计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (9): 16-24.doi: 10.11896/jsjkx.250300159

• 智能医学工程 • 上一篇    下一篇

基于深度学习的肾脏超声图像分割:现状与挑战

尹诗1, 施振扬1, 吴梦麟1,2, 蔡金燕1, 余德3   

  1. 1 南京工业大学计算机与信息工程学院 南京 211816
    2 深圳卡本医疗器械有限公司 广东 深圳 518000
    3 江苏开放大学信息工程学院 南京 210036
  • 收稿日期:2025-03-31 修回日期:2025-06-19 出版日期:2025-09-15 发布日期:2025-09-11
  • 通讯作者: 余德(yude@jsou.edu.cn)
  • 作者简介:(yinshi2021@njtech.edu.cn)
  • 基金资助:
    江苏省自然科学基金(BK20230312)

Deep Learning-based Kidney Segmentation in Ultrasound Imaging:Current Trends and Challenges

YIN Shi1, SHI Zhenyang1, WU Menglin1,2, CAI Jinyan1, YU De3   

  1. 1 College of Computer and Information Engineering,Nanjing Tech University,Nanjing 211816,China
    2 Carbon Medical Device Ltd.,Shenzhen,Guangdong 518000,China
    3 School of Information Technology,Jiangsu Open University,Nanjing 210036,China
  • Received:2025-03-31 Revised:2025-06-19 Online:2025-09-15 Published:2025-09-11
  • About author:YIN Shi,born in 1991,associate researcher,is a member of CCF(No.50991G).Her main research interests include medical image analysis and deep learning.
    YU De,born in 1987,lecturer.His main research interests include signal processing and medical image analysis.
  • Supported by:
    Natural Science Foundation of Jiangsu Province(BK20230312).

PDF (PC)

20

摘要： 肾脏超声图像分割作为一项关键的临床任务,在疾病诊断和治疗规划中发挥着重要作用。该综述系统回顾了2017至2024年间肾脏超声图像分割领域的重要研究成果,重点分析了二维和三维分割技术及异常病变肾脏分割方法。对于二维超声图像,总结了4类分割技术方法:1)基于纹理特征提取的传统分割方法;2)U-Net及其改进架构;3)融合肾脏形状和边界先验知识的深度监督学习方法;4)多模态信息融合分割技术。此外,详细梳理了当前公开可用的数据集和标准化评估指标,为后续研究提供了可靠的比较基准。尽管当前二维分割方法已取得显著进展,但仍面临诸多挑战:精细解剖结构的分割精度有待提升,三维分割技术尚未成熟,异常病变分割研究明显不足,以及高质量训练数据严重匮乏等关键问题。这些技术瓶颈的突破将直接决定该领域研究成果的临床转化前景。展望未来,需要重点发展精细结构与三维分割技术、探索跨模态学习方法、深化组织特征信息融合策略,并着力构建大模型和高质量数据集,以全面提升肾脏超声分割技术的临床应用价值。

关键词: 肾脏超声分割, 深度学习, 肾脏异常, 评估指标, 公开数据集

Abstract: Kidney ultrasound segmentation plays a pivotal role in clinical diagnosis and treatment planning.This review systema-tically reviews key developments in renal segmentation techniques from 2017 to 2024,focusing on 2D/3D approaches and pathological tissue analysis.Current 2D methods encompass four categories:traditional texture-based techniques,U-Net variants,shape-prior integrated deep learning,and multimodal fusion approaches.The study comprehensively evaluates available datasets and standardized metrics,establishing critical benchmarks for the field.While significant progress has been made in 2D segmentation,persistent challenges include limited precision in fine structures,immature 3D techniques,inadequate pathological analysis,and data scarcity.Overcoming these limitations is crucial for clinical translation.Future directions emphasize refining structural segmentation,advancing 3D reconstruction,developing cross-modal learning,and creating comprehensive datasets.These efforts will enhance the clinical utility of renal ultrasound segmentation,bridging the gap between technical innovation and medical application.

