首页 / 院系成果 / 成果详情页

Automated segmentation of complicated cystic renal masses using 3D V-Net convolutional neural network on MRI 期刊论文

编号：

6683E5EB113C71125F20A2E43A32D4BE
作者：

Kang, Huanhuan#^[1]Jia, Chuang^[2];Wang, Zhongyi(王中一)^[3]Huang, Bin^[2];Wang, He^[4];Jiang, Jiahui^[5];Liu, Zhe^[6];Cui, Mengqiu^[1];Zhao, Jian^[7];Bai, Xu^[1];Li, Lin^[8];Guo, Huiping^[1];Ning, Xueyi^[1];Ye, Huiyi^[1];Yang, Dawei^[5];Guo, Hao(郭浩)^[3]Xue, Jian*^[2]Wang, Haiyi*^[1]
语种：

英文
期刊：

BRITISH JOURNAL OF RADIOLOGY ISSN：0007-1285 2026 年 ; 2026 MAR 8
收录：
高质量科技期刊分级目录：

T2
关键词：

cystic renal mass convolutional neural network segmentation MRI
摘要：

Objectives: To develop and test a convolutional neural network model for automated segmentation of complicated cystic renal masses (cCRMs) on MRI. Methods: This multicenter retrospective study analysed 210 cCRMs between October 2019 and May 2021, divided into training/internal validation (n = 150, Institution 1) and test sets (n = 60, Institutions 2-4). Comparative 3D V-Net and U-Net models were developed across 7 MRI sequences (T2-weighted, diffusion-weighted, apparent diffusion coefficient maps, unenhanced T1-weighted, and enhanced corticomedullary, nephrographic, and excretory phases images). A total of 14 models were developed, and 7 pairwise comparisons were performed between the 3D V-Net and U-Net models. Segmentation performance was evaluated using Dice similarity coefficient (DSC) and Hausdorff distance (HD), with subgroup analysis of small cCRMs (<= 40 mm). Results: In the test set, the excretory-phase V-Net (EPV-Net model) showed the highest DSC, and perform better than the corresponding U-Net (EPU-Net model) across all cCRMs (DSC: 0.74 +/- 0.05 vs 0.70 +/- 0.06, P < .001; HD: 27.41 +/- 7.44 mm vs 39.18 +/- 11.07 mm, P < .001) and the 35 small cCRMs subgroup (DSC: 0.74 +/- 0.05 vs 0.70 +/- 0.06, P < .001; HD: 27.48 mm +/- 6.32 vs 38.72 +/- 10.69 mm, P < .001). Conclusions: The 3D EPV-Net model demonstrated good segmentation accuracy, even for small lesions, supporting its clinical utility for cCRMs evaluation. Advances in knowledge: This automated approach may streamline workflow compared to manual segmentation in cCRMs assessment.
推荐引用方式
GB/T 7714：

Kang Huanhuan,Jia Chuang,Wang Zhongyi, et al. Automated segmentation of complicated cystic renal masses using 3D V-Net convolutional neural network on MRI [J].BRITISH JOURNAL OF RADIOLOGY,2026.
APA：

Kang Huanhuan,Jia Chuang,Wang Zhongyi,Huang Bin,&Wang Haiyi.(2026).Automated segmentation of complicated cystic renal masses using 3D V-Net convolutional neural network on MRI .BRITISH JOURNAL OF RADIOLOGY.
MLA：

Kang Huanhuan, et al. "Automated segmentation of complicated cystic renal masses using 3D V-Net convolutional neural network on MRI" .BRITISH JOURNAL OF RADIOLOGY(2026).
入库时间：

3/16/2026 9:08:17 AM
更新时间：

3/31/2026 9:50:05 PM
条目包含文件：

文件类型： , 文件大小：

正在加载全文

未找到全文

浏览次数：5  下载次数：0

分享到：

浏览次数：5

下载次数：0

打印次数：0