Enhancing deep learning methods for brain metastasis detection through cross-technique annotations on SPACE MRI
Delineating small BMs on SPACE MRI sequence results in higher quality annotations than on MPRAGE sequence due to enhanced conspicuity. Leveraging cross-technique ground truth annotations during training improved the accuracy of DL models in detecting and segmenting BMs. Cross-technique annotation may enhance DL models by integrating benefits from specialized, time-intensive MRI sequences while not relying on them. Further validation in prospective studies is needed.
Authors
- Rami A. El Shafie (ORCID: https://orcid.org/0000-0002-1355-7882)
- Julius C. Holzschuh (ORCID: https://orcid.org/0009-0003-6768-2544)
- Thomas Welzel (ORCID: https://orcid.org/0000-0001-9960-3052)
- Bálint Kovàcs (ORCID: https://orcid.org/0000-0002-1191-0646)
- Irada Pflüger
- Tassilo Wald (ORCID: https://orcid.org/0009-0007-5222-2683)
- Philipp Kickingereder (ORCID: https://orcid.org/0000-0002-6224-0064)
- Andreas Kudak (ORCID: https://orcid.org/0000-0002-1474-1403)
- Bastian von Nettelbladt
- Constantin Ulrich (ORCID: https://orcid.org/0000-0003-3002-8170)
- Michael Baumgartner (ORCID: https://orcid.org/0000-0003-4455-9917)
- Benjamin Hamm (ORCID: https://orcid.org/0009-0003-4818-8700)
- Klaus H. Maier-Hein
- Jürgen Debus
Institutions
- University of Bonn (DE)
- German Cancer Research Center (DE)
- Heidelberg University (DE)
- University Hospital Bonn (DE)
- University Hospital Heidelberg (DE)
- Heidelberger Institut für Radioonkologie (DE)
- National Center for Tumor Diseases (DE)
- Universitätsmedizin Göttingen (DE)
- German Center for Lung Research (DE)
- DKFZ-ZMBH Alliance (DE)
- Deutsches Konsortium für Translationale Krebsforschung (DE)
- Heidelberg University (US)
Publication Details
- Journal
- European Radiology Experimental
- Published
- 2025-02-06
- DOI
- https://doi.org/10.1186/s41747-025-00554-5
- Citations
- 3
- Primary Topic
- Brain Metastases and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 4.56
Funders
- Deutsche Forschungsgemeinschaft