Key words: Kidney US segmentation, Deep learning, Renal abnormalities, Evaluation metrics, Open datasets

中图分类号: 

  • TP391.41
[1]TORRES H R,QUEIRÓS S,MORAIS P,et al.Kidney segmentation in ultrasound,magnetic resonance and computed tomography images:A systematic review[J].Computer Methods and Programs in Biomedicine,2018,157:49-67.
[2]MENG X,LUO D,MO R.Application value of surgical navigation system based on deep learning and mixed reality for guiding puncture in percutaneous nephrolithotomy:a retrospective study[J].BMC Urology,2024,24(1):230.
[3]KIM D W,AHN H G,KIM J,et al.Advanced Kidney Volume Measurement Method Using Ultrasonography with Artificial Intelligence-Based Hybrid Learning in Children[J].Sensors,2021,21(20):6846.
[4]YIN S,PENG Q,LI H,et al.Multi-instance deep learning of ultrasound imaging data for pattern classification of congenital abnormalities of the kidney and urinary tract in children[J].Uro-logy,2020,142:183-189.
[5]YIN S,PENG Q,LI H,et al.Multi-instance deep learning with graph convolutional neural networks for diagnosis of kidney di-seases using ultrasound imaging[C]//Uncertainty for Safe Utilization of Machine Learning in Medical Imaging and Clinical Image-Based Procedures:First International Workshop,UNSURE 2019,and 8th International Workshop,CLIP 2019.Springer,2019:146-154.
[6]GUNABUSHANAM G,SCOUTT L M.Ultrasound Image Optimization for the Interventional Radiologist[J].Techniques in Vascular and Interventional Radiology,2021,24(3):100766.
[7]MARSOUSI M,PLATANIOTIS K N,STERGIOPOULOS S.An Automated Approach for Kidney Segmentation in Three-Dimensional Ultrasound Images[J].IEEE Journal of Biomedical and Health Informatics,2017,21(4):1079-1094.
[8]RANDLES R,FINNEGAN A.Guidelines for writing a syste-matic review[J].Nurse Education Today,2023,125:105803.
[9]GUO S,CHEN H,SHENG X,et al.Cross-modal transfer learning based on an improved CycleGAN model for accurate kidney segmentation in ultrasound images[J].Ultrasound in Medicine &Biology,2024,50(11):1638-1645.
[10]OGHLI M G,BAGHERI S M,SHABANZADEH A,et al.Fully automated kidney image biomarker prediction in ultrasound scans using Fast-Unet++[J].Scientific Reports,2024,14(1):4782.
[11]CHANG Y C,LO C M,CHEN Y K,et al.W-Net:Two-stagesegmentation for multi-center kidney ultrasound[C]//2024 IEEE Conference on Artificial Intelligence(CAI).IEEE,2024:1522-1523.
[12]MAHMUD S,ABBAS T O,CHOWDHURY M E,et al.Automated grading of prenatal hydronephrosis severity from segmented kidney ultrasounds using deep learning[J].Expert Systems with Applications,2024,255:124594.
[13]WANG Z,GUAN Y,CHEN Z,et al.A Kidney Dynamic Ultrasound Image Segmentation Method Based on STDC Network[C]//2024 36th Chinese Control and Decision Conference(CCDC).IEEE,2024:501-505.
[14]PENG T,RUAN Y,GU Y,et al.Coarse-to-fine approach:Automatic delineation of kidney ultrasound data[J].Big Data Mining and Analytics,2024,7(4):1321-1332.
[15]ALEXA R,KRANZ J,KRAMANN R,et al.Harnessing Artificial Intelligence for Enhanced Renal Analysis:Automated Detection of Hydronephrosis and Precise Kidney Segmentation[J].European Urology Open Science,2024,62:19-25.
[16]DAOUD M I,SHTAIYAT A,YOUNES H A,et al.Improved kidney outlining in ultrasound images by combining deep lear-ning semantic segmentation with conventional active contour[C]//10th International Conference on Electrical and Electronics Engineering(ICEEE 2023).IEEE,2023:74-78.
[17]CHEN G P,ZHAO Y,DAI Y,et al.Asymmetric U-shaped network with hybrid attention mechanism for kidney ultrasound images segmentation[J].Expert Systems with Applications,2023,212:118847.
