1.上海理工大学健康科学与工程学院(上海 200093)
2.上海理工大学上海介入医疗器械工程技术研究中心(上海 200093)
3.上海健康医学院医学影像学院(上海 201318)
4.同济大学附属第十人民医院泌尿外科(上海 200072)
5.上海申康医院发展中心(上海 200041)
沈兵。E-mail:urodrshenbing@shsmu.edu.cn
收稿:2025-12-22,
修回:2026-02-25,
纸质出版:2026-06-25
移动端阅览
王文韬,张海康,黄清明,等.基于nnU-Net的肾脏区域分割方法及其术中应用[J].北京生物医学工程,2026,45(3):221-228,246.
WANG Wentao,ZHANG Haikang,HUANG Qingming,et al.Kidney region segmentation method based on nnU-Net and its intraoperative application[J].Beijing Biomedical Engineering,2026,45(3):221-228,246.
王文韬,张海康,黄清明,等.基于nnU-Net的肾脏区域分割方法及其术中应用[J].北京生物医学工程,2026,45(3):221-228,246. DOI: 10.3969/j.issn.1002-3208.2026.03.001.
WANG Wentao,ZHANG Haikang,HUANG Qingming,et al.Kidney region segmentation method based on nnU-Net and its intraoperative application[J].Beijing Biomedical Engineering,2026,45(3):221-228,246. DOI: 10.3969/j.issn.1002-3208.2026.03.001.
目的
2
探索基于nnU-Net的深度学习方法在肾脏手术中的应用,特别是术中图像分割的可行性。随着肾脏肿瘤手术的复杂性增加,精准的术中图像分割成为手术导航的关键。研究目标是通过nnU-Net框架,提高术中肾脏区域的自动识别和定位精度,从而为智能辅助手术系统提供技术支持。
方法
2
基于术中获取的左目内镜图像,采用nnU-Net框架构建了二维语义分割模型。训练数据通过逐帧截取术中图像,并结合手工掩码标注生成。标注区域包括肾脏实质、肿瘤及表面粘连脂肪,作为一个统一的感兴趣区域(region of interest,ROI)进行识别。在数据验证方面,使用了独立的测试集,包括未参与训练和验证的5例肾脏手术病例,最终通过与人工标注的掩码进行对比分析,以评估模型在术中图像分割中的表现。
结果
2
模型在独立测试集中表现良好,平均 Dice相似系数(Dice similarity coefficient,DSC)为0.933 5,交并比(intersection over union, IoU)为0.878 4,精确率(precision)为0.951 7,召回率(recall)为0.921 1,
F
1 分数(
F
1 score)为0.933 5。nnU-Net能够有效地分割肾脏区域,并在不同情况下保持较高的分割精度。推理速度方面,模型的平均单帧分割时间约为0.38 s,显示出较强的处理能力,适用于术中快速图像更新与定位识别需求。
结论
2
nnU-Net模型在复杂术中图像环境中依然保持较高的准确性与鲁棒性,验证了nnU-Net在实时图像分割任务中的应用潜力。nnU-Net可以为智能辅助手术系统提供可靠的技术支撑,具有较好的临床应用前景。
Objective
2
Explore the application of deep learning methods based on nnU-Net in kidney surgery, particularly the feasibility of intraoperative image segmentation. As the complexity of kidney tumor surgery increases, precise intraoperative image segmentation becomes crucial for surgical navigation. The research goal is to improve the automatic recognition and localization accuracy of the renal region during surgery using the nnU-Net framework, thereby providing technical support for intelligent assisted surgery systems.
Methods
2
Based on the left-eye endoscopic images obtained during surgery, a 2D semantic segmentation model was constructed using the nnU-Net framework. The training data were generated by extracting individual frames from the intraoperative images and combining them with manually annotated masks. The annotated regions include renal parenchyma, tumors, and surface-adhered fat, which were identified as a unified region of Interest (ROI). For data validation, an independent test set was used, consisting of five kidney surgery cases that did not participate in the training and validation. The model's performance in intraoperative image segmentation was evaluated by comparing it with manually annotated masks.
Results
2
The model performed well on the independent test set, with an average Dice similarity coefficient (DSC) of 0.9335, intersection over union (IoU) of 0.878 4, precision of 0.951 7, recall of 0.921 1, and
F
1 score of 0.933 5. These results indicate that nnU-Net can effectively segment the renal region and maintain high segmentation accuracy under various conditions. In terms of inference speed, the average single-frame segmentation time of the model is approximately 0.38 seconds, demonstrating strong processing capability, making it suitable for rapid intraoperative image updates and localization tasks.
Conclusions
2
The nnU-Net model maintains high accuracy and robustness even in complex intraoperative image environments, validating its potential for real-time image segmentation tasks. nnU-Net can provide reliable technical support for intelligent assisted surgery systems, showing promising clinical application prospects.
Stefano P, Ahmed E, Enrico C, et al. New imaging technologies for robotic kidney cancer surgery[J]. Asian Journal of Urology, 2022, 9(3): 253-262.
