SANGRIA: Surgical Video Scene Graph Optimization for Surgical Workflow Prediction

• URL: http://arxiv.org/abs/2407.20214v2
• Date: Sat, 5 Oct 2024 11:18:09 GMT
• Title: SANGRIA: Surgical Video Scene Graph Optimization for Surgical Workflow Prediction
• Authors: Çağhan Köksal, Ghazal Ghazaei, Felix Holm, Azade Farshad, Nassir Navab,
• Abstract summary: We introduce an end-to-end framework for the generation and optimization of surgical scene graphs. Our solution outperforms the SOTA on the CATARACTS dataset by 8% accuracy and 10% F1 score in surgical workflow.
• Score: 37.86132786212667
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Graph-based holistic scene representations facilitate surgical workflow understanding and have recently demonstrated significant success. However, this task is often hindered by the limited availability of densely annotated surgical scene data. In this work, we introduce an end-to-end framework for the generation and optimization of surgical scene graphs on a downstream task. Our approach leverages the flexibility of graph-based spectral clustering and the generalization capability of foundation models to generate unsupervised scene graphs with learnable properties. We reinforce the initial spatial graph with sparse temporal connections using local matches between consecutive frames to predict temporally consistent clusters across a temporal neighborhood. By jointly optimizing the spatiotemporal relations and node features of the dynamic scene graph with the downstream task of phase segmentation, we address the costly and annotation-burdensome task of semantic scene comprehension and scene graph generation in surgical videos using only weak surgical phase labels. Further, by incorporating effective intermediate scene representation disentanglement steps within the pipeline, our solution outperforms the SOTA on the CATARACTS dataset by 8% accuracy and 10% F1 score in surgical workflow recognition

Related papers

• Revisiting Surgical Instrument Segmentation Without Human Intervention: A Graph Partitioning View [7.594796294925481]
  We propose an unsupervised method by reframing the video frame segmentation as a graph partitioning problem. A self-supervised pre-trained model is firstly leveraged as a feature extractor to capture high-level semantic features. On the "deep" eigenvectors, a surgical video frame is meaningfully segmented into different modules like tools and tissues, providing distinguishable semantic information.
  arXiv  Detail & Related papers  (2024-08-27T05:31:30Z)
• Encoding Surgical Videos as Latent Spatiotemporal Graphs for Object and Anatomy-Driven Reasoning [2.9724186623561435]
  We use latent graphs to represent a surgical video in terms of the constituent anatomical structures and tools over time. We introduce a novel graph-editing module that incorporates prior knowledge temporal coherence to correct errors in the graph.
  arXiv  Detail & Related papers  (2023-12-11T20:42:27Z)
• Dynamic Scene Graph Representation for Surgical Video [37.22552586793163]
  We exploit scene graphs as a more holistic, semantically meaningful and human-readable way to represent surgical videos. We create a scene graph dataset from semantic segmentations from the CaDIS and CATARACTS datasets. We demonstrate the benefits of surgical scene graphs regarding the explainability and robustness of model decisions.
  arXiv  Detail & Related papers  (2023-09-25T21:28:14Z)
• Local-Global Information Interaction Debiasing for Dynamic Scene Graph Generation [51.92419880088668]
  We propose a novel DynSGG model based on multi-task learning, DynSGG-MTL, which introduces the local interaction information and global human-action interaction information. Long-temporal human actions supervise the model to generate multiple scene graphs that conform to the global constraints and avoid the model being unable to learn the tail predicates.
  arXiv  Detail & Related papers  (2023-08-10T01:24:25Z)
• Deep Temporal Graph Clustering [77.02070768950145]
  We propose a general framework for deep Temporal Graph Clustering (GC) GC introduces deep clustering techniques to suit the interaction sequence-based batch-processing pattern of temporal graphs. Our framework can effectively improve the performance of existing temporal graph learning methods.
  arXiv  Detail & Related papers  (2023-05-18T06:17:50Z)
• LABRAD-OR: Lightweight Memory Scene Graphs for Accurate Bimodal Reasoning in Dynamic Operating Rooms [39.11134330259464]
  holistic modeling of the operating room (OR) is a challenging but essential task. We introduce memory scene graphs, where the scene graphs of previous time steps act as the temporal representation guiding the current prediction. We design an end-to-end architecture that intelligently fuses the temporal information of our lightweight memory scene graphs with the visual information from point clouds and images.
  arXiv  Detail & Related papers  (2023-03-23T14:26:16Z)
• EasyDGL: Encode, Train and Interpret for Continuous-time Dynamic Graph Learning [92.71579608528907]
  This paper aims to design an easy-to-use pipeline (termed as EasyDGL) composed of three key modules with both strong ability fitting and interpretability. EasyDGL can effectively quantify the predictive power of frequency content that a model learn from the evolving graph data.
  arXiv  Detail & Related papers  (2023-03-22T06:35:08Z)
• Scene Graph Modification as Incremental Structure Expanding [61.84291817776118]
  We focus on scene graph modification (SGM), where the system is required to learn how to update an existing scene graph based on a natural language query. We frame SGM as a graph expansion task by introducing the incremental structure expanding (ISE) We construct a challenging dataset that contains more complicated queries and larger scene graphs than existing datasets.
  arXiv  Detail & Related papers  (2022-09-15T16:26:14Z)
• Iterative Scene Graph Generation [55.893695946885174]
  Scene graph generation involves identifying object entities and their corresponding interaction predicates in a given image (or video) Existing approaches to scene graph generation assume certain factorization of the joint distribution to make the estimation iteration feasible. We propose a novel framework that addresses this limitation, as well as introduces dynamic conditioning on the image.
  arXiv  Detail & Related papers  (2022-07-27T10:37:29Z)
• Learning and Reasoning with the Graph Structure Representation in Robotic Surgery [15.490603884631764]
  Learning to infer graph representations can play a vital role in surgical scene understanding in robotic surgery. We develop an approach to generate the scene graph and predict surgical interactions between instruments and surgical region of interest.
  arXiv  Detail & Related papers  (2020-07-07T11:49:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.