[18]SONG Z,LIU X,GONG Y,et al.A Two-Stage Framework for Kidney Segmentation in Ultrasound Images[C]//International Conference on Neural Computing for Advanced Applications.Springer,2023:60-74.
[19]CHEN S H,WU Y L,PAN C Y,et al.Renal ultrasound image segmentation method based on channel attention and GL-UNet11[J].Journal of Radiation Research and Applied Sciences,2023,16(3):100631.
[20]PENG T,GU Y,RUAN S J,et al.Novel solution for using neural networks for kidney boundary extraction in 2D ultrasound data[J].Biomolecules,2023,13(10):1548.
[21]PENG T,WU Y,GU Y,et al.Intelligent contour extraction approach for accurate segmentation of medical ultrasound images[J].Frontiers in Physiology,2023,14:1177351.
[22]CHEN G,LIU Y,QIAN J,et al.DSEU-net:A novel deep supervision SEU-net for medical ultrasound image segmentation[J].Expert Systems with Applications,2023,223:119939.
[23]CHEN G,DAI Y,ZHANG J,et al.MBANet:Multi-branchaware network for kidney ultrasound images segmentation[J].Computers in Biology and Medicine,2022,141:105140.
[24]CHEN G,YIN J,DAI Y,et al.A novel convolutional neural network for kidney ultrasound images segmentation[J].Computer Methods and Programs in Biomedicine,2022,218:106712.
[25]SONG Y,ZHENG J,LEI L,et al.CT2US:Cross-modal transfer learning for kidney segmentation in ultrasound images with synthesized data[J].Ultrasonics,2022,122:106706.
[26]ALEX D M,ABRAHAM CHANDY D,HEPZIBAH CHRISTINAL A,et al.YSegNet:a novel deep learning network for kidney segmentation in 2D ultrasound images[J].Neural Computing and Applications,2022,34(24):22405-22416.
[27]CHEN G,DAI Y,ZHANG J,et al.Mbdsnet:Automatic segmentation of kidney ultrasound images using a multi-branch and deep supervision network[J].Digital Signal Processing,2022,130:103742.
[28]CHEN J,JIN P,SONG Y,et al.Auto-segmentation ultrasound-based radiomics technology to stratify patient with diabetic kidney disease:A multi-center retrospective study[J].Frontiers in Oncology,2022,12:876967.
[29]LEE S,KANG M,BYEON K,et al.Machine Learning-Aided Chronic Kidney Disease Diagnosis Based on Ultrasound Imaging Integrated with Computer-Extracted Measurable Features[J].Journal of Digital Imaging,2022,35(5):1091-1100.
[30]SINGLA R,RINGSTROM C,HU R,et al.Speckle and sha-dows:ultrasound-specific physics-based data augmentation applied to kidney segmentation[C]//Medical Imaging with Deep Learning.2022.
[31]FENG W,LIU J,GUAN Z,et al.Renal Ultrasound Image Segmentation Based on U-Net and Generative Adversarial Nets[C]//2022 7th Asia-Pacific Conference on Intelligent Robot Systems(ACIRS).IEEE,2022:96-100.
[32]VALENTE S,MORAIS P,TORRES H R,et al.A deep learningmethod for kidney segmentation in 2D ultrasound images[C]//2022 44th Annual International Conference of the IEEE Engineering in Medicine & Biology Society(EMBC).IEEE,2022:3911-3914.
[33]WANG S,SINGH V K,CHEAH E,et al.Stacked dilated convolutions and asymmetric architecture for U-Net-based medical image segmentation[J].Computers in Biology and Medicine,2022,148:105891.
[34]SONG S H,HAN J H,KIM K S,et al.Deep-learning segmentation of ultrasound images for automated calculation of the hydronephrosis area to renal parenchyma ratio[J].Investigative and Clinical Urology,2022,63(4):455.
[35]YIN S,ZHANG Z,LI H,et al.Fully-automatic segmentation of kidneys in clinical ultrasound images using a boundary distance regression network[C]//2019 IEEE 16th International Symposium on Biomedical Imaging(ISBI 2019).IEEE,2019:1741-1744.
[36]PENG H,GUAN Y,LI J,et al.MwUnet:A semantic segmentation deep learning method for the ultrasonic image of hydronephrosis in children[C]//2021 IEEE International Conference on Systems,Man,and Cybernetics(SMC).IEEE,2021:1894-1899.
