Artificial intelligence (AI)‑based diagnosis has become an important auxiliary method for detecting lung infections. However, most existing approaches rely on deep learning, which are often plagued by issues such as insufficient model stability, high complexity, and low accuracy. This paper presents a shallow model which incorporates a multi‑scale attention mechanism to achieve both high accuracy and a simple structure for diagnosing COVID‑19 from CT scans. Firstly, two datasets of COVID‑19 CT images are combined into a single dataset to address the issue of model instability caused by insufficient data. Secondly, by introducing multi‑scale attention(MA) in the final three layers of the shallow ResNet18 network, the model’s feature extraction capability is enhanced. Finally, classifier with three fully connected layers (CTFCL) is constructed to improve the classification performance of the model, thereby increasing the accuracy of lung CT classification. Experimental results show that the proposed model achieves an accuracy of 95.41%, outperforming networks such as ResNet50, ResNet101, VGG16, and DenseNet169. Furthermore, the model has only 12.24×106 parameters, which is approximately 50% fewer than networks like ResNet50 and VGG16.
由表1可知,在与主流算法模型对比中RMCNet网络在测试集上实现了95.41%的准确率,96.25%的召回率以及95.45%的F1指标.同时随着模型的深度增加,其分类性能也越好. RMCNet则通过浅层的网络结构达到了最优的效果,展现了对大规模数据集依赖性低和模型训练简单的优势.在图3a的雷达图中,可以直观看出RMCNet模型在准确率、召回率和F1得分3个指标上达到最好,在AUC指标上仅次于DenseNet169和DenseNet201,并且该模型的多边形面积占比最大,表明其综合性能最好.同时表1中RMCNet模型参数数量仅为12.24×106,这比ResNet50和ResNet101模型参数数量少50%以上,比具有网络结构优势的DenseNet169和DenseNet201少9×106~24×106,是VGG16模型参数量的十分之一.此外,RMCNet模型的每秒浮点操作数(floating‑point operations per second,FLOPS)[22]仅为1.82×109,比其他模型低50%以上.从参数数量和FLOPS的对比结果来看,RMCNet模型具有更少的模型参数和更小的计算量,证明了该模型在保证性能的同时更具有实用性.
YeDuan‑sheng, HuaYun‑song.Research on pneumonia image segmentation based on deep learning[J].Software Engineering and Applications,2022(6):1490-1499.
[4]
KhanA I, ShahJ L, BhatM M.CoroNet:a deep neural network for detection and diagnosis of COVID-19 from chest X‑ray images[J].Computer Methods and Programs in Biomedicine,2020,196:105581.
[5]
CholletF.Xception:deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,2017:1251-1258.
[6]
WangS, ZhaY F, LiW M,et al.A fully automatic deep learning system for COVID-19 diagnostic and prognostic analysis[J].European Respiratory Journal,2020,56(2):2000775.
[7]
HuangG, LiuZ, Van Der MaatenL,et al.Densely connected convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,2017:4700-4708.
[8]
LinT Y, DollárP, GirshickR,et al.Feature pyramid networks for object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,2017:2117-2125.
[9]
WuZ F, ShenC H, Van Den HengelA.Wider or deeper:revisiting the ResNet model for visual recognition[J].Pattern Recognition,2019,90:119-133.
[10]
LiL, QinL X, XuZ G,et al.Using artificial intelligence to detect COVID-19 and community‑acquired pneumonia based on pulmonary CT:evaluation of the diagnostic accuracy[J].Radiology,2020,296(2):65-71.
[11]
LucasM, LermaM, FurstJ,et al.RSI-Grad-CAM:visual explanations from deep networks via Riemann‑Stieltjes integrated gradient‑based localization[C]// International Symposium on Visual Computing.Cham:Springer,2022:262-274.
[12]
TiwariS, JainA.Convolutional capsule network for COVID-19 detection using radiography images[J].International Journal of Imaging Systems and Technology,2021,31(2):525-539.
[13]
QuanH, XuX S, ZhengT T,et al.DenseCapsNet:detection of COVID-19 from X‑ray images using a capsule neural network[J].Computers in Biology and Medicine,2021,133:104399.
[14]
BanerjeeA, BhattacharyaR, BhatejaV,et al.COFE-Net:an ensemble strategy for computer‑aided detection for COVID-19[J].Measurement,2022,187:110289.
[15]
XiaX L, XuC, NanB.Inception-v3 for flower classification[C]//2017 2nd International Conference on Image,Vision and Computing (ICIVC).Chengdu,2017:783-787.
[16]
FerreiraC A, MeloT, SousaP,et al.Classification of breast cancer histology images through transfer learning using a pre‑trained inception ResNet V2[C]// International Conference on Image Analysis and Recognition.Cham:Springer,2018:763-770.
[17]
LiJ X, WangY Q, WangS,et al.Multiscale attention guided network for COVID-19 diagnosis using chest X-ray images[J].IEEE Journal of Biomedical and Health Informatics,2021,25(5):1336-1346.
[18]
FuJ L, ZhengH L, MeiT.Look closer to see better:recurrent attention convolutional neural network for fine‑grained image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,2017:4438-4446.
[19]
SrivastavaN, HintonG, KrizhevskyA,et al.Dropout:a simple way to prevent neural networks from overfitting[J].The Journal of Machine Learning Research,2014,15(1):1929-1958.
[20]
JaiswalA, GianchandaniN, SinghD,et al.Classification of the COVID-19 infected patients using DenseNet201 based deep transfer learning[J].Journal of Biomolecular Structure and Dynamics,2021,39(15):5682-5689.
[21]
PolsinelliM, CinqueL, PlacidiG.A light CNN for detecting COVID-19 from CT scans of the chest[J].Pattern Recognition Letters,2020,140:95-100.
[22]
LinT Y, GoyalP, GirshickR,et al.Focal loss for dense object detection[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,2017:2980-2988.
[23]
LiY S, ChenY P, DaiX Y,et al.Micronet:improving image recognition with extremely low flops[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Montreal,2021:468-477.
[24]
HuJ, ShenL, SunG.Squeeze‑and‑excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Salt Lake City,2018:7132-7141.
[25]
WooS, ParkJ, LeeJ Y,et al.CBAM:Convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision (ECCV).Munich:Springer,2018:3-19.