Deep Learning-based Detection for COVID-19 from Chest CT using Weak Label

Chuansheng Zheng; Xianbo Deng; Qing Fu; Qiang Zhou; Jiapei Feng; Hui Ma; Wenyu Liu; Xinggang Wang

is the (which was not peer-reviewed) The copyright holder for this preprint . Table 1 : Detailed structure of the proposed DeCovNet. The number after the symbol "@", e.g., 5×7×7, denotes the kernel size of the convolution layer or the residual block. "&" means that there are two types of kernel size in the residual block. "T" denotes the length of the input CT volume. The number in "Output size" is in the order of "channel, length, height, width". The input size is 2 × T × 192 × 288. segmented the lung regions using an unsupervised learning method [20] , removed the failure cases manually, and the rest segmentation results were taken as ground-truth masks. The 3D lung mask of each CT volume was obtained by testing the trained 2D UNet frame-by-frame without using any temporal information. The overall training and testing procedures of UNet and DeCoVNet for COVID-19 detection were illustrated in Fig. 2 .