surgical-video

This repo contains an implementation code for the weakly supervised surgical tool tracker. In this research, the temporal dependency in surgical video data is modeled using a convolutional LSTM which is trained only on image level labels to detect, localize and track surgical instruments.

Code for Surgical Skill Assessment via Video Semantic Aggregation (MICCAI 2022)

SG2VID: Scene Graphs Enable Fine-Grained Control for Video Synthesis (MICCAI 2025 - ORAL)

Visualization of dataset splits for surgical phase and instrument recognition

A curated list of surgical AI resources (continually updated)

Domain-adaptive CLIP retrieval system for surgical video keyframe–text matching. Includes adapter training, INT8 acceleration, PQ compression, and Reliability analysis.

Automated Assessment of Simulated Laparoscopic Surgical Performance using 3DCNN

Two open-source web tools for annotating surgical video: frame-annotator (clip classification) and surgical-annotator (multi-task masks, keypoints & phases). Built for trustworthy, reproducible surgical-AI datasets (MICCAI 2026 SafeSurg).

Official code and models for SurgRec: Scaling Video Pretraining for Surgical Foundation Models (arXiv:2603.29966)

Browser tool for labelling surgical video on two layers: Phase (task stages) and Gesture (JIGSAWS-style sub-segmentation). Vite dev server + stdlib-only Python scripts, no backend.

digitalsurgery — independent third-party profile of a public API surface, by API Evangelist. Digital Surgery (originally Touch Surgery / Kinosis) is a London-based surgical artificial intelligence company backed by Balderton Capital and acquired by Medtronic in 2020, where it now operates as part of Medtronic's Covidien surgical business.

Qwen3-VL with timestamp-aware frame routing for ORena SAVE FOCUS 2026 SEGMENT — Incision Impossible

Two-stage W64-to-NW2 Qwen3-VL temporal cascade for ORena SAVE FOCUS 2026 PROCEDURE — Incision Impossible