• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

使用改进的UNET架构对医学图像进行语义分割的综述

Review of Semantic Segmentation of Medical Images Using Modified Architectures of UNET.

作者信息

Krithika Alias AnbuDevi M, Suganthi K

机构信息

Vellore Institute of Technology, Chennai 600127, India.

出版信息

Diagnostics (Basel). 2022 Dec 6;12(12):3064. doi: 10.3390/diagnostics12123064.

DOI:10.3390/diagnostics12123064
PMID:36553071
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC9777361/
Abstract

In biomedical image analysis, information about the location and appearance of tumors and lesions is indispensable to aid doctors in treating and identifying the severity of diseases. Therefore, it is essential to segment the tumors and lesions. MRI, CT, PET, ultrasound, and X-ray are the different imaging systems to obtain this information. The well-known semantic segmentation technique is used in medical image analysis to identify and label regions of images. The semantic segmentation aims to divide the images into regions with comparable characteristics, including intensity, homogeneity, and texture. UNET is the deep learning network that segments the critical features. However, UNETs basic architecture cannot accurately segment complex MRI images. This review introduces the modified and improved models of UNET suitable for increasing segmentation accuracy.

摘要

在生物医学图像分析中，肿瘤和病变的位置及外观信息对于帮助医生治疗和识别疾病严重程度而言不可或缺。因此，对肿瘤和病变进行分割至关重要。磁共振成像（MRI）、计算机断层扫描（CT）、正电子发射断层显像（PET）、超声和X射线是获取此类信息的不同成像系统。医学图像分析中采用了著名的语义分割技术来识别和标记图像区域。语义分割旨在将图像划分为具有可比特征的区域，包括强度、均匀性和纹理。U-Net是用于分割关键特征的深度学习网络。然而，U-Net的基本架构无法准确分割复杂的MRI图像。本综述介绍了适用于提高分割精度的U-Net改进模型。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/4a9fea318745/diagnostics-12-03064-g020.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/d4087f80a898/diagnostics-12-03064-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/1860a67291e3/diagnostics-12-03064-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/51e74b39b346/diagnostics-12-03064-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/1aab803ecd3a/diagnostics-12-03064-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/a7f1340e91c1/diagnostics-12-03064-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/8c16fdaaa075/diagnostics-12-03064-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/7665b95f154e/diagnostics-12-03064-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/d81c2bc6d210/diagnostics-12-03064-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/f85344cde2c3/diagnostics-12-03064-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/5e75e8900c34/diagnostics-12-03064-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/9adb64ca84d8/diagnostics-12-03064-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/e824b9f7e036/diagnostics-12-03064-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/7538e2e76758/diagnostics-12-03064-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/2d151175e4b9/diagnostics-12-03064-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/fe471c1aca7b/diagnostics-12-03064-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/3ae310943094/diagnostics-12-03064-g016.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/aee21094f8f6/diagnostics-12-03064-g017.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/c2ece7aa0784/diagnostics-12-03064-g018.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/f8bf70ae4c36/diagnostics-12-03064-g019.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/4a9fea318745/diagnostics-12-03064-g020.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/d4087f80a898/diagnostics-12-03064-g001.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/1860a67291e3/diagnostics-12-03064-g002.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/51e74b39b346/diagnostics-12-03064-g003.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/1aab803ecd3a/diagnostics-12-03064-g004.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/a7f1340e91c1/diagnostics-12-03064-g005.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/8c16fdaaa075/diagnostics-12-03064-g006.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/7665b95f154e/diagnostics-12-03064-g007.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/d81c2bc6d210/diagnostics-12-03064-g008.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/f85344cde2c3/diagnostics-12-03064-g009.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/5e75e8900c34/diagnostics-12-03064-g010.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/9adb64ca84d8/diagnostics-12-03064-g011.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/e824b9f7e036/diagnostics-12-03064-g012.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/7538e2e76758/diagnostics-12-03064-g013.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/2d151175e4b9/diagnostics-12-03064-g014.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/fe471c1aca7b/diagnostics-12-03064-g015.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/3ae310943094/diagnostics-12-03064-g016.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/aee21094f8f6/diagnostics-12-03064-g017.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/c2ece7aa0784/diagnostics-12-03064-g018.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/f8bf70ae4c36/diagnostics-12-03064-g019.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/b0b8/9777361/4a9fea318745/diagnostics-12-03064-g020.jpg