Wang Y, Wilder S, Hijazi M, et al. Surgeon skill and perioperative outcomes in robot-assisted partial nephrectomy[J]. JAMA Network Open, 2024, 7(7): e2421696.
Soomro NA, Hashimoto DA, Porteous AJ, et al. Systematic review of learning curves in robot-assisted surgery[J]. BJS Open, 2020, 4(1): 27-44.
Gao YH, Jiang Y, Peng YA, et al. Medical image segmentation: a comprehensive review of deep learning-based methods[J]. Tomography (Ann Arbor, Mich.), 2025, 11(5): 52.
Yao WJ, Bai JJ, Liao W, et al. From CNN to transformer: a review of medical image segmentation models[J].Journal of Imaging Informatics in Medicine,2024,37(4):1529-1547.
Jeon KS, Joo I, Park J, et al. Fully-automated multi-organ segmentation tool applicable to both non-contrast and post-contrast abdominal CT: deep learning algorithm developed using dual-energy CT images[J]. Scientific Reports, 2024, 14(1): 4378.
Krishnan C, Schmidt E, Onuoha E, et al. nnUNet for automatic kidney and cyst segmentation in autosomal dominant polycystic kidney disease[J]. Current Medical Imaging, 2024, 20(1): 1-9.
Luo H, Li JZ, Huang HY, et al. AI-based segmentation of renal enhanced CT images for quantitative evaluation of chronic kidney disease[J]. Scientific Reports, 2024, 14(1): 16890.
Nie ZH, Xu MH, Wang ZY, et al. A review of application of deep learning in endoscopic image processing[J]. Journal of Imaging, 2024, 10(11): 275.
Zhang M, Ye Z, Yuan EY, et al. Imaging-based deep learning in kidney diseases: recent progress and future prospects[J].Insights into Imaging,2024,15(1):50.
Yushkevich PA, Gao Y, Guido G. ITK-SNAP: An interactive tool for semi-automatic segmentation of multi-modality biomedical images[J]. Annual International Conference of the IEEE Engineering in Medicine and Biology Society,2016,2016:3342-3345.
Tanzi L, Piazzolla P, Porpiglia F, et al. Real-time deep learning semantic segmentation during intra-operative surgery for 3D augmented reality assistance[J].International Journal of Computer Assisted Radiology and Surgery,2021,16(9):1435-1445.
Ronneberger O, Fischer P, Brox T. U-Net: convolutional networks for biomedical image segmentation[M]// Navab N, Hornegger J, Wells W,et al,eds. Medical image computing and computer-assisted intervention-MICCAI 2015. Cham:Springer,2015:234-241.
Park GS, Park J, Choi RH, et al. Deep learning model for real-time semantic segmentation during intraoperative robotic prostatectomy[J]. European Urology Open Science, 2024, 62: 47-53.
Kilian C, Rasoul S, Salomé M, et al. Kidney tracking for live augmented reality in stereoscopic mini-invasive partial nephrectomy[J].Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization,2023,11(4):1251-1260.
Isensee F, Jaeger PF, Kohl SA-A, et al. nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation[J]. Nature Methods, 2020, 18(2): 203-211.
He KY, Peng B, Yu WB, et al. A novel Mis-Seg-Focus loss function based on a two-stage nnU-Net framework for accurate brain tissue segmentation[J]. Bioengineering, 2024, 11(5): 427.
Marcos F, Bruno S, Sandro Q, et al. Exploring optical flow inclusion into nnU-Net framework for surgical instrument segmentation[C]//Life and Health Sciences Research Institute (Portugal);IPCA (Portugal);KARL STORZ SE Co. KG (Germany),2024: 1292827.
Lunardo F, Baker L, Tan A, et al. How much data do you need? An analysis of pelvic multi-organ segmentation in a limited data context[J]. Physical and Engineering Sciences in Medicine, 2025, 48(1): 409-419.
Zhang X, Liu X, Wang ZY. DRA U-Net: An attention based U-Net framework for 2D medical image segmentation[C]//Proceedings of the 2021 IEEE International Conference on Big Data (Big Data). Los Angeles: IEEE, 2021: 3936-3942.
Chen Y, Wu Y, Wu JH, et al. Redefining contextual and boundary synergy: a boundary-guided fusion network for medical image segmentation[J].Electronics,2024,13(24):4986-4986.
Evgin G. Medical image data augmentation: techniques, comparisons and interpretations[J].Artificial Intelligence Review,2023,56(11):41-45.
Chen J, Chen R, Wang W, et al. TinyU-Net: Lighter yet better U-Net with Cascaded Multi-receptive fields[C]//Linguraru MG, Dou Q, Feragen A, et al,eds. Medical image computing and computer assisted intervention-MICCAI 2024. Cham: Springer, 2024: 626-635.
Liu XY, Qu LH, Xie ZY, et al. Towards more precise automatic analysis: a systematic review of deep learning-based multi-organ segmentation[J].BioMedical Engineering OnLine,2024,23(1):52-52.
0
浏览量
25
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621