Optimizing MRI sequence classification performance: insights from domain shift analysis
Question Domain shift between adult and pediatric MRI data limits deep learning model accuracy, requiring solutions for reliable sequence classification across diverse patient populations. Findings The MedViT model outperformed ResNet-18 in pediatric imaging; expert domain knowledge adjustment further improved accuracy, demonstrating robustness across diverse datasets. Clinical relevance This study enhances MRI sequence classification by leveraging advanced neural networks and expert domain knowledge to mitigate domain shift, boosting diagnostic precision and efficiency across diverse patient populations in multicenter environments.
Authors
- Gianluca Brugnara (ORCID: https://orcid.org/0000-0003-2461-1407)
- Dominik Sturm (ORCID: https://orcid.org/0000-0003-0250-1696)
- Martin Bendszus (ORCID: https://orcid.org/0000-0002-9094-6769)
- Philipp Kickingereder (ORCID: https://orcid.org/0000-0002-6224-0064)
- Stefan M. Pfister (ORCID: https://orcid.org/0000-0002-5447-5322)
- Marianne Schell (ORCID: https://orcid.org/0000-0002-9754-1350)
- Felix Sahm (ORCID: https://orcid.org/0000-0001-5441-1962)
- Martha Foltyn (ORCID: https://orcid.org/0000-0002-1754-1765)
- Mustafa Ahmed Mahmutoglu (ORCID: https://orcid.org/0000-0002-4891-5547)
- Aditya Rastogi (ORCID: https://orcid.org/0000-0001-6103-4886)
Institutions
- German Cancer Research Center (DE)
- Heidelberg University (DE)
- University Hospital Bonn (DE)
- University Hospital Heidelberg (DE)
- Hopp Children's Cancer Center Heidelberg (DE)
- Deutsches Konsortium für Translationale Krebsforschung (DE)
Publication Details
- Journal
- European Radiology
- Published
- 2025-05-26
- DOI
- https://doi.org/10.1007/s00330-025-11671-5
- Citations
- 2
- Primary Topic
- Radiomics and Machine Learning in Medical Imaging
- Type
- article
- Field-Weighted Citation Impact
- 2.72
Funders
- Merck KGaA
- Deutsche Forschungsgemeinschaft
- Else Kröner-Fresenius-Stiftung
- Medizinischen Fakultät Heidelberg, Universität Heidelberg