What are you looking at? Modality contribution in multimodal medical deep learning
Our metric offers valuable insights that can support the advancement of multimodal model development and dataset creation. By introducing this method, we contribute to the growing field of interpretability in deep learning for multimodal research. This approach helps to facilitate the integration of multimodal AI into clinical practice. Our code is publicly available at https://github.com/ChristianGappGit/MC_MMD.
Authors
- Christian Gapp (ORCID: https://orcid.org/0000-0002-4520-298X)
- Elke R. Gizewski (ORCID: https://orcid.org/0000-0001-6859-8377)
- Martin Welk (ORCID: https://orcid.org/0000-0002-6268-7050)
- Karl Fritscher (ORCID: https://orcid.org/0000-0003-2593-6203)
- Elias Tappeiner (ORCID: https://orcid.org/0000-0003-1034-8361)
- Rainer Schubert (ORCID: https://orcid.org/0000-0002-8026-7500)
Institutions
- Innsbruck Medical University (AT)
- Universität Innsbruck (AT)
- UMIT - Private Universität für Gesundheitswissenschaften, Medizinische Informatik und Technik (AT)
- Tirol Kliniken (AT)
Publication Details
- Journal
- International Journal of Computer Assisted Radiology and Surgery
- Published
- 2025-10-02
- DOI
- https://doi.org/10.1007/s11548-025-03523-w
- Primary Topic
- Explainable Artificial Intelligence (XAI)
- Type
- article
- Field-Weighted Citation Impact
- 0.00