A generic video foundation model, finetuned on unlabeled surgical footage and optionally fused with OR sensor streams, reaches competitive phase recognition on HeiCo and improves in-house outcome prediction.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Leveraging Generic Foundation Models for Multimodal Surgical Data Analysis
A generic video foundation model, finetuned on unlabeled surgical footage and optionally fused with OR sensor streams, reaches competitive phase recognition on HeiCo and improves in-house outcome prediction.