融合多尺度特征和注意力机制的肺炎病灶分割方法

Pneumonia lesion segmentation method integrating multi-scale feature fusion and attention mechanisms

摘要

针对肺炎病灶区域大小形状差异大、区域边界不清晰等问题，提出一种融合多尺度特征和注意力机制的分割模型，该模型由多尺度视觉状态空间块构成编码器与解码器的主要结构，利用级联的不同大小的卷积核捕捉不同病灶的多尺度上下文信息，有效应对肺炎病灶形状大小的显著变异性。同时，在跳跃连接层中引入CBAM模块，通过空间和通道维度的双分支交互机制，增强编码器与解码器特征的语义一致性，减少特征融合时的信息丢失。在多个公开肺部病理图像数据集上进行实验，在Large COVID-19 CT scan slice与CNCB两个肺炎病灶分割数据集上的Dice相似系数分别达到87.7%与85.1%，IOU分别达到79.7%与75.3%，并在LUNA16肺结节数据集上验证模型的泛化能力，实验结果表明，相较于传统U-Net架构与transformer架构该模型分割结果具有显著提升。

Abstract

A segmentation model integrating multi-scale features and attention mechanisms is developed to address the challenges of large variations in the size and shape of pneumonia lesions as well as unclear boundaries. The model employs multi-scale visual state space blocks as the main components of both the encoder and decoder. It utilizes cascaded convolution kernels of different sizes to capture multi-scale contextual information from varying-sized lesions, thereby effectively coping with the significant variations in the shape and scale of pneumonia lesions. Meanwhile, a convolutional block attention module is embedded in the skip connection layers. Through a dual-branch interaction mechanism across spatial and channel dimensions, the semantic consistency of encoder-decoder features is improved, and the information loss during feature fusion is alleviated. Experiments are conducted on several public lung pathological image datasets. On the Large COVID-19 CT scan slice dataset and the CNCB pneumonia lesion segmentation dataset, the proposed model achieves Dice similarity coefficients of 87.7% and 85.1%, and IoU of 79.7% and 75.3%, respectively. In addition, the generalization performance of the model is verified on the LUNA16 lung nodule dataset. Experimental results demonstrate that the proposed model exhibits superior segmentation performance over conventional U-Net architectures and Transformer-based models.

关键词

肺炎病灶 / 图像分割 / 特征融合 / 多尺度 / 注意力机制 / 医学图像处理

Key words

pneumonia lesion / image segmentation / feature fusion / multi-scale / mixed attention / medical image processing

参考文献

[1] Machkovech HM, Hahn AM, Garonzik Wang J, et al. Persistent SARS—CoV—2 infection: significance and implications[J]. Lancet Infect Dis, 2024, 24(7): e453-e462.

[2] Oulefki A, Agaian S, Trongtirakul T, et al. Automatic COVID—19 lung infected region segmentation and measurement using CT—scans images[J]. Pattern Recognit, 2021, 114: 107747.

[3] Hertel R, Benlamri R . Deep learning techniques for COVID—19 diagnosis and prognosis based on radiological imaging[J]. ACM Comput Surv, 2023, 55(12): 260.

[4] Ehab W, Huang LN, Li YM . UNet and variants for medical image segmentation[J]. Int J Netw Dyn Intell, 2024, 3: 100009.

[5] 陈梦飞, 王娆芬, 王海玲, 等 . 基于多尺度卷积与并行反向注意力的医学图像分割[J]. 中国医学物理学杂志, 2025, 42(1): 27-36.

[6] Chen MF, Wang RF, Wang HL, et al. Medical image segmentation based on multi—scale convolution and parallel reverse attention[J]. Chinese Journal of Medical Physics, 2025, 42(1): 27-36.

[7] Alshomrani S, Arif M, Al Ghamdi MA . SAA—UNet: spatial attention and attention gate UNet for COVID—19 pneumonia segmentation from computed tomography[J]. Diagnostics, 2023, 13(9): 1658.

