%0 Journal Article
%T SAM2U-Net：基于SAM2与U-Net的医学图像分割模型
SAM2U-Net: A U-Shaped Medical Image Segmentation Model Based on SAM2 and U-Net
%A 候俊伟
%A 刘磊
%J Modeling and Simulation
%P 337-347
%@ 2324-870X
%D 2025
%I Hans Publishing
%R 10.12677/mos.2025.143227
%X 医学图像分割在医学诊断中起着关键作用。尽管新兴视觉模型在各种医学分割任务中表现优异，但大多仅针对特定任务设计，缺乏普适性。本研究提出了一种新型SAM2学习模型，旨在实现通用医学图像分割。该模型基于U型架构，创新性地将SAM2的Hiera骨干网络与CNN模块并行结合，通过多尺度特征提取机制增强分割精度。在甲状腺结节诊断、结肠镜息肉分割等六个数据集上的实验表明，本模型在Dice系数和IoU上平均提高了1.46%，优于现有方法。结果证实，该模型能有效提取医学图像的病理特征，实现准确的区域分割，为广泛的临床诊断任务提供支持。
Medical image segmentation plays a crucial role in medical diagnosis. Although emerging visual models perform excellently in various medical segmentation tasks, most are designed for specific tasks and lack universality. This study proposes a novel SAM2 learning model aimed at achieving general medical image segmentation. The model is based on a U-shaped architecture and innovatively combines the Hiera backbone network of SAM2 with CNN modules in parallel, enhancing segmentation accuracy through a multi-scale feature extraction mechanism. Experiments on six datasets, including thyroid nodule diagnosis and colonoscopic polyp segmentation, demonstrate that this model improves the average Dice coefficient and IoU by 1.45% compared to existing methods. The results confirm that the model can effectively extract pathological features from medical images, achieving accurate regional segmentation and supporting a wide range of clinical diagnostic tasks.
%K 甲状腺结节，
%K 息肉分割，
%K 深度学习，
%K 超声图像，
%K 分割算法，
%K 医学图像，
%K U型架构，
%K Transformer模块，
%K 并行网络
Thyroid Nodule
%K Polyp Segmentation
%K Deep Learning
%K Ultrasound Imaging
%K Segmentation Algorithm
%K Medical Imaging
%K U-Shaped Architecture
%K Transformer Module
%K Parallel Network
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=109993