Real-time visual feedback from catheterization analysis is crucial for enhancing surgical safety and efficiency during endovascular interventions. However, existing datasets are often limited to specific tasks, small scale, and lack the comprehensive annotations necessary for broader endovascular intervention understanding. To tackle these limitations, we introduce CathAction, a large-scale dataset for catheterization understanding. Our CathAction dataset encompasses approximately 500,000 annotated frames for catheterization action understanding and collision detection, and 25,000 ground truth masks for catheter and guidewire segmentation. For each task, we benchmark recent related works in the field. We further discuss the challenges of endovascular intentions compared to traditional computer vision tasks and point out open research questions. We hope that CathAction will facilitate the development of endovascular intervention understanding methods that can be applied to real-world applications. The dataset is available at https://airvlab.github.io/cathaction/.
@article{arxiv.2408.13126,
title = {CathAction: A Benchmark for Endovascular Intervention Understanding},
author = {Baoru Huang and Tuan Vo and Chayun Kongtongvattana and Giulio Dagnino and Dennis Kundrat and Wenqiang Chi and Mohamed Abdelaziz and Trevor Kwok and Tudor Jianu and Tuong Do and Hieu Le and Minh Nguyen and Hoan Nguyen and Erman Tjiputra and Quang Tran and Jianyang Xie and Yanda Meng and Binod Bhattarai and Zhaorui Tan and Hongbin Liu and Hong Seng Gan and Wei Wang and Xi Yang and Qiufeng Wang and Jionglong Su and Kaizhu Huang and Angelos Stefanidis and Min Guo and Bo Du and Rong Tao and Minh Vu and Guoyan Zheng and Yalin Zheng and Francisco Vasconcelos and Danail Stoyanov and Daniel Elson and Ferdinando Rodriguez y Baena and Anh Nguyen},
journal= {arXiv preprint arXiv:2408.13126},
year = {2024}
}