::EC::H2020::RIA","description":"Research and Innovation action","name":"RIA"},"level2":{"id":null,"description":null,"name":null}},"startDate":"2020-10-01","endDate":"2025-09-30"}],"otherTitles":null,"green":true,"inDiamondJournal":false,"isGreen":true,"isInDiamondJournal":false}