{"title":"HiEndo:在两阶段框架下利用大规模数据生成高分辨率腹腔镜视频。","authors":"Zhao Wang, Yeqian Zhang, Jiayi Gu, Yueyao Chen, Yonghao Long, Xiang Xia, Puhua Zhang, Chunchao Zhu, Zerui Wang, Qi Dou, Zheng Wang, Zizhen Zhang","doi":"10.1080/24699322.2025.2536643","DOIUrl":null,"url":null,"abstract":"<p><p>Recent success in generative AI has demonstrated great potential in various medical scenarios. However, how to generate realistic and high-fidelity gastrointestinal laparoscopy videos still lacks exploration. A recent work, Endora, proposes a basic generation model for a gastrointestinal laparoscopy scenario, producing low-resolution laparoscopy videos, which can not meet the real needs in robotic surgery. Regarding this issue, we propose an innovative two-stage video generation architecture HiEndo for generating high-resolution gastrointestinal laparoscopy videos with high fidelity. In the first stage, we build a diffusion transformer for generating a low-resolution laparoscopy video upon the basic capability of Endora as an initial start. In the second stage, we further design a super resolution module to improve the resolution of initial video and refine the fine-grained details. With these two stages, we could obtain high-resolution realistic laparoscopy videos with high fidelity, which can meet the real-world clinical usage. We also collect a large-scale gastrointestinal laparoscopy video dataset with 61,270 video clips for training and validation of our proposed method. Extensive experimental results have demonstrate the effectiveness of our proposed framework. For example, our model achieves 15.1% Fréchet Video Distance and 3.7% F1 score improvements compared with the previous state-of-the-art method.</p>","PeriodicalId":56051,"journal":{"name":"Computer Assisted Surgery","volume":"30 1","pages":"2536643"},"PeriodicalIF":1.9000,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"HiEndo: harnessing large-scale data for generating high-resolution laparoscopy videos under a two-stage framework.\",\"authors\":\"Zhao Wang, Yeqian Zhang, Jiayi Gu, Yueyao Chen, Yonghao Long, Xiang Xia, Puhua Zhang, Chunchao Zhu, Zerui Wang, Qi Dou, Zheng Wang, Zizhen Zhang\",\"doi\":\"10.1080/24699322.2025.2536643\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Recent success in generative AI has demonstrated great potential in various medical scenarios. However, how to generate realistic and high-fidelity gastrointestinal laparoscopy videos still lacks exploration. A recent work, Endora, proposes a basic generation model for a gastrointestinal laparoscopy scenario, producing low-resolution laparoscopy videos, which can not meet the real needs in robotic surgery. Regarding this issue, we propose an innovative two-stage video generation architecture HiEndo for generating high-resolution gastrointestinal laparoscopy videos with high fidelity. In the first stage, we build a diffusion transformer for generating a low-resolution laparoscopy video upon the basic capability of Endora as an initial start. In the second stage, we further design a super resolution module to improve the resolution of initial video and refine the fine-grained details. With these two stages, we could obtain high-resolution realistic laparoscopy videos with high fidelity, which can meet the real-world clinical usage. We also collect a large-scale gastrointestinal laparoscopy video dataset with 61,270 video clips for training and validation of our proposed method. Extensive experimental results have demonstrate the effectiveness of our proposed framework. For example, our model achieves 15.1% Fréchet Video Distance and 3.7% F1 score improvements compared with the previous state-of-the-art method.</p>\",\"PeriodicalId\":56051,\"journal\":{\"name\":\"Computer Assisted Surgery\",\"volume\":\"30 1\",\"pages\":\"2536643\"},\"PeriodicalIF\":1.9000,\"publicationDate\":\"2025-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computer Assisted Surgery\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1080/24699322.2025.2536643\",\"RegionNum\":4,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2025/7/25 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q3\",\"JCRName\":\"SURGERY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Assisted Surgery","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1080/24699322.2025.2536643","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2025/7/25 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"SURGERY","Score":null,"Total":0}
HiEndo: harnessing large-scale data for generating high-resolution laparoscopy videos under a two-stage framework.
Recent success in generative AI has demonstrated great potential in various medical scenarios. However, how to generate realistic and high-fidelity gastrointestinal laparoscopy videos still lacks exploration. A recent work, Endora, proposes a basic generation model for a gastrointestinal laparoscopy scenario, producing low-resolution laparoscopy videos, which can not meet the real needs in robotic surgery. Regarding this issue, we propose an innovative two-stage video generation architecture HiEndo for generating high-resolution gastrointestinal laparoscopy videos with high fidelity. In the first stage, we build a diffusion transformer for generating a low-resolution laparoscopy video upon the basic capability of Endora as an initial start. In the second stage, we further design a super resolution module to improve the resolution of initial video and refine the fine-grained details. With these two stages, we could obtain high-resolution realistic laparoscopy videos with high fidelity, which can meet the real-world clinical usage. We also collect a large-scale gastrointestinal laparoscopy video dataset with 61,270 video clips for training and validation of our proposed method. Extensive experimental results have demonstrate the effectiveness of our proposed framework. For example, our model achieves 15.1% Fréchet Video Distance and 3.7% F1 score improvements compared with the previous state-of-the-art method.
期刊介绍:
omputer Assisted Surgery aims to improve patient care by advancing the utilization of computers during treatment; to evaluate the benefits and risks associated with the integration of advanced digital technologies into surgical practice; to disseminate clinical and basic research relevant to stereotactic surgery, minimal access surgery, endoscopy, and surgical robotics; to encourage interdisciplinary collaboration between engineers and physicians in developing new concepts and applications; to educate clinicians about the principles and techniques of computer assisted surgery and therapeutics; and to serve the international scientific community as a medium for the transfer of new information relating to theory, research, and practice in biomedical imaging and the surgical specialties.
The scope of Computer Assisted Surgery encompasses all fields within surgery, as well as biomedical imaging and instrumentation, and digital technology employed as an adjunct to imaging in diagnosis, therapeutics, and surgery. Topics featured include frameless as well as conventional stereotactic procedures, surgery guided by intraoperative ultrasound or magnetic resonance imaging, image guided focused irradiation, robotic surgery, and any therapeutic interventions performed with the use of digital imaging technology.