分割一切模型(SAM)在医学图像分割中的应用  

Application of Segment Anything Model in Medical Image Segmentation

作　　者：吴曈 胡浩基[1,2] 冯洋 罗琼[4] 徐栋[5,6] 郑伟增[7] 金能 杨琛 姚劲草[5,6] Wu Tong;Hu Haoji;Feng Yang;Luo Qiong;Xu Dong;Zheng Weizeng;Jin Neng;Yang Chen;Yao Jincao(University of Illinois Urbana-Champaign Institute,Zhejiang University,Hangzhou 314400,Zhejiang,China;College of Information Science and Electronic Engineering,Zhejiang University,Hangzhou 310027,Zhejiang,China;Angelalign Research Institute,Anglealign Technology Inc.,200433,Shanghai,China;Department of Obstetrics,Women’s Hospital School of Medicine Zhejiang University,Hangzhou 310006,Zhejiang,China;Zhejiang Cancer Hospital,Hangzhou 331022,Zhejiang,China;Hangzhou Institute of Medicine(HIM),Chinese Academy of Sciences,Hangzhou 310000,Zhejiang,China;Department of Radiology,Women’s Hospital School of Medicine Zhejiang University,Hangzhou 310006,Zhejiang,China)

机构地区：[1]浙江大学伊利诺伊大学厄巴纳香槟校区联合学院,浙江杭州314400 [2]浙江大学信息与电子工程学院,浙江杭州310027 [3]上海时代天使医疗器械有限公司天使研究院,上海200433 [4]浙江大学医学院附属妇产科医院产科,浙江杭州310006 [5]浙江省肿瘤医院,浙江杭州331022 [6]中国科学院杭州医学研究所,浙江杭州310000 [7]浙江大学医学院附属妇产科放射科,浙江杭州310006

出　　处：《中国激光》2024年第21期19-34,共16页Chinese Journal of Lasers

基　　金：国家自然科学基金(U21B2004,62106222);浙江省重点研发计划(2021C01119);浙江省自然科学基金(LZ23F020008);浙江大学-时代天使智慧健康研究项目。

摘　　要：医学图像分割是计算机辅助医疗流程中的关键步骤,精准的医学图像分割可以为诊断与治疗提供帮助。分割一切模型(SAM)利用提示驱动的基础大模型进行下游的分割任务,它的出现为医学图像分割提出了与神经网络不同的新方向。但是,SAM是以自然图像为基础的模型,对医学图像的处理效果还有待提高。本文介绍了SAM在医学图像上直接应用的效果,并总结了将SAM应用到医学图像分割任务的研究工作。与此同时,介绍了本课题组在乳腺肿瘤数据集与孕妇骨盆数据集上进行的两个实验,验证了大模型经过大量数据微调后具有更好的泛化能力。半监督网络与SAM结合生成高质量的伪标签能够有效提高分割效果。虽然目前SAM在医学图像分割领域已取得较好效果,但进一步提升存在一定困难。本文最后分析了SAM面临的挑战并讨论了SAM在医学图像分割中的潜在发展方向,希望有助于医疗分割技术的进步。Significance The application of deep neural networks to image segmentation is one of the most prevalent topics in medical imaging.As an initial step in computer-aided detection processes,medical image segmentation aims to identify contours or regions of interest within images,thereby providing valuable assistance to clinicians in image interpretation,surgical planning,and clinical decision-making.Deep neural networks,which leverage their powerful ability to learn complex image features,have demonstrated outstanding performance in medical image segmentation.However,the use of deep neural networks for medical image segmentation has two significant limitations.First,different medical imaging modalities and specific segmentation tasks exhibit diverse image characteristics,leading to the low generalization capabilities of deep neural networks,which are often tailored to specific tasks.Second,increasingly complex network architectures with notable segmentation efficacy demand significant amounts of annotated image data,particularly those that require laborious manual annotation by medical experts.With the rapid advancement of large-scale pretrained foundation models(LPFMs)in the field of artificial intelligence,an increasing number of tasks have achieved superior results through the fine-tuning of LPFMs.LPFMs are generic models trained on massive amounts of data and acquire foundational and versatile representational capabilities that can be transferred across different domains.Consequently,various downstream tasks can be easily fine-tuned using universal models.Considering the challenges in medical image segmentation,including low model generalization and difficulty in dataset acquisition,universal LPFMs are urgently needed in the field of medical image segmentation to facilitate breakthroughs in artificial intelligence applied to medical imaging.Since its introduction as a foundational large model in the field of natural image segmentation,the segment anything model(SAM)has been applied across various domains with re

关 键 词：分割一切模型 医学图像分割 基础模型 深度学习 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]