Detection and Severity Classification of COVID-19 in CT Images Using Deep Learning.
Detection and Severity Classification of COVID-19 in CT Images Using Deep Learning.
复制标题
DOI：
10.3390/diagnostics11050893
复制
发表时间：
2021-05-17
期刊：
影响因子：
--
通讯作者：
Ayari MA
中科院分区：
文献类型：
--
作者：
Qiblawey Y;Tahir A;Chowdhury MEH;Khandakar A;Kiranyaz S;Rahman T;Ibtehaz N;Mahmud S;Maadeed SA;Musharavati F;Ayari MA
及早发现COVID-19对于降低患者死亡风险至关重要。在这项研究中，提出了一个级联系统来分割肺部，从计算机断层扫描图像中检测、定位和量化COVID-19感染。使用编码器-解码器卷积神经网络（ed - cnn）、UNet和特征金字塔网络（FPN）进行了一系列广泛的实验，使用DenseNet和ResNet的变体，使用不同的骨干（编码器）结构。采用DenseNet 161编码器进行的U-Net模型肺区域分割实验表明，该模型的Dice Similarity Coefficient （DSC）为97.19%,Intersection over Union （IoU）为95.10%。此外，使用DenseNet201编码器的FPN，该系统在COVID-19感染分割方面取得了良好的性能，DSC为94.13%，IoU为91.85%。所提出的系统可以可靠地定位各种形状和大小的感染，特别是小的感染区域，这在最近的研究中很少考虑。此外，该系统的检测灵敏度为99.64%，特异性为98.72%，具有较高的检测性能。最后，该系统能够在1110名受试者的数据集中区分不同严重程度的COVID-19感染，轻度、中度、重度和危重的敏感性分别为98.3%、71.2%、77.8%和100%。
Detecting COVID-19 at an early stage is essential to reduce the mortality risk of the patients. In this study, a cascaded system is proposed to segment the lung, detect, localize, and quantify COVID-19 infections from computed tomography images. An extensive set of experiments were performed using Encoder–Decoder Convolutional Neural Networks (ED-CNNs), UNet, and Feature Pyramid Network (FPN), with different backbone (encoder) structures using the variants of DenseNet and ResNet. The conducted experiments for lung region segmentation showed a Dice Similarity Coefficient (DSC) of 97.19% and Intersection over Union (IoU) of 95.10% using U-Net model with the DenseNet 161 encoder. Furthermore, the proposed system achieved an elegant performance for COVID-19 infection segmentation with a DSC of 94.13% and IoU of 91.85% using the FPN with DenseNet201 encoder. The proposed system can reliably localize infections of various shapes and sizes, especially small infection regions, which are rarely considered in recent studies. Moreover, the proposed system achieved high COVID-19 detection performance with 99.64% sensitivity and 98.72% specificity. Finally, the system was able to discriminate between different severity levels of COVID-19 infection over a dataset of 1110 subjects with sensitivity values of 98.3%, 71.2%, 77.8%, and 100% for mild, moderate, severe, and critical, respectively.
登录
查看更多内容
影响因子：
5.9
作者：
Pu J;Leader JK;Bandos A;Ke S;Wang J;Shi J;Du P;Guo Y;Wenzel SE;Fuhrman CR;Wilson DO;Sciurba FC;Jin C
通讯作者：
Jin C
DOI：
10.1148/ryai.2020200048
发表时间：
2020-07-01
期刊：
RADIOLOGY-ARTIFICIAL INTELLIGENCE
影响因子：
--
作者：
Chaganti, Shikha;Grenier, Philippe;Comaniciu, Dorin
通讯作者：
Comaniciu, Dorin
DOI：
10.1148/ryct.2020200075
发表时间：
2020-04-01
期刊：
RADIOLOGY-CARDIOTHORACIC IMAGING
影响因子：
--
作者：
Huang, Lu;Han, Rui;Xia, Liming
通讯作者：
Xia, Liming
影响因子：
64.8
作者：
Esteva A;Kuprel B;Novoa RA;Ko J;Swetter SM;Blau HM;Thrun S
通讯作者：
Thrun S
影响因子：
16.6
作者：
Jin C;Chen W;Cao Y;Xu Z;Tan Z;Zhang X;Deng L;Zheng C;Zhou J;Shi H;Feng J
通讯作者：
Feng J