深度学习下的医学图像分割综述

doi:10.3778/j.issn.1002-8331.2409-0142

摘要/Abstract

摘要： 针对医学图像的高维性、复杂性和高精度要求等特性，深度学习下的医学图像分割方法凭借其强大的特征提取能力和对复杂模式的学习能力，能够自适应地从大量数据中学习并提取多层次特征，展现出高精度、高鲁棒性和可扩展性强等优势。通过端到端地提取感兴趣的器官、组织或病变区域，为医生在疾病诊断、制定治疗策略和临床研究等领域提供有力帮助。重点综述了U-Net、Transformer、Mamba、分割一切模型（segment anything model，SAM）及其各自变体模型在医学图像分割中的应用情况和发展脉络，从多个维度进行了综合对比分析，对开展医学影像研究、临床疾病诊断与治疗决策，以及医疗技术创新产品开发均具有一定参考价值。在此基础上，总结了目前医学图像分割研究中面临的挑战，并对该领域未来的研究前景进行展望。

关键词: 医学图像分割, 深度学习, U-Net, Transformer, Mamba, 分割一切模型（SAM）

Abstract: In response to the high-dimensional, complex nature and high-precision demands of medical images, deep learning-based segmentation methods excel in feature extraction and complex pattern recognition. These methods adaptively learn and extract multi-level features from vast datasets, demonstrating high accuracy, robustness, and scalability. By end-to-end extraction of organs, tissues, or lesion areas of interest, they provide substantial assistance to physicians in disease diagnosis, treatment planning, and clinical research. This review focuses on the application and development trajectory of U-Net, Transformer, Mamba, and segment anything model (SAM) and their variants in medical image segmentation, offering a comprehensive comparative analysis across multiple dimensions. It holds reference value for medical imaging research, clinical diagnosis and treatment decision-making, and the development of innovative medical technology products. Building on this, the review summarizes current challenges in medical image segmentation research and prospects the future research landscape.

Key words: medical image segmentation, deep learning, U-Net, Transformer, Mamba, segment anything model (SAM)

邢素霞, 李珂娴, 方俊泽, 郭正, 赵士杭. 深度学习下的医学图像分割综述[J]. 计算机工程与应用, 2025, 61(7): 25-41.

XING Suxia, LI Kexian, FANG Junze, GUO Zheng, ZHAO Shihang. Survey of Medical Image Segmentation in Deep Learning[J]. Computer Engineering and Applications, 2025, 61(7): 25-41.

参考文献

2024.
[49] ZHANG M Y, YU Y, GU L M, et al. VM-UNET-V2 rethinking vision Mamba UNet for medical image segmentation[J]. arXiv:2403.09157, 2024.
[50] WANG Z, ZHENG J Q, ZHANG Y, et al. Mamba-UNet: UNet-like pure visual Mamba for medicalimage segmentation[J]. arXiv:2402.07245, 2024.
[51] LIU J R, YANG H, ZHOU H Y, et al. Swin-UMamba: Mamba-based UNet with ImageNet-based pretraining[J]. arXiv: 2402. 03302, 2024.
[52] WU R K, LIU Y H, LIANG P C, et al. UltraLight VM-UNet: parallel vision Mamba significantly reduces parameters for skin lesion segmentation[J]. arXiv:2403.20035, 2024.
[53] XING Z H, YE T, YANG Y J, et al. SegMamba: long-range sequential modeling Mamba for 3D medical image segmentation[J]. arXiv:2401.13560, 2024.
[54] WANG J, CHEN J, CHEN D, et al. Large window-based Mamba unet for medical image segmentation: beyond convolution and self-attention[J]. arXiv:2305.03678, 2023.
[55] ZHANG Y, JIAO R. Towards segment anything model(SAM) for medical image segmentation: a survey[J]. arXiv:2305.03678, 2023.
[56] WANG D, ZHANG J, DU B, et al. SAMRS: scaling-up remote sensing segmentation dataset with segment anything model[J]. arXiv:2305.02034, 2023.
[57] TANG L, XIAO H K, LI B. Can SAM segment anything? when SAM meets camouflaged object detection[J]. arXiv: 2304.04709, 2023.
[58] HE S, BAO R N, LI J P, et al. Computer-vision benchmark segment-anything model(SAM) in medical images: accuracy in 12 datasets[J]. arXiv:2304.09324, 2023.
[59] HU M Z, LI Y H, YANG X F. SkinSAM: empowering skin cancer segmentation with segment anything model[J]. arXiv:2304.13973, 2023.
[60] LI Y H, HU M Z, YANG X F. Polyp-SAM: transfer SAM for polyp segmentation[J]. arXiv:2305.00293, 2023.
[61] MA J, HE Y T, LI F F, et al. Segment anything in medical images[J]. Nature Communications, 2024, 15(1): 654.
[62] CHEN L C, PAPANDREOU G, SCHROFF F, et al. Rethinking atrous convolution for semantic image segmentation[J]. arXiv:1706.05587, 2017.
[63] WU J D, JI W, LIU Y P, et al. Medical SAM adapter: adapting segment anything model for medical image segmentation[J]. arXiv:2304.12620, 2023.
[64] CHENG J L, YE J, DENG Z Y, et al. SAM-Med2D[J]. arXiv:2308.16184, 2023.
[65] ZHANG J W, MA K, KAPSE S, et al. SAM-path: a segment anything model for semantic segmentation in digital pathology[C]//Proceedings of the International Conference on Medical Image Computing and Computer-Assisted Intervention. Cham: Springer, 2023: 161-170.
[66] WANG H, GUO S, YE J, et al. SAM-Med3D[J]. arXiv: 2310.15161, 2023.
[67] ZHANG Y C, YANG J, LIU Y C, et al. SemiSAM: enhancing semi-supervised medical image segmentation via SAM-assisted consistency regularization[J]. arXiv:2312.06316, 2023.
[68] GOODFELLOW I, POUGET-ABADIE J, MIRZA M, et al. Generative adversarial networks[J]. Communications of the ACM, 2020, 63(11): 139-144.
[69] GUO J X, LU S D, CAI H, et al. Long text generation via adversarial training with leaked information[J]. arXiv:1709.
08624, 2017.
[70] LI S J, ZHANG Y F, YANG X. Semi-supervised cardiac MRI segmentation based on generative adversarial network and variational auto-encoder[C]//Proceedings of the 2021 IEEE International Conference on Bioinformatics and Biomedicine, 2021: 1402-1405.
[71] HUNG W C, TSAI Y H, LIOU Y T, et al. Adversarial learning for semi-supervised semantic segmentation[J]. arXiv: 1802.07934, 2018.
[72] GUAN H, YAP P T, BOZOKI A, et al. Federated learning for medical image analysis: a survey[J]. Pattern Recognition, 2024, 151: 110424.
[73] LI Y Q, WANG D D, YUAN C, et al. Enhancing agricultural image segmentation with an agricultural segment anything model adapter[J]. Sensors, 2023, 23(18): 7884.
[74] LI G J, HUANG Q H, WANG W, et al. Selective and multi-scale fusion Mamba for medical image segmentation[J]. Expert Systems with Applications, 2025, 261: 125518.
[75] PUTZ F, BEIRAMI S, SCHMIDT M A, et al. The segment anything foundation model achieves favorable brain tumor auto-segmentation accuracy in MRI to support radiotherapy treatment planning[J]. arXiv:2304.07875, 2023.
[76] 孙兴, 蔡肖红, 李明, 等. 视觉大模型SAM在医学图像分割中的应用综述[J]. 计算机工程与应用, 2024, 60(17): 1-16.
SUN X, CAI X H, LI M, et al. Review of application of visual foundation model SAM in medical image segmentation[J]. Computer Engineering and Applications, 2024, 60(17): 1-16.
[77] HUANG Z, BIANCHI F, YUKSEKGONUL M, et al. A visual-language foundation model for pathology image analysis using medical Twitter[J]. Nature Medicine, 2023, 29(9): 2307-2316.

