Task type
Workflow recognition, segmentation, detection, action recognition, assessment, and multimodal summarization.
Model Explorer · Research implementations
Find the models shaping surgical video research.
A source-linked catalog for workflow, action recognition, segmentation, depth, and surgical representation learning.
These are research artifacts, not validated clinical products. Open code or weights do not establish safety, generalizability, or regulatory status.
Input modality
Video, still frames, instrument trajectories, structured operative events, text, or multimodal combinations.
Evidence maturity
Prototype, benchmark model, externally tested method, prospective study, or regulated product claim.
Use context
Education, annotation support, retrospective research, intraoperative assistance, quality review, or documentation.
Searching catalog...
How to compare
Architecture is only one part of the evidence.
Compare training data, external validation, task definitions, metrics, failure modes, compute requirements, licenses, and the workflow in which an output would be used.
The Weekly Signal
Get the weekly Surgical AI signal.
A concise email with practical explainers, useful papers, dataset updates, tools, and ways to participate.