Please use this identifier to cite or link to this item:
http://hdl.handle.net/2445/195373
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Garrucho Moras, Lidia | - |
dc.contributor.author | Kushibar, Kaisar | - |
dc.contributor.author | Jouide El Kaderi, Socayna | - |
dc.contributor.author | Díaz, Oliver | - |
dc.contributor.author | Igual Muñoz, Laura | - |
dc.contributor.author | Lekadir, Karim, 1977- | - |
dc.date.accessioned | 2023-03-16T11:47:50Z | - |
dc.date.available | 2023-03-16T11:47:50Z | - |
dc.date.issued | 2022-10 | - |
dc.identifier.issn | 0933-3657 | - |
dc.identifier.uri | http://hdl.handle.net/2445/195373 | - |
dc.description.abstract | Computer-aided detection systems based on deep learning have shown great potential in breast cancer detection. However, the lack of domain generalization of artificial neural networks is an important obstacle to their deployment in changing clinical environments. In this study, we explored the domain generalization of deep learning methods for mass detection in digital mammography and analyzed in-depth the sources of domain shift in a large-scale multi-center setting. To this end, we compared the performance of eight state-of-the-art detection methods, including Transformer based models, trained in a single domain and tested in five unseen domains. Moreover, a single-source mass detection training pipeline was designed to improve the domain generalization without requiring images from the new domain. The results show that our workflow generalized better than state-of-the-art transfer learning based approaches in four out of five domains while reducing the domain shift caused by the different acquisition protocols and scanner manufacturers. Subsequently, an extensive analysis was performed to identify the covariate shifts with the greatest effects on detection performance, such as those due to differences in patient age, breast density, mass size, and mass malignancy. Ultimately, this comprehensive study provides key insights and best practices for future research on domain generalization in deep learning based breast cancer detection. | - |
dc.format.extent | 14 p. | - |
dc.format.mimetype | application/pdf | - |
dc.language.iso | eng | - |
dc.publisher | Elsevier | - |
dc.relation.isformatof | Reproducció del document publicat a: https://doi.org/10.1016/j.artmed.2022.102386 | - |
dc.relation.ispartof | Artificial Intelligence in Medicine, 2022, vol. 132 | - |
dc.relation.uri | https://doi.org/10.1016/j.artmed.2022.102386 | - |
dc.rights | cc by-nc-nd (c) Lidia Garrucho Moras et al., 2022 | - |
dc.rights.uri | https://creativecommons.org/licenses/by-nc-nd/4.0/ | - |
dc.source | Articles publicats en revistes (Matemàtiques i Informàtica) | - |
dc.subject.classification | Càncer de mama | - |
dc.subject.classification | Diagnòstic per la imatge | - |
dc.subject.classification | Aprenentatge automàtic | - |
dc.subject.classification | Processament digital d'imatges | - |
dc.subject.classification | Mamografia | - |
dc.subject.other | Breast cancer | - |
dc.subject.other | Diagnostic imaging | - |
dc.subject.other | Machine learning | - |
dc.subject.other | Digital image processing | - |
dc.subject.other | Mammography | - |
dc.title | Domain generalization in deep learning based mass detection in mammography: A large-scale multi-center study | - |
dc.type | info:eu-repo/semantics/article | - |
dc.type | info:eu-repo/semantics/publishedVersion | - |
dc.identifier.idgrec | 725492 | - |
dc.date.updated | 2023-03-16T11:47:50Z | - |
dc.rights.accessRights | info:eu-repo/semantics/openAccess | - |
Appears in Collections: | Articles publicats en revistes (Matemàtiques i Informàtica) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
725492.pdf | 2.59 MB | Adobe PDF | View/Open |
This item is licensed under a Creative Commons License