相似文献

1
Review of Semantic Segmentation of Medical Images Using Modified Architectures of UNET.使用改进的UNET架构对医学图像进行语义分割的综述
Diagnostics (Basel). 2022 Dec 6;12(12):3064. doi: 10.3390/diagnostics12123064.
2
Attention-VGG16-UNet: a novel deep learning approach for automatic segmentation of the median nerve in ultrasound images.注意力-VGG16-UNet：一种用于超声图像中正中神经自动分割的新型深度学习方法。
Quant Imaging Med Surg. 2022 Jun;12(6):3138-3150. doi: 10.21037/qims-21-1074.
3
Automatic intraprostatic lesion segmentation in multiparametric magnetic resonance images with proposed multiple branch UNet.利用提出的多分支U-Net在多参数磁共振图像中实现前列腺内病变的自动分割。
Med Phys. 2020 Dec;47(12):6421-6429. doi: 10.1002/mp.14517. Epub 2020 Oct 24.
4
Dense-UNet: a novel multiphoton cellular image segmentation model based on a convolutional neural network.密集型U-Net：一种基于卷积神经网络的新型多光子细胞图像分割模型。
Quant Imaging Med Surg. 2020 Jun;10(6):1275-1285. doi: 10.21037/qims-19-1090.
5
Deep Neural Network for Cardiac Magnetic Resonance Image Segmentation.用于心脏磁共振图像分割的深度神经网络
J Imaging. 2022 May 23;8(5):149. doi: 10.3390/jimaging8050149.
6
[Application of Improved Unet Network in the Recognition and Segmentation of Hemorrhage Regions in Brain CT Images].改进的Unet网络在脑CT图像出血区域识别与分割中的应用
Sichuan Da Xue Xue Bao Yi Xue Ban. 2022 Jan;53(1):114-120. doi: 10.12182/20220160302.
7
Connected-UNets: a deep learning architecture for breast mass segmentation.连接式UNet：一种用于乳腺肿块分割的深度学习架构。
NPJ Breast Cancer. 2021 Dec 2;7(1):151. doi: 10.1038/s41523-021-00358-x.
8
Lung and Infection CT-Scan-Based Segmentation with 3D UNet Architecture and Its Modification.基于3D UNet架构及其改进的肺部与感染CT扫描分割
Healthcare (Basel). 2023 Jan 10;11(2):213. doi: 10.3390/healthcare11020213.
9
Attention-UNet architectures with pretrained backbones for multi-class cardiac MR image segmentation.具有预训练骨干网络的注意力UNet架构用于多类心脏磁共振图像分割。
Curr Probl Cardiol. 2024 Jan;49(1 Pt C):102129. doi: 10.1016/j.cpcardiol.2023.102129. Epub 2023 Oct 20.
10
Comparing different CT, PET and MRI multi-modality image combinations for deep learning-based head and neck tumor segmentation.比较基于深度学习的头颈部肿瘤分割的不同 CT、PET 和 MRI 多模态图像组合。
Acta Oncol. 2021 Nov;60(11):1399-1406. doi: 10.1080/0284186X.2021.1949034. Epub 2021 Jul 15.

引用本文的文献

1
An Ensemble Learning for Automatic Stroke Lesion Segmentation Using Compressive Sensing and Multi-Resolution U-Net.一种使用压缩感知和多分辨率U-Net的自动中风病变分割集成学习方法。
Biomimetics (Basel). 2025 Aug 4;10(8):509. doi: 10.3390/biomimetics10080509.
2
Deep Learning Spinal Cord Segmentation Based on B0 Reference for Diffusion Tensor Imaging Analysis in Cervical Spondylotic Myelopathy.基于B0参考的深度学习脊髓分割在颈椎病性脊髓病扩散张量成像分析中的应用
Bioengineering (Basel). 2025 Jun 28;12(7):709. doi: 10.3390/bioengineering12070709.
3
PA OmniNet: A retraining-free, generalizable deep learning framework for robust photoacoustic image reconstruction.

