Large-scale pretrained frame generative model enables real-time low-dose DSA imaging: An AI system development and multi-center validation study.

IF 12.8 Q1 MEDICINE, RESEARCH & EXPERIMENTAL

Med Pub Date : 2024-08-14 DOI:10.1016/j.medj.2024.07.025

Huangxuan Zhao, Ziyang Xu, Lei Chen, Linxia Wu, Ziwei Cui, Jinqiang Ma, Tao Sun, Yu Lei, Nan Wang, Hongyao Hu, Yiqing Tan, Wei Lu, Wenzhong Yang, Kaibing Liao, Gaojun Teng, Xiaoyun Liang, Yi Li, Congcong Feng, Tong Nie, Xiaoyu Han, Dongqiao Xiang, Charles B L M Majoie, Wim H van Zwam, Aad van der Lugt, P Matthijs van der Sluijs, Theo van Walsum, Yun Feng, Guoli Liu, Yan Huang, Wenyu Liu, Xuefeng Kan, Ruisheng Su, Weihua Zhang, Xinggang Wang, Chuansheng Zheng

{"title":"Large-scale pretrained frame generative model enables real-time low-dose DSA imaging: An AI system development and multi-center validation study.","authors":"Huangxuan Zhao, Ziyang Xu, Lei Chen, Linxia Wu, Ziwei Cui, Jinqiang Ma, Tao Sun, Yu Lei, Nan Wang, Hongyao Hu, Yiqing Tan, Wei Lu, Wenzhong Yang, Kaibing Liao, Gaojun Teng, Xiaoyun Liang, Yi Li, Congcong Feng, Tong Nie, Xiaoyu Han, Dongqiao Xiang, Charles B L M Majoie, Wim H van Zwam, Aad van der Lugt, P Matthijs van der Sluijs, Theo van Walsum, Yun Feng, Guoli Liu, Yan Huang, Wenyu Liu, Xuefeng Kan, Ruisheng Su, Weihua Zhang, Xinggang Wang, Chuansheng Zheng","doi":"10.1016/j.medj.2024.07.025","DOIUrl":null,"url":null,"abstract":"Background: Digital subtraction angiography (DSA) devices are commonly used in numerous interventional procedures across various parts of the body, necessitating multiple scans per procedure, which results in significant radiation exposure for both doctors and patients. Inspired by generative artificial intelligence techniques, this study proposes GenDSA, a large-scale pretrained multi-frame generative model-based real-time and low-dose DSA imaging system.Methods: GenDSA was developed to generate 1-, 2-, and 3-frame sequences following each real frame. A large-scale dataset comprising ∼3 million DSA images from 27,117 patients across 10 hospitals was constructed to pretrain, fine-tune, and validate GenDSA. Two other datasets from 25 hospitals were used for evaluation. Objective evaluations included SSIM and PSNR. Five interventional radiologists independently assessed the quality of the generated frames using the Likert scale and visual Turing test. Scoring consistency among the radiologists was measured using the Kendall coefficient of concordance (W). The Fleiss' kappa values were used for inter-rater agreement analysis for visual Turing tests.Findings: Using only one-third of the clinical radiation dose, videos generated by GenDSA were perfectly consistent with real videos. Objective evaluations demonstrated that GenDSA's performance (PSNR = 36.83, SSIM = 0.911, generation time = 0.07 s/frame) surpassed state-of-the-art algorithms. Subjective ratings and statistical results from five doctors indicated no significant difference between real and generated videos. Furthermore, the generated videos were comparable to real videos in overall quality (4.905 vs. 4.935) and lesion assessment (4.825 vs. 4.860).Conclusions: With clear clinical and translational values, the developed GenDSA can significantly reduce radiation damage to both doctors and patients during DSA-guided procedures.Funding: This study was supported by the National Key R&D Program and the National Natural Science Foundation of China.","PeriodicalId":29964,"journal":{"name":"Med","volume":" ","pages":""},"PeriodicalIF":12.8000,"publicationDate":"2024-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Med","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1016/j.medj.2024.07.025","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"MEDICINE, RESEARCH & EXPERIMENTAL","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Digital subtraction angiography (DSA) devices are commonly used in numerous interventional procedures across various parts of the body, necessitating multiple scans per procedure, which results in significant radiation exposure for both doctors and patients. Inspired by generative artificial intelligence techniques, this study proposes GenDSA, a large-scale pretrained multi-frame generative model-based real-time and low-dose DSA imaging system.