[8] Huang XP, Chen JX, Chen MZ, et al. TDD—UNet: transformer with double decoder UNet for COVID—19 lesions segmentation[J]. Comput Biol Med, 2022, 151(Pt A): 106306.

[9] Zhang SL, Liu JY, Qian TY, et al. Prompt—guided dual—path UNet with Mamba for medical image segmentation[EB/OL]. (2025—03—25)[2026—01—10]. https://arxiv.org/abs/2503.19589.

[10] Wang ZY, Zheng JQ, Zhang YC, et al. Mamba—UNet: UNet—like pure visual mamba for medical image segmentation[EB/OL]. (2024—03—30)[2026—01—10]. https://arxiv.org/abs/2402.05079.

[11] Yang XF, Luo ZY, Wu YL, et al. TMU: transmission—enhanced mamba—UNet for medical image segmentation[C]// Advanced Intelligent Computing Technology and Applications. Singapore: Springer Nature Singapore, 2024: 428-438.

[12] He KM, Zhang XY, Ren SQ, et al. Deep residual learning for image recognition[C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ, USA: IEEE, 2016: 770-778.

[13] Ramachandran P, Zoph B, Le QV . Searching for activation functions[EB/OL]. (2017—10—17)[2026—01—10]. https://arxiv.org/abs/1710.05941.

[14] Chollet F. Xception: deep learning with depthwise separable convolutions[C]// 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Piscataway, NJ, USA: IEEE, 2017: 1800-1807.

[15] Hendrycks D, Gimpel K . Gaussian error linear units (GELUs)[EB/OL]. (2023—06—06)[2026—01—10]. https://arxiv.org/abs/1606.08415.

[16] Woo S, Park J, Lee JY, et al. CBAM: convolutional block attention module[C]// Computer Vision—ECCV 2018. Cham: Springer International Publishing, 2018: 3-19.

[17] Maftouni M, Law ACC, Shen B, et al. A robust ensemble—deep learning model for COVID—19 diagnosis based on an integrated CT scan images database[C]// IISE Annual Conference and Expo 2021. Norcross, GA, USA: Institute of Industrial and Systems Engineers, 2021: 632-637.

[18] Zhang K, Liu XH, Shen J, et al. Clinically applicable AI system for accurate diagnosis, quantitative measurements, and prognosis of COVID—19 pneumonia using computed tomography[J]. Cell, 2020, 181(6): 1423-1433.e11.

[19] Setio AAA, Traverso A, de Bel T, et al. Validation, comparison, and combination of algorithms for automatic detection of pulmonary nodules in computed tomography images: the LUNA16 challenge[J]. Med Image Anal, 2017, 42: 1-13.

[20] Ronneberger O, Fischer P, Brox T . U—Net: convolutional networks for biomedical image segmentation[C]// Medical Image Computing and Computer—Assisted Intervention—MICCAI 2015. Cham: Springer International Publishing, 2015: 234-241.

[21] Diakogiannis FI, Waldner F, Caccetta P, et al. ResUNet—a: a deep learning framework for semantic segmentation of remotely sensed data[J]. ISPRS J Photogramm Remote Sens, 2020, 162: 94-114.

[22] Isensee F, Jaeger PF, Kohl SAA, et al. nnU—Net: a self—configuring method for deep learning—based biomedical image segmentation[J]. Nat Methods, 2021, 18(2): 203-211.

[23] Chen JN, Mei JR, Li XH, et al. TransUNet: rethinking the U—Net architecture design for medical image segmentation through the lens of transformers[J]. Med Image Anal, 2024, 97: 103280.

[24] Cao H, Wang YY, Chen J, et al. Swin—Unet: Unet—like pure transformer for medical image segmentation[C]// Computer Vision—ECCV 2022 Workshops. Cham: Springer Nature Switzerland, 2023: 205-218.

基金资助

上海市科学技术委员会地方院校能力建设项目(23010502700)