[1]韩冰,丁秀婷.腔内镜与MR影像的配准融合算法在胃肠道疾病中的应用[J].中国医学物理学杂志,2026,43(7):910-915.[doi:DOI:10.3969/j.issn.1005-202X.2026.07.011]
　HAN Bing,DING Xiuting.Application of endoscopic and MRI image registration and fusion algorithm in gastrointestinal diseases[J].Chinese Journal of Medical Physics,2026,43(7):910-915.[doi:DOI:10.3969/j.issn.1005-202X.2026.07.011]
点击复制

腔内镜与MR影像的配准融合算法在胃肠道疾病中的应用()

《中国医学物理学杂志》[ISSN:1005-202X/CN:44-1351/R]

卷:
43卷
期数:
2026年第7期
页码:
910-915
栏目:
医学影像物理
出版日期:
2026-07-22

文章信息/Info

Title:
Application of endoscopic and MRI image registration and fusion algorithm in gastrointestinal diseases
文章编号:
1005-202X（2026）07-0910-06
作者:
韩冰; 丁秀婷
北京大学第三医院秦皇岛医院消化内科， 河北 秦皇岛 066000
Author(s):
HAN Bing;  DING Xiuting
Department of Gastroenterology, Qinhuangdao Hospital, Peking University Third Hospital, Qinhuangdao 066000, China
关键词:
医学影像配准; 多模态融合; 腔内镜; 磁共振成像; 胃肠道疾病; 深度学习; 计算机辅助诊断
Keywords:
Keywords: medical image registration multimodal fusion endoscopy magnetic resonance imaging gastrointestinal disease deep learning computer-aided diagnosis
分类号:
R318
DOI:
DOI:10.3969/j.issn.1005-202X.2026.07.011
文献标志码:
A
摘要:
目的：开发一种鲁棒、精准的腔内镜-MR影像配准融合算法，并系统验证其在胃肠道肿瘤精确定位与边界界定、炎症性肠病评估等临床场景中的应用价值，以推动多模态影像融合在消化内镜领域的实用化进程。方法：提出一种基于混合特征与深度强化学习的非刚性配准框架（Hybrid-DRL）。采用包含120例患者的配对腔内镜（白光、窄带成像、蓝光成像模式）与三维T2加权MR影像数据集进行训练与测试。算法首先提取基于深度学习的稠密特征图与改进的加速稳健特征关键点作为混合特征，随后通过近端策略优化深度强化学习代理优化非刚性形变场参数。结果：该算法的平均靶点配准误差相比传统基于互信息的算法和单纯基于深度学习的VoxelMorph算法显著降低（P<0.01）。在临床应用中，融合影像使早期胃癌边界的观察者间一致性从0.72提高至0.91，并辅助发现12%的病例存在传统内镜难以识别的黏膜下浸润迹象。结论：提出的Hybrid-DRL算法能有效实现腔内镜与MR影像的高精度配准，并在提升胃肠道疾病诊疗精准性方面具有明确价值，为胃肠道疾病的精准诊疗提供了可靠的技术框架与实证依据，为多模态智能融合从算法创新向临床实用转化提供关键支持。
Abstract:
Abstract: Objective To develop a robust and precise registration and fusion algorithm for endoscopic and MRI images, and systematically validate its application potential in clinical scenarios including precise localization and boundary definition of gastrointestinal tumors and the evaluation of inflammatory bowel diseases, thereby promoting the practical application of multimodal image fusion in digestive endoscopy. Methods A non-rigid registration framework (Hybrid-DRL) was established based on hybrid features and deep reinforcement learning. A paired dataset consisting of endoscopic images (white light, narrow band imaging, and blue light imaging modes) and three-dimensional T2-weighted MR images from 120 patients was used for model training and testing. The algorithm first extracted dense feature maps based on deep learning and improved speeded up robust features key points to construct hybrid features, and then optimized non-rigid deformation field parameters through the proximal policy optimization-based deep reinforcement learning agent. Results The average target registration error of the proposed algorithm was significantly lower than that of the traditional mutual information-based algorithm and the deep learning-based VoxelMorph algorithm (P<0.01). In clinical applications, the fusion images improved the inter-observer consistency of boundary observation for early gastric cancer from 0.72 to 0.91, and assisted in detecting 12% of cases with submucosal infiltration signs that were difficult to identify by traditional endoscopy. Conclusion The proposed Hybrid-DRL algorithm can effectively achieve high-precision registration of endoscopic and MRI images, and possesses promising value in improving the diagnostic and therapeutic accuracy of gastrointestinal diseases. These findings provide a reliable technical framework and empirical basis for the precise diagnosis and treatment of gastrointestinal diseases, as well as key support for the transformation of multimodal intelligent fusion from algorithm innovation to clinical application.

