Prior-Attention Residual Learning for More Discriminative COVID-19 Screening in CT Images

Abstract

We propose a conceptually simple framework for fast COVID-19 screening in 3D chest CT images. The framework can efficiently predict whether or not a CT scan contains pneumonia while simultaneously identifying pneumonia types between COVID-19 and Interstitial Lung Disease (ILD) caused by other viruses. In the proposed method, two 3D-ResNets are coupled together into a single model for the two above-mentioned tasks via a novel prior-attention strategy. We extend residual learning with the proposed prior-attention mechanism and design a new so-called priorattention residual learning (PARL) block. The model can be easily built by stacking the PARL blocks and trained endto-end using multi-task losses. More specifically, one 3D-ResNet branch is trained as a binary classifier using lung images with and without pneumonia so that it can highlight the lesion areas within the lungs. Simultaneously, inside the PARL blocks, prior-attention maps are generated from this branch and used to guide another branch to learn more discriminative representations for the pneumoniatype classification. Experimental results demonstrate that the proposed framework can significantly improve the performance of COVID-19 screening. Compared to other meth-Manuscript

Automated Assessment of COVID-19 Reporting and Data System and Chest CT Severity Scores in Patients Suspected of Having COVID-19 Using Artificial Intelligence

“…Artificial Intelligence (AI) using deep learning has been advocated for automated reading of COVID-19 CT scans, including diagnosing COVID-19 (7)(8)(9)(10)(11)(12)(13)(14) and quantifying parenchymal involvement (15)(16)(17)(18). While these studies illustrate the potential of AI algorithms, their practical value is debatable (19).…”

Auto-diagnosis of COVID-19 using Lung CT Images with Semi-supervised Shallow Learning Network

“…It is evident from the experimental data provided in Table 1 and from the statistical significance test (KS test) [19] conducted on the results that in spite of being a self-supervised network, the proposed PQIS-Net attains similar performance in segmentation tasks on the data set [30] in comparison with the pre-trained CNN models (ResNet50 [9] and 3D-UNet [10]) under the four evaluation parameters (ACC, DS, P P V, SS). Table 2 presents the numerical results obtained using the proposed semi-supervised shallow neural network model, ResNet50 [9], 3D-UNet [10], Kang et al [15] and Wang et al [18] for COVID-19 detection on the Brazilian data Figure 6: PQIS-Net segmented lung CT slice#171 [30] with the three different masks. set [29].…”

“…In addition to this segmentation, experiments are also set up for classifications the proposed Semi-supervised model, ResNet50 [9], 3D-UNet [10]. The other state of the art techniques include Kang et al [15] and Wang et al [18] for COVID-19 detection on the Brazilian data set [29]. The evaluation process involves the manually segmented lesion mask as ground truth and each 2D pixel is predicted as either True Positive (T RP ) or True Negative (T RN ) or False Positive (T RN ) or False Negative (F LN ).…”

Applications of artificial intelligence in battling against covid-19: A literature review

“…In order to identify covid-19 cases a new framework for processing CT images is proposed in [316] . In this method, two 3D-ResNets are combined to build a prior-attention residual learning.…”