Surgical instrument-tissue interaction recognition with multi-task-attention video transformer
In this work, we demonstrate the benefits of using a fine-grained temporal context rather than static images or coarse temporal context for the task of surgical instrument-tissue interaction recognition. We also show that leveraging cross-attention with spatio-temporal features from various subtasks leads to improved surgical instrument-tissue interaction recognition performance. The project is available at: https://lennart-maack.github.io/InstrTissRec-MTAM .
Authors
- Lennart Maack (ORCID: https://orcid.org/0009-0002-0097-0157)
- Tobias Maurer (ORCID: https://orcid.org/0000-0002-5660-7691)
- Sarah Latus (ORCID: https://orcid.org/0000-0002-9796-7001)
- Alexander Schlaefer (ORCID: https://orcid.org/0000-0001-9201-8854)
- Berk Cam
Institutions
- Universität Hamburg (DE)
- University Medical Center Hamburg-Eppendorf (DE)
- Martini-Klinik (DE)
- Hamburg University of Technology (DE)
Publication Details
- Journal
- International Journal of Computer Assisted Radiology and Surgery
- Published
- 2025-11-11
- DOI
- https://doi.org/10.1007/s11548-025-03546-3
- Primary Topic
- Surgical Simulation and Training
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- European Commission
- HORIZON EUROPE Framework Programme