
浏览全部资源
扫码关注微信
1.贵州大学 电气工程学院,贵州 贵阳 550025
2.贵阳市公共卫生救治中心,贵州 贵阳 550000
3.贵阳市花溪区人民医院,贵州 贵阳 550025
闫小丽,女,硕士研究生,2025年于河南工程学院获得学士学位,主要从事医学影像处理方面的研究。E-mail: gs.xlyan25@gzu.edu.cn
何志琴,女,硕士,教授,2005年于贵州大学获得硕士学位,主要从事人工智能、机器人控制以及医学影像诊断方面的研究。E-mail: zqhe@gzu.edu.cn
收稿:2026-08-02,
修回:2026-08-19,
纸质出版:2026-09-05
移动端阅览
闫小丽, 张蔚, 刘庭亭, 等. 基于多视角注意力融合的肺结核影像分类[J]. 液晶与显示, 2026,41(9):1339-1351.
YAN Xiaoli, ZHANG Wei, LIU Tingting, et al. Classification of pulmonary tuberculosis images based on multi-view attention fusion[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(9): 1339-1351.
闫小丽, 张蔚, 刘庭亭, 等. 基于多视角注意力融合的肺结核影像分类[J]. 液晶与显示, 2026,41(9):1339-1351. DOI: 10.3724/CJLCD.2026-0148. CSTR: 32172.14.CJLCD.2026-0148.
YAN Xiaoli, ZHANG Wei, LIU Tingting, et al. Classification of pulmonary tuberculosis images based on multi-view attention fusion[J]. Chinese Journal of Liquid Crystals and Displays, 2026, 41(9): 1339-1351. DOI: 10.3724/CJLCD.2026-0148. CSTR: 32172.14.CJLCD.2026-0148.
肺结核计算机断层扫描(Computed Tomography, CT)影像存在噪声干扰、病灶特征弱及三维空间信息利用不足等问题,影响分类性能。针对上述问题,本文提出一种融合图像增强与多维特征学习的肺结核CT三分类方法。该方法采用多尺度八方向各向异性扩散增强与渐进式N2F(Noise2Fast)去噪联合预处理策略,提高病灶纹理可见性;随后构建轻量化PRISM-Net(Parallel Representation with Iterative Spatial Multi-view Fusion Network),通过迭代式多视角融合策略(IMVF)挖掘不同空间方向的互补信息,并引入多维注意力模块(MDAM)增强三维特征表达能力。在150例独立测试集上进行验证,所提方法精确率达到94.68%,加权F1值达到94.63%,正常、活动性及非活动性肺结核类别AUC分别达到1.000 0、0.992 6和0.992 3,性能优于MambaOut-Kobe、MobileNetV3、MedVit及ResNet50等模型。
Computed Tomography(CT) images of pulmonary tuberculosis suffer from issues including noise interference, weak lesion features, and insufficient utilization of three-dimensional spatial information, which impair classification performance. To address these problems, this paper proposes a three-class classification method for pulmonary tuberculosis CT images that integrates image enhancement and multi-dimensional feature learning. This method adopts a joint preprocessing strategy combining multi-scale eight-direction anisotropic diffusion enhancement and progressive N2F (Noise2Fast) denoising, which improves the visibility of lesion textures. Subsequently, a lightweight Parallel Representation with PRISM-Net(Iterative Spatial Multi-view Fusion Network) is constructed. The network utilizes an Iterative Multi-view Fusion(IMVF) strategy to extract complementary information from different spatial orientations, and introduces a MDAM (Multi-Dimensional Attention Module) to enhance the expression capability of three-dimensional features. Validated on an independent test set comprising 150 cases, the proposed method achieves a precision of 94.68% and a weighted F1-score of 94.63%. The AUC (Area Under the Curve) for normal samples, active pulmonary tuberculosis, and inactive pulmonary tuberculosis reaches 1.000 0, 0.992 6, and 0.992 3 respectively, outperforming benchmark models including MambaOut-Kobe, MobileNetV3, MedVit and ResNet50.
World Health Organization . Global tuberculosis report 2023 [R]. Geneva : World Health Organization , 2023 .
NIJIATI M , GUO L , ABULIZI A , et al . Deep learning and radiomics of longitudinal CT scans for early prediction of tuberculosis treatment outcomes [J]. European Journal of Radiology , 2023 , 169 : 111180 . doi: 10.1016/j.ejrad.2023.111180 http://dx.doi.org/10.1016/j.ejrad.2023.111180
马依迪丽·尼加提 , 米日古丽·达毛拉 , 张斌 , 等 . 基于CT深度学习的活动性/非活动性肺结核分类模型构建及验证 [J]. 暨南大学学报(自然科学与医学版) , 2023 , 44 ( 1 ): 69 - 77,86 .