[37]ROSHANITABRIZI P,ZEMBER J,SPRAGUE B M,et al.Standardized analysis of kidney ultrasound images for the prediction of pediatric hydronephrosis severity[C]//Machine Learning in Medical Imaging:12th International Workshop,MLMI 2021.Springer,2021:366-375.
[38]LIN Y,KHONG P L,ZOU Z,et al.Evaluation of pediatric hydronephrosis using deep learning quantification of fluid-to-kidney-area ratio by ultrasonography[J].Abdominal Radiology,2021,46:5229-5239.
[39]CHEN G,DAI Y,LI R,et al.SDFNet:Automatic segmentation of kidney ultrasound images using multi-scale low-level structu-ral feature[J].Expert Systems with Applications,2021,185:115619.
[40]WEN P,GUAN Y,LI J,et al.A-PSPNet:A novel segmentation method of renal ultrasound image[C]//2021 IEEE International Conference on Systems,Man,and Cybernetics(SMC).IEEE,2021:40-45.
[41]YIN S,PENG Q,LI H,et al.Automatic kidney segmentation in ultrasound images using subsequent boundary distance regression and pixelwise classification networks[J].Medical Image Analysis,2020,60:101602.
[42]ALEX D M,CHANDY D A.Investigations on performances of pre-trained U-Net models for 2D ultrasound kidney image segmentation[C]//International Conference for Emerging Techno-logies in Computing.Cham:Springer,2020:185-195.
[43]WU Z,HAI J,ZHANG L,et al.Cascaded fully convolutionalDenseNet for automatic kidney segmentation in ultrasound images[C]//2019 2nd International Conference on Artificial Intelligence and Big Data(ICAIBD).IEEE,2019:384-388.
[44]LI X,LI C,LIU H,et al.A modified level set algorithm based on point distance shape constraint for lesion and organ segmentation[J].Physica Medica,2019,57:123-36.
[45]MEENAKSHI S,SUGANTHI M,SURESHKUMAR P.Seg-mentation and boundary detection of fetal kidney images in second and third trimesters using kernel-based fuzzy clustering[J].Journal of Medical Systems,2019,43(7):203.
[46]ZHENG Q,WARNER S,TASIAN G,et al.A Dynamic Graph Cuts Method with Integrated Multiple Feature Maps for Segmenting Kidneys in 2D Ultrasound Images[J].Academic Radiology,2018,25(9):1136-1145.
[47]SELVATHI D,BAMA S.Phase based distance regularized level set for the segmentation of ultrasound kidney images[J].Pattern Recognition Letters,2017,86:9-17.
[48]RAVISHANKAR H,THIRUVENKADAM S,VENKATARAMANI R,et al.Joint deep learning of foreground,background and shape for robust contextual segmentation[C]//International Conference on Information Processing in Medical Imaging.Cham:Springer,2017:622-632.
[49]RAVISHANKAR H,VENKATARAMANI R,THIRUVENKADAM S,et al.Learning and incorporating shape models for semantic segmentation[C]//International Conference on Medical Image Computing and Computer-assisted Intervention.Cham:Springer,2017:203-211.
[50]SINGLA R,RINGSTROM C,HU G,et al.The open kidney ultrasound data set[C]//International Workshop on Advances in Simplifying Medical Ultrasound.Cham:Springer,2023:155-164.
[51]PAWAR M,DOSHI P,SHINDE R.Detection and Segmentation of Kidney from Ultrasound Image Using GVF[C]//Techno-Societal 2018:Proceedings of the 2nd International Conference on Advanced Technologies for Societal Applications-Volume 1.Cham:Springer,2019:217-229.
[52]JAGTAP J M,GREGORY A V,HOMES H L,et al.Automated measurement of total kidney volume from 3D ultrasound images of patients affected by polycystic kidney disease and comparison to MR measurements[J].Abdominal Radiology,2022,47(7):2408-2419.
[53]TABRIZI P R,MANSOOR A,CERROLAZA J J,et al.Auto-matic kidney segmentation in 3D pediatric ultrasound images using deep neural networks and weighted fuzzy active shape model[C]//2018 IEEE 15th International Symposium on Biomedical Imaging(ISBI 2018).IEEE,2018:1170-1173.
