# Optimizing Latent Graph Representations of Surgical Scenes for Zero-Shot Domain Transfer

arXiv: https://arxiv.org/abs/2403.06953 · HTML: https://mlsanity.com/paper/2403.06953/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 22 references that are in the map.
- First posted on arXiv 11 Mar 2024.

## Builds on (most cited first, 15 of 15)

- [EndoNet: A Deep Architecture for Recognition Tasks on Laparoscopic Videos](https://mlsanity.com/paper/1602.03012/) (2016)
- [RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective Whitening](https://mlsanity.com/paper/2103.15597/) (2021)
- [Latent Graph Representations for Critical View of Safety Assessment](https://mlsanity.com/paper/2212.04155/) (2022)
- [Surgical Action Triplet Detection by Mixed Supervised Learning of Instrument-Tissue Interactions](https://mlsanity.com/paper/2307.09548/) (2023)
- [LABRAD-OR: Lightweight Memory Scene Graphs for Accurate Bimodal Reasoning in Dynamic Operating Rooms](https://mlsanity.com/paper/2303.13293/) (2023)
- [Dynamic Scene Graph Representation for Surgical Video](https://mlsanity.com/paper/2309.14538/) (2023)
- [The Endoscapes Dataset for Surgical Scene Segmentation, Object Detection, and Critical View of Safety Assessment: Official Splits and Benchmark](https://mlsanity.com/paper/2312.12429/) (2023)
- [FUN-SIS: a Fully UNsupervised approach for Surgical Instrument Segmentation](https://mlsanity.com/paper/2202.08141/) (2022)
- [Federated Cycling (FedCy): Semi-supervised Federated Learning of Surgical Phases](https://mlsanity.com/paper/2203.07345/) (2022)
- [Encoding Surgical Videos as Latent Spatiotemporal Graphs for Object and Anatomy-Driven Reasoning](https://mlsanity.com/paper/2312.06829/) (2023)
- [Unifying Unsupervised Domain Adaptation and Zero-Shot Visual Recognition](https://mlsanity.com/paper/1903.10601/) (2019)
- [Semi-supervised Domain Adaptive Medical Image Segmentation through Consistency Regularized Disentangled Contrastive Learning](https://mlsanity.com/paper/2307.02798/) (2023)
- [ST(OR)2: Spatio-Temporal Object Level Reasoning for Activity Recognition in the Operating Room](https://mlsanity.com/paper/2312.12250/) (2023)
- [Adaptation of Surgical Activity Recognition Models Across Operating Rooms](https://mlsanity.com/paper/2207.03083/) (2022)
- [ACT-Net: Anchor-context Action Detection in Surgery Videos](https://mlsanity.com/paper/2310.03377/) (2023)

## Built on by (most cited first, 1 of 1)

- [The SAGES Critical View of Safety Challenge: A Global Benchmark for AI-Assisted Surgical Quality Assessment](https://mlsanity.com/paper/2509.17100/) (2025)

## Abstract (arXiv)

Purpose: Advances in deep learning have resulted in effective models for surgical video analysis; however, these models often fail to generalize across medical centers due to domain shift caused by variations in surgical workflow, camera setups, and patient demographics. Recently, object-centric learning has emerged as a promising approach for improved surgical scene understanding, capturing and disentangling visual and semantic properties of surgical tools and anatomy to improve downstream task performance. In this work, we conduct a multi-centric performance benchmark of object-centric approaches, focusing on Critical View of Safety assessment in laparoscopic cholecystectomy, then propose an improved approach for unseen domain generalization. Methods: We evaluate four object-centric approaches for domain generalization, establishing baseline performance. Next, leveraging the disentangled nature of object-centric representations, we dissect one of these methods through a series of ablations (e.g. ignoring either visual or semantic features for downstream classification). Finally, based on the results of these ablations, we develop an optimized method specifically tailored for domain generalization, LG-DG, that includes a novel disentanglement loss function. Results: Our optimized approach, LG-DG, achieves an improvement of 9.28% over the best baseline approach. More broadly, we show that object-centric approaches are highly effective for domain generalization thanks to their modular approach to representation learning. Conclusion: We investigate the use of object-centric methods for unseen domain generalization, identify method-agnostic factors critical for performance, and present an optimized approach that substantially outperforms existing methods.