NIJIATI·MAYIDILI , DAMAOLA·MIRIGULI , ZHANG B , et al . Establishment and verification of active/non-active tuberculosis classification model based on CT deep learning [J]. Journal of Jinan University (Natural Science & Medicine Edition) , 2023 , 44 ( 1 ): 69 - 77, 86 . (in Chinese)
WU N F , SUN Y X , HU J T , et al . Intelligent nanophotonics: when machine learning sheds light [J]. eLight , 2025 , 5 ( 1 ): 5 . doi: 10.1186/s43593-025-00085-x http://dx.doi.org/10.1186/s43593-025-00085-x
万丙辰 , 张选德 . 基于元学习和失真感知的图像质量评价 [J]. 液晶与显示 , 2024 , 39 ( 11 ): 1519 - 1531 . doi: 10.37188/cjlcd.2024-0141 http://dx.doi.org/10.37188/cjlcd.2024-0141
WAN B C , ZHANG X D . Image quality assessment based on meta-learning and distortion perception [J]. Chinese Journal of Liquid Crystals and Displays , 2024 , 39 ( 11 ): 1519 - 1531 . (in Chinese) . doi: 10.37188/cjlcd.2024-0141 http://dx.doi.org/10.37188/cjlcd.2024-0141
CHENG S Y , CHANG S B , LI Y Z , et al . Enhanced multiscale human brain imaging by semi-supervised digital staining and serial sectioning optical coherence tomography [J]. Light: Science & Applications , 2025 , 14 ( 1 ): 57 . doi: 10.1038/s41377-024-01658-0 http://dx.doi.org/10.1038/s41377-024-01658-0
ZHANG Z L , KONG L Y , ZHANG L H Y , et al . Structured light meets machine intelligence [J]. eLight , 2025 , 5 ( 1 ): 26 . doi: 10.1186/s43593-025-00102-z http://dx.doi.org/10.1186/s43593-025-00102-z
梁礼明 , 朱晨锟 , 阳渊 , 等 . 跨级特征自适应融合的暗光图像增强算法 [J]. 液晶与显示 , 2024 , 39 ( 6 ): 856 - 866 . doi: 10.37188/cjlcd.2023-0210 http://dx.doi.org/10.37188/cjlcd.2023-0210
LIANG L M , ZHU C K , YANG Y , et al . Cross-level feature adaptive fusion network for low-light image enhancement [J]. Chinese Journal of Liquid Crystals and Displays , 2024 , 39 ( 6 ): 856 - 866 . (in Chinese) . doi: 10.37188/cjlcd.2023-0210 http://dx.doi.org/10.37188/cjlcd.2023-0210
KAMATH V , BHAT V G , RAJU G , et al . Application of fluorescence lifetime imaging-integrated deep learning analysis for cancer research [J]. Light: Advanced Manufacturing , 2025 , 6 ( 3 ): 522 - 545 . doi: 10.37188/lam.2025.049 http://dx.doi.org/10.37188/lam.2025.049
ESTEVA A , CHOU K , YEUNG S , et al . Deep learning-enabled medical computer vision [J]. npj Digital Medicine , 2021 , 4 ( 1 ): 5 .
ARDILA D , KIRALY A P , BHARADWAJ S , et al . End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography [J] . Nature Medicine , 2019 , 25 ( 6 ): 954 - 961 . doi: 10.1038/s41591-019-0447-x http://dx.doi.org/10.1038/s41591-019-0447-x
SIDDIQUI E A , CHAURASIA V , SHANDILYA M . Classification of lung cancer computed tomography images using a 3-dimensional deep convolutional neural network with multi-layer filter [J]. Journal of Cancer Research and Clinical Oncology , 2023 , 149 ( 13 ): 11279 - 11294 .