本文引用的文献

1
A Computationally Efficient Approach to Segmentation of the Aorta and Coronary Arteries Using Deep Learning.一种使用深度学习对主动脉和冠状动脉进行分割的计算高效方法。
IEEE Access. 2021 Jul 21;9:108873-108888. doi: 10.1109/ACCESS.2021.3099030. eCollection 2021.
2
HD-RDS-UNet: Leveraging Spatial-Temporal Correlation Between the Decoder Feature Maps for Lymphoma Segmentation.HD-RDS-Unet：利用解码器特征图之间的时空相关性进行淋巴瘤分割。
IEEE J Biomed Health Inform. 2022 Mar;26(3):1116-1127. doi: 10.1109/JBHI.2021.3102612. Epub 2022 Mar 7.
3
HF-UNet: Learning Hierarchically Inter-Task Relevance in Multi-Task U-Net for Accurate Prostate Segmentation in CT Images.
PA OmniNet：一种用于稳健光声图像重建的无需重新训练的通用深度学习框架。
Photoacoustics. 2025 Jul 8;45:100740. doi: 10.1016/j.pacs.2025.100740. eCollection 2025 Oct.
4
ProtoSAM-2D: 2D Semantic Segment Anything Model with Mask-Level Prototype-Learning and Distillation.ProtoSAM-2D：具有掩码级原型学习与蒸馏的二维语义分割通用模型
Proc SPIE Int Soc Opt Eng. 2025 Feb;13406. doi: 10.1117/12.3047044. Epub 2025 Apr 11.
5
Comparing UNet configurations for anthropogenic geomorphic feature extraction from land surface parameters.比较用于从地表参数中提取人为地貌特征的UNet配置。
PLoS One. 2025 Jun 10;20(6):e0325904. doi: 10.1371/journal.pone.0325904. eCollection 2025.
6
Developments in Deep Learning Artificial Neural Network Techniques for Medical Image Analysis and Interpretation.用于医学图像分析与解读的深度学习人工神经网络技术进展
Diagnostics (Basel). 2025 Apr 23;15(9):1072. doi: 10.3390/diagnostics15091072.
7
Advances in Deep Learning for Semantic Segmentation of Low-Contrast Images: A Systematic Review of Methods, Challenges, and Future Directions.低对比度图像语义分割的深度学习进展：方法、挑战及未来方向的系统综述
Sensors (Basel). 2025 Mar 25;25(7):2043. doi: 10.3390/s25072043.
8
Multi-scale conv-attention U-Net for medical image segmentation.用于医学图像分割的多尺度卷积注意力U型网络
Sci Rep. 2025 Apr 8;15(1):12041. doi: 10.1038/s41598-025-96101-8.
9
Advancements in Hysteroscopic Diagnosis and Management of Endometritis.宫腔镜检查诊断和治疗子宫内膜炎的进展
Diagnostics (Basel). 2025 Jan 21;15(3):243. doi: 10.3390/diagnostics15030243.
10
Artificial Intelligence-Based Classification and Segmentation of Bladder Cancer in Cystoscope Images.基于人工智能的膀胱镜图像中膀胱癌的分类与分割
Cancers (Basel). 2024 Dec 28;17(1):57. doi: 10.3390/cancers17010057.
HF-UNet：在 CT 图像中进行准确前列腺分割的多任务 U-Net 中学习层次化的任务间相关性
IEEE Trans Med Imaging. 2021 Aug;40(8):2118-2128. doi: 10.1109/TMI.2021.3072956. Epub 2021 Jul 30.
4
Image Segmentation Using Deep Learning: A Survey.基于深度学习的图像分割技术综述。
IEEE Trans Pattern Anal Mach Intell. 2022 Jul;44(7):3523-3542. doi: 10.1109/TPAMI.2021.3059968. Epub 2022 Jun 3.
5
COVID-19 lung CT image segmentation using deep learning methods: U-Net versus SegNet.使用深度学习方法进行COVID-19肺部CT图像分割：U-Net与SegNet对比
BMC Med Imaging. 2021 Feb 9;21(1):19. doi: 10.1186/s12880-020-00529-5.
6
FFU-Net: Feature Fusion U-Net for Lesion Segmentation of Diabetic Retinopathy.FFU-Net：用于糖尿病视网膜病变病变分割的特征融合 U-Net。
Biomed Res Int. 2021 Jan 2;2021:6644071. doi: 10.1155/2021/6644071. eCollection 2021.
7
Deep learning-based medical image segmentation with limited labels.基于深度学习的有限标签医学图像分割。
Phys Med Biol. 2020 Nov 20;65(23). doi: 10.1088/1361-6560/abc363.
8
Severity and Consolidation Quantification of COVID-19 From CT Images Using Deep Learning Based on Hybrid Weak Labels.基于混合弱标签的深度学习对 COVID-19 CT 图像严重程度和实变量化
IEEE J Biomed Health Inform. 2020 Dec;24(12):3529-3538. doi: 10.1109/JBHI.2020.3030224. Epub 2020 Dec 4.
9
Multi-Res-Attention UNet: A CNN Model for the Segmentation of Focal Cortical Dysplasia Lesions from Magnetic Resonance Images.多分辨率注意力 U-Net：一种用于从磁共振图像中分割局灶性皮质发育不良病变的 CNN 模型。
IEEE J Biomed Health Inform. 2021 May;25(5):1724-1734. doi: 10.1109/JBHI.2020.3024188. Epub 2021 May 11.
10
MI-UNet: Multi-Inputs UNet Incorporating Brain Parcellation for Stroke Lesion Segmentation From T1-Weighted Magnetic Resonance Images.MI-UNet：一种多输入 U 型网络，结合脑区划分用于从 T1 加权磁共振图像中分割中风病灶。
IEEE J Biomed Health Inform. 2021 Feb;25(2):526-535. doi: 10.1109/JBHI.2020.2996783. Epub 2021 Feb 5.