Methods: GenDSA was developed to generate 1-, 2-, and 3-frame sequences following each real frame. A large-scale dataset comprising ∼3 million DSA images from 27,117 patients across 10 hospitals was constructed to pretrain, fine-tune, and validate GenDSA. Two other datasets from 25 hospitals were used for evaluation. Objective evaluations included SSIM and PSNR. Five interventional radiologists independently assessed the quality of the generated frames using the Likert scale and visual Turing test. Scoring consistency among the radiologists was measured using the Kendall coefficient of concordance (W). The Fleiss' kappa values were used for inter-rater agreement analysis for visual Turing tests.

Findings: Using only one-third of the clinical radiation dose, videos generated by GenDSA were perfectly consistent with real videos. Objective evaluations demonstrated that GenDSA's performance (PSNR = 36.83, SSIM = 0.911, generation time = 0.07 s/frame) surpassed state-of-the-art algorithms. Subjective ratings and statistical results from five doctors indicated no significant difference between real and generated videos. Furthermore, the generated videos were comparable to real videos in overall quality (4.905 vs. 4.935) and lesion assessment (4.825 vs. 4.860).

Conclusions: With clear clinical and translational values, the developed GenDSA can significantly reduce radiation damage to both doctors and patients during DSA-guided procedures.

Funding: This study was supported by the National Key R&D Program and the National Natural Science Foundation of China.

查看原文本刊更多论文

大规模预训练帧生成模型可实现实时低剂量 DSA 成像：人工智能系统开发和多中心验证研究。

背景：数字减影血管造影（DSA）设备常用于身体各部位的众多介入手术，每次手术需要进行多次扫描，这对医生和患者都造成了巨大的辐射暴露。受生成人工智能技术的启发，本研究提出了基于生成模型的大规模预训练多帧实时低剂量 DSA 成像系统 GenDSA：方法：开发的 GenDSA 可在每个真实帧后生成 1、2 和 3 帧序列。为了对 GenDSA 进行预训练、微调和验证，建立了一个大规模数据集，其中包括来自 10 家医院 27,117 名患者的 300 万张 DSA 图像。另外两个来自 25 家医院的数据集用于评估。客观评估包括 SSIM 和 PSNR。五位介入放射科医生使用李克特量表和视觉图灵测试独立评估生成帧的质量。放射科医生之间的评分一致性采用 Kendall 一致性系数 (W) 进行测量。弗莱斯卡帕值用于视觉图灵测试的评分者间一致性分析：结果：仅使用三分之一的临床辐射剂量，GenDSA 生成的视频与真实视频完全一致。客观评估表明，GenDSA 的性能（PSNR = 36.83、SSIM = 0.911、生成时间 = 0.07 秒/帧）超过了最先进的算法。来自五位医生的主观评价和统计结果表明，真实视频和生成视频之间没有明显差异。此外，生成的视频在总体质量（4.905 对 4.935）和病变评估（4.825 对 4.860）方面与真实视频不相上下：结论：所开发的GenDSA具有明确的临床和转化价值，可在DSA引导手术中显著减少对医生和患者的辐射伤害：本研究得到了国家重点研发计划和国家自然科学基金的支持。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Med MEDICINE, RESEARCH & EXPERIMENTAL-

CiteScore

17.70

自引率

0.60%

发文量

102

期刊介绍： Med is a flagship medical journal published monthly by Cell Press, the global publisher of trusted and authoritative science journals including Cell, Cancer Cell, and Cell Reports Medicine. Our mission is to advance clinical research and practice by providing a communication forum for the publication of clinical trial results, innovative observations from longitudinal cohorts, and pioneering discoveries about disease mechanisms. The journal also encourages thought-leadership discussions among biomedical researchers, physicians, and other health scientists and stakeholders. Our goal is to improve health worldwide sustainably and ethically. Med publishes rigorously vetted original research and cutting-edge review and perspective articles on critical health issues globally and regionally. Our research section covers clinical case reports, first-in-human studies, large-scale clinical trials, population-based studies, as well as translational research work with the potential to change the course of medical research and improve clinical practice.