YADAV V K , SINGHAI J . Adaptive gamma correction for automatic contrast enhancement of chest-X-ray images affected by various lung diseases [J]. Multimedia Tools and Applications , 2024 , 83 ( 29 ): 73457 - 73475 . doi: 10.1007/s11042-023-18083-x http://dx.doi.org/10.1007/s11042-023-18083-x
YANG S H , PU Q , LEI C T , et al . Low-dose CT denoising with a high-level feature refinement and dynamic convolution network [J]. Medical Physics , 2023 , 50 ( 6 ): 3597 - 3611 . doi: 10.1002/mp.16175 http://dx.doi.org/10.1002/mp.16175
FU M H , DUAN Y H , CHENG Z P , et al . Total-body low-dose CT image denoising using a prior knowledge transfer technique with a contrastive regularization mechanism [J]. Medical Physics , 2023 , 50 ( 5 ): 2971 - 2984 . doi: 10.1002/mp.16163 http://dx.doi.org/10.1002/mp.16163
LI Z Y , LIU Y , CHEN Y , et al . Dual-domain fusion deep convolutional neural network for low-dose CT denoising [J]. Journal of X-Ray Science and Technology , 2023 , 31 ( 4 ): 757 - 775 . doi: 10.3233/xst-230020 http://dx.doi.org/10.3233/xst-230020
LI H , NAN Y , DEL SER J , et al . Large-kernel attention for 3D medical image segmentation [J]. Cognitive Computation , 2024 , 16 ( 4 ): 2063 - 2077 .
PERONA P , MALIK J . Scale-space and edge detection using anisotropic diffusion [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 1990 , 12 ( 7 ): 629 - 639 . doi: 10.1109/34.56205 http://dx.doi.org/10.1109/34.56205
CHEN H , ZHANG Y , KALRA M K , et al . Low-dose CT with a residual encoder-decoder convolutional neural network [J]. IEEE Transactions on Medical Imaging , 2017 , 36 ( 12 ): 2524 - 2535 . doi: 10.1109/tmi.2017.2715284 http://dx.doi.org/10.1109/tmi.2017.2715284
张建强 , 何秋生 . 上下文感知的低光图像增强算法 [J]. 液晶与显示 , 2025 , 40 ( 5 ): 751 - 760 . doi: 10.37188/cjlcd.2024-0277 http://dx.doi.org/10.37188/cjlcd.2024-0277
ZHANG J Q , HE Q S . Context aware low-light image enhancement algorithm [J]. Chinese Journal of Liquid Crystals and Displays , 2025 , 40 ( 5 ): 751 - 760 . (in Chinese) . doi: 10.37188/cjlcd.2024-0277 http://dx.doi.org/10.37188/cjlcd.2024-0277
LEQUYER J , PHILIP R , SHARMA A , et al . A fast blind zero-shot denoiser [J]. Nature Machine Intelligence , 2022 , 4 ( 11 ): 953 - 963 . doi: 10.1038/s42256-022-00547-8 http://dx.doi.org/10.1038/s42256-022-00547-8
HUANG Z , LI Y , WANG Y , et al . Lightweight three-dimensional convolutional neural network for medical image classification [J]. IEEE Journal of Biomedical and Health Informatics , 2023 , 27 ( 8 ): 3955 - 3966 .
周涛 , 程倩茹 , 张祥祥 , 等 . 基于DCIF-GAN的肺部肿瘤PET/CT跨模态医学图像融合 [J]. 光学 精密工程 , 2024 , 32 ( 2 ): 221 - 236 . doi: 10.37188/ope.20243202.0221 http://dx.doi.org/10.37188/ope.20243202.0221
ZHOU T , CHENG Q R , ZHANG X X , et al . PET/CT cross-modal medical image fusion of lung tumors based on DCIF-GAN [J]. Optics and Precision Engineering , 2024 , 32 ( 2 ): 221 - 236 . (in Chinese) . doi: 10.37188/ope.20243202.0221 http://dx.doi.org/10.37188/ope.20243202.0221
邸敬 , 郭文庆 , 任莉 , 等 . 基于结构功能交叉神经网络的多模态医学图像融合 [J]. 光学 精密工程 , 2024 , 32 ( 2 ): 252 - 267 . doi: 10.37188/ope.20243202.0252 http://dx.doi.org/10.37188/ope.20243202.0252
DI J , GUO W Q , REN L , et al . Multimodal medical image fusion method based on structural functional cross neural network [J]. Optics and Precision Engineering , 2024 , 32 ( 2 ): 252 - 267 . (in Chinese) . doi: 10.37188/ope.20243202.0252 http://dx.doi.org/10.37188/ope.20243202.0252