[54]TORRES H R,QUEIRÓS S,MORAIS P,et al.Kidney segmentation in 3-D ultrasound images using a fast phase-based approach[J].IEEE Transactions on Ultrasonics,Ferroelectrics,and Frequency Control,2020,68(5):1521-1531.
[55]WEERASINGHE N H,LOVELL N H,WELSH A W,et al.Multi-parametric fusion of 3D power Doppler ultrasound for fetal kidney segmentation using fully convolutional neural networks[J].IEEE Journal of Biomedical and Health Informatics,2020,25(6):2050-2057.
[56]BOUSSAID H,ROUET L.Shape Feature Loss for Kidney Segmentation in 3D Ultrasound Images[C]//BMVC.2021:427.
[57]NDZIMBONG W,FOURNIOL C,THEMYR L,et al.TRUSTED:The Paired 3D Transabdominal Ultrasound and CT Human Data for Kidney Segmentation and Registration Research[J].Scientific Data,2025,12(1):615.
[58]TABRIZI P R,MANSOOR A,CERROLAZA J J,et al.Auto-matic segmentation of the renal collecting system in 3D pediatric ultrasound to assess the severity of hydronephrosis[C]//2019 IEEE 16th International Symposium on Biomedical Imaging(ISBI 2019).IEEE,2019:1717-1720.
[59]NITHYA A,APPATHURAI A,VENKATADRI N,et al.Kidney disease detection and segmentation using artificial neural network and multi-kernel k-means clustering for ultrasound images[J].Measurement,2020,149:106952.
[60]ESKANDARI S,MESHGINI S,FARZAMNIA A.Using a no-vel algorithm in ultrasound images to detect renal stones[C]//Proceedings of the 12th National Technical Seminar on Unmanned System Technology 2020:NUSYS’20.Springer,2022:755-767.
[61]RAJU P,RAO V M,RAO B P.Optimal GLCM combined FCM segmentation algorithm for detection of kidney cysts and tumor[J].Multimedia Tools and Applications,2019,78:18419-18441.
[62]QI H,WANG Z,QI X,et al.Ultrasound image segmentation of renal tumors based on UNet++ with fusion of multiscale residuals and dual attention[J].Physics in Medicine & Biology,2024,69(7):075002.
[63]AKKASALIGAR P T,BIRADAR S,BADIGER S.Segmentation of kidney stones in medical ultrasound images[C]//International Conference on Recent Trends in Image Processing and Pattern Recognition.Springer,2019:200-208.
[64]SELVARANI S,RAJENDRAN P.Detection of Renal Calculi in Ultrasound Image Using Meta-Heuristic Support Vector Machine[J].Journal of Medical Systems,2019,43(9):300.
[65]JEEVAW S.Ultrasound Normal Kidney Image Dataset[EB/OL].https://universe.roboflow.com/jeevaws/ultrasound-normal-kidney-image.
[66]VALENTE S,MORAIS P,TORRES H R,et al.A Comparative Study of Deep Learning Methods for Multi-Class Semantic Segmentation of 2D Kidney Ultrasound Images[C]//2023 45th Annual International Conference of the IEEE Engineering in Medicine & Biology Society(EMBC).IEEE,2023:1-4.
[67]NOBLE J A,BOUKERROUI D.Ultrasound image segmenta-tion:a survey[J].IEEE Transactions on Medical Imaging,2006,25(8):987-1010.
[68]HELLER N,ISENSEE F,MAIER-HEIN K H,et al.The state of the art in kidney and kidney tumor segmentation in contrast-enhanced CT imaging:Results of the KiTS19 challenge[J].Medical Image Analysis,2021,67:101821.
[69]NOBLE J A.Ultrasound image segmentation and tissue characterization[J].Proceedings of the Institution of Mechanical Engineers,2010,224(2):307-316.
[70]WU P,LIU Y,LI Y,et al.Robust Prostate Segmentation Using Intrinsic Properties of TRUS Images[J].IEEE Transactions on Medical Imaging,2015,34(6):1321-1335.
[71]QIAN K,ANDO T,NAKAMURA K,et al.Ultrasound imaging method for internal jugular vein measurement and estimation of circulating blood volume[J].International Journal of Computer Assisted Radiology and Surgery,2014,9(2):231-239.
[72]MAZUROWSKI M A,DONG H,GU H,et al.Segment any-thing model for medical image analysis:An experimental study[J].Medical Image Analysis,2023,89:102918.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发