相似文献/References:

[1]张文妙韵,周武,戴培山.三维旋转估计及其在医学影像配准中的应用[J].中国医学物理学杂志,2015,32(04):542.[doi:10.3969/j.issn.1005-202X.2015.04.020]
　[J].Chinese Journal of Medical Physics,2015,32(7):542.[doi:10.3969/j.issn.1005-202X.2015.04.020]
[2]康帅,奚峥皓,黄陈,等.基于DPU-Net的直肠癌肿瘤分割与T分期研究[J].中国医学物理学杂志,2023,40(10):1189.[doi:DOI:10.3969/j.issn.1005-202X.2023.10.001]
　KANG Shuai,XI Zhenghao,HUANG Chen,et al.Rectal tumor segmentation and T staging based on DPU-Net[J].Chinese Journal of Medical Physics,2023,40(7):1189.[doi:DOI:10.3969/j.issn.1005-202X.2023.10.001]
[3]沈逸凡,宁瑞鹏,李任任,等.基于多模态磁共振成像探讨阿尔茨海默症和轻度认知障碍的功能-结构共变模式[J].中国医学物理学杂志,2025,42(10):1298.[doi:DOI:10.3969/j.issn.1005-202X.2025.10.006]
　SHEN Yifan,NING Ruipeng,et al.Exploring function-structure covariant patterns in Alzheimers disease and mild cognitive impairment based on multimodal magnetic resonance imaging[J].Chinese Journal of Medical Physics,2025,42(7):1298.[doi:DOI:10.3969/j.issn.1005-202X.2025.10.006]
[4]刘文哲,郑一博,刘强,等.基于信号质量指数动态加权融合的多模态呼吸率估计方法[J].中国医学物理学杂志,2026,43(1):99.[doi:DOI:10.3969/j.issn.1005-202X.2026.01.013]
　LIU Wenzhe,ZHENG Yibo,LIU Qiang,et al.Signal quality index-based dynamic weighted fusion method for multimodal respiratory rate estimation[J].Chinese Journal of Medical Physics,2026,43(7):99.[doi:DOI:10.3969/j.issn.1005-202X.2026.01.013]
[5]代妮娜,张文君.基于多模态融合技术提升乳腺超声与磁共振成像联合诊断乳腺癌的效能[J].中国医学物理学杂志,2026,43(3):317.[doi:DOI:10.3969/j.issn.1005-202X.2026.03.006]
　DAI Nina,ZHANG Wenjun.Improving the diagnostic efficacy of breast cancer by integrating breast ultrasound and magnetic resonance imaging with multimodal fusion technology[J].Chinese Journal of Medical Physics,2026,43(7):317.[doi:DOI:10.3969/j.issn.1005-202X.2026.03.006]

备注/Memo

备注/Memo:
【收稿日期】2026-03-01 【基金项目】河北省医学科学研究课题（20241389） 【作者简介】韩冰，硕士研究生，主治医师，研究方向：肝硬化发病机制及治疗，E-mail: H80295b@163.com 【通信作者】丁秀婷，主任医师，研究方向：肝硬化发病机制及治疗，E-mail: 13803240800@163.com
更新日期/Last Update: 2026-07-22