WOO S , PARK J , LEE J Y , et al . CBAM: convolutional block attention module [C]// Proceedings of the 15th European Conference on Computer Vision . Munich : Springer , 2018 : 3 - 19 . doi: 10.1007/978-3-030-01234-2_1 http://dx.doi.org/10.1007/978-3-030-01234-2_1
闫纲琦 , 梁宗林 , 宋延嵩 , 等 . 基于注意力残差网络的快照式多光谱相机图像重构 [J]. 中国光学(中英文) , 2024 , 17 ( 6 ): 1316 - 1328 . doi: 10.37188/co.2023-0196 http://dx.doi.org/10.37188/co.2023-0196
YAN G Q , LIANG Z L , SONG Y S , et al . Image reconstruction of snapshot multispectral camera based on an attention residual network [J]. Chinese Optics , 2024 , 17 ( 6 ): 1316 - 1328 . (in Chinese) . doi: 10.37188/co.2023-0196 http://dx.doi.org/10.37188/co.2023-0196
邸敬 , 梁婵 , 刘冀钊 , 等 . 基于跨域交互注意力和对比学习引导的红外与可见光图像融合 [J]. 中国光学(中英文) , 2025 , 18 ( 2 ): 317 - 332 . doi: 10.37188/co.2024-0147 http://dx.doi.org/10.37188/co.2024-0147
DI J , LIANG C , LIU J Z , et al . Infrared and visible image fusion guided by cross-domain interactive attention and contrastive learning [J]. Chinese Optics , 2025 , 18 ( 2 ): 317 - 332 . (in Chinese) . doi: 10.37188/co.2024-0147 http://dx.doi.org/10.37188/co.2024-0147
朱新军 , 赵浩淼 , 王红一 , 等 . 基于轻型自限制注意力的结构光相位及深度估计混合网络 [J]. 中国光学(中英文) , 2024 , 17 ( 1 ): 118 - 127 . doi: 10.37188/co.2023-0066 http://dx.doi.org/10.37188/co.2023-0066
ZHU X J , ZHAO H M , WANG H Y , et al . A hybrid network based on light self-limited attention for structured light phase and depth estimation [J]. Chinese Optics , 2024 , 17 ( 1 ): 118 - 127 . (in Chinese) . doi: 10.37188/co.2023-0066 http://dx.doi.org/10.37188/co.2023-0066
HENDRIX W , HENDRIX N , SCHOLTEN E T , et al . Deep learning for the detection of benign and malignant pulmonary nodules in non-screening chest CT scans [J]. Communications Medicine , 2023 , 3 ( 1 ): 156 . doi: 10.1038/s43856-023-00388-5 http://dx.doi.org/10.1038/s43856-023-00388-5
MAFTOUNI M , LAW A C C , SHEN B . A robust ensemble-deep learning model for COVID-19 diagnosis based on an integrated CT scan images database [C]// Proceedings of the IISE Annual Conference and Expo 2021 . Online : IISE , 2021 : 632 - 637 .
JENSSEN H B , SAKINIS T . MedSeg COVID dataset 1 [DB/OL]. ( 2021-01-05 ). https://doi.org/10.6084/m9.figshare.13521488 https://doi.org/10.6084/m9.figshare.13521488 .
上官泓廷 , 刘玉红 . 医学图像增强方法综述 [J]. 中国医学物理学杂志 , 2023 , 40 ( 4 ): 410 - 415 .
SHANGGUAN H T , LIU Y H . A review of medical image enhancement methods [J]. Chinese Journal of Medical Physics , 2023 , 40 ( 4 ): 410 - 415 . (in Chinese)
LIUFU Y L , SU R H , WEN Y H , et al . A CT-based deep learning approach to differentiate multiple primary lung cancers, metastases, and benign nodules [J]. BMC Cancer , 2026 , 26 ( 1 ): 177 . doi: 10.1186/s12885-025-15501-1 http://dx.doi.org/10.1186/s12885-025-15501-1
HOWARD A , SANDLER M , CHEN B , et al . Searching for MobileNetV3 [C]// Proceedings of 2019 IEEE/CVF International Conference on Computer Vision . Seoul : IEEE , 2019 : 1314 - 1324 . doi: 10.1109/iccv.2019.00140 http://dx.doi.org/10.1109/iccv.2019.00140
MANZARI O N , AHMADABADI H , KASHIANI H , et al . MedViT: a robust vision transformer for generalized medical image classification [J]. Computers in Biology and Medicine , 2023 , 157 : 106791 . doi: 10.1016/j.compbiomed.2023.106791 http://dx.doi.org/10.1016/j.compbiomed.2023.106791
HE K M , ZHANG X Y , REN S Q , et al . Deep residual learning for image recognition [C]// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition . Las Vegas : IEEE , 2016 : 770 - 778 . doi: 10.1109/cvpr.2016.90 http://dx.doi.org/10.1109/cvpr.2016.90
0
浏览量
48
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621