[1]	甄彤, 张威振, 李智慧. 遥感影像中种植作物结构分类方法综述[J]. 计算机工程与应用, 2025, 61(8): 35-48.
[2]	李仝伟, 仇大伟, 刘静, 逯英航. 基于RGB与骨骼数据的人体行为识别综述[J]. 计算机工程与应用, 2025, 61(8): 62-82.
[3]	温浩, 杨洋. 融合ERNIE与知识增强的临床短文本分类研究[J]. 计算机工程与应用, 2025, 61(8): 108-116.
[4]	王燕, 卢鹏屹, 他雪. 结合特征融合注意力的规范化卷积图像去雾网络[J]. 计算机工程与应用, 2025, 61(8): 226-238.
[5]	周佳妮, 刘春雨, 刘家鹏. 融合通道与多头注意力的股价趋势预测模型[J]. 计算机工程与应用, 2025, 61(8): 324-338.
[6]	白雪冰, 车进, 吴金蔓. 多尺度特征融合的图像描述算法[J]. 计算机工程与应用, 2025, 61(7): 288-296.
[7]	史昕, 王浩泽, 纪艺, 马峻岩. 融合时空特征的多模态车辆轨迹预测方法[J]. 计算机工程与应用, 2025, 61(7): 325-333.
[8]	陈宇, 权冀川. 伪装目标检测：发展与挑战[J]. 计算机工程与应用, 2025, 61(7): 42-60.
[9]	翟慧英, 郝汉, 李均利, 占志峰. 铁路设施无人机自主巡检算法研究综述[J]. 计算机工程与应用, 2025, 61(7): 61-80.
[10]	韩佰轩, 彭月平, 郝鹤翔, 叶泽聪. DMU-YOLO：机载视觉的多类异常行为检测算法[J]. 计算机工程与应用, 2025, 61(7): 128-140.
[11]	刘红玉, 高见. 融合CBAM的违法犯罪类安卓恶意软件检测与分类模型研究[J]. 计算机工程与应用, 2025, 61(6): 317-327.
[12]	王文举, 唐邦, 顾泽骅, 王森. 深度学习的多视角三维重建技术综述[J]. 计算机工程与应用, 2025, 61(6): 22-35.
[13]	孙宇, 刘川, 周扬. 深度学习在知识图谱构建及推理中的应用[J]. 计算机工程与应用, 2025, 61(6): 36-52.
[14]	王敬凯, 秦董洪, 白凤波, 李路路, 孔令儒, 徐晨. 语音识别与大语言模型融合技术研究综述[J]. 计算机工程与应用, 2025, 61(6): 53-63.
[15]	侯颖, 胡鑫, 赵瑞瑞, 张楠, 徐艳红, 马莉. 感兴趣区域YOLO_BFROI的扶梯乘客安全检测算法[J]. 计算机工程与应用, 2025, 61(6): 84-95.