User: Guest  Login
Title:

Joint Learning of Localized Representations from Medical Images and Reports

Document type:
Proceedings Paper
Author(s):
Mueller, Philip; Kaissis, Georgios; Zou, Congyu; Rueckert, Daniel
Abstract:
Contrastive learning has proven effective for pre-training image models on unlabeled data with promising results for tasks such as medical image classification. Using paired text (like radiological reports) during pre-training improves the results even further. Still, most existing methods target image classification downstream tasks and may not be optimal for localized tasks like semantic segmentation or object detection. We therefore propose Localized representation learning from Vision and Te...     »
Journal title abbreviation:
Med Image Comput Comput Assist Interv Int Conf Med Image Comput Comput Assist Interv
Year:
2022
Journal volume:
13686
Pages contribution:
685-701
Fulltext / DOI:
doi:10.1007/978-3-031-19809-0_39
Print-ISSN:
0302-9743
TUM Institution:
Institut für KI und Informatik in der Medizin (Prof. Rückert)
 BibTeX