Diagnosis of COVID-19 by using structural attention graph neural network

Objective: The primary routine clinical diagnosis of COVID-19(corona virus disease 2019) is usually conducted based on epidemiological history, clinical manifestations and various laboratory detection methods, including nucleic acid amplification test (NAAT), computed tomography (CT) scan and serological techniques. However, manual detection is costly, time-consuming and leads to the potential increase of the infection risk of clinicians. As a good alternative, artificial intelligence techniques on available data from laboratory tests play an important role in the confirmation of COVID-19 cases. Some studies have been designed for distinguishing between novel coronavirus pneumonia, community-acquired pneumonia and normal people by graph neural network. However, these studies leverage the relationships between features to build a topological structure graph (e.g., connecting the nodes with high similarity), while ignoring the inner relationships between different parts of the lung, and thus limiting the performance of their models. To address this issue, we propose a graph neural network with hierarchical information inherent to the physical structure of lungs for the improved diagnosis of COVID-19. Besides, an attention mechanism is introduced to capture the discriminative features of different severities of infection in the left and right lobes of different patients. Method: Firstly, the topological structure is constructed based on the lungs' physical structure, and different lung segments are regarded as different nodes. Each node in the graph contains three kinds of handcrafted features, such as volume, density and mass feature, which reflect the infection in each lung segment and can be extracted from chest CT images using VB-Net. Secondly, based on graph neural network (GNN) and attention mechanism, we propose a novel structural attention graph neural network (SAGNN), which can perform the graph classification task. The SAGNN first aggregates the features in a given sample graph, and then uses the attention mechanism to effectively fuse the different features to obtain the final graph representation. This representation is then fed into a linear layer with softmax activation function that performs graph classification, so that the corresponding sample graph can be finally classified as a mild case or a severe one. To alleviate the effect of category imbalance on the classification results, we use the focal loss function. We optimize the proposed model via back propagation and learn the representations of graphs. Result: To verify the effectiveness of the proposed method, we compared SAGNN with several classical machine learning methods and graph classification methods on a real COVID-19 dataset, which includes 358 severe cases and 1 329 mild cases, provided by Shanghai Public Health Clinical Center. The result of comparative experiments was measured using three evaluation metrics including the sensitivity (SEN), the specificity (SPE) and the area under the receiver operating characteristic(ROC) curve (AUC). In the experiments, our model had a good performance, indicating the effectiveness of our model. Based on comparison with the classical machine learning methods and the graph neural network methods, SAGNN outperformed by 14.2%42.0% and 3.6%4.8% in terms of SEN, respectively. In terms of AUC, the performance of SAGNN increased by 8.9%18.7% and 3.1%3.6%, respectively. In addition, through the ablation experiments of SAGNN, we found that the SAGNN with attention mechanism outperformed by 2.4%, 1.4% and 1.1% in SPE, SEN and AUC than the SAGNN not with attention mechanism, respectively. The SAGNN with focal loss function outperformed by 2.1%, 1.1% and 0.9% in SPE, SEN and AUC than the SAGNN with cross-entropy loss function, respectively. Conclusion: In this work, we propose SAGNN, a new architecture for the diagnosis of severe and mild cases of COVID-19. Experimental results show the superior performance of SAGNN on classification task. Experimental results show that concatenating features of lung segments by t eir structure is effective. Moreover, we introduce an attention mechanism to distinguish the infection degree of right and left lungs. The focal loss is used to solve the issue of imbalanced group distribution, which further improves the overall network performance. We thus demonstrate the potential of SAGNN as clinical diagnosis support in this highly critical domain of medical intervention. We believe that our architecture provides a valuable case study for the early diagnosis of COVID-19, which is helpful for improvement in the field of computer-aided diagnosis and clinical practice. © 2022, Editorial Office of Journal of Image and Graphics. All right reserved.

CMC-COV19D: Contrastive Mixup Classification for COVID-19 Diagnosis

Deep learning methods have been extensively investigated for rapid and precise computer-aided diagnosis during the outbreak of the COVID-19 epidemic. However, there are still remaining issues to be addressed, such as distinguishing COVID-19 in the complex scenario of multi-type pneumonia classification. In this paper, we aim to boost the COVID-19 diagnostic performance with more discriminative deep representations of COVID and non-COVID categories. We propose a novel COVID-19 diagnosis approach with contrastive representation learning to effectively capture the intra-class similarity and inter-class difference. Besides, we design an adaptive joint training strategy to integrate the classification loss, mixup loss, and contrastive loss. Through the joint loss function, we obtain the high-level representations which are highly discriminative in COVID-19 screening. Extensive experiments on two chest CT image datasets, i.e., CC-CCII dataset and COV19-CT-DB database, demonstrate the effectiveness of our proposed approach in COVID-19 diagnosis. Our method won the first prize in the ICCV 2021 Covid-19 Diagnosis Competition of AI-enabled Medical Image Analysis Workshop. Our code is publicly available at https://github.com/houjunlin/Team-FDVTS-COVID-Solution.

Multi-DeepNet: A Novel Weakly-Supervised Multi-Task and Multi-View-Oriented Convolution Neural Network for COVID-19 Diagnosis from CT Images

Currently, manual analysis performed by professional radiologists is required for COVID-19 diagnosis given the patient's chest Computed Tomography (CT) images, but this process is inefficient and costly. Deep learning methods can provide computer vision-based solutions to help guide radiologists perform faster and more accurate diagnosis. However, current well performed methods require training on large and balanced datasets with pixel level lung lesion annotations, both of which are not easily accessible. Moreover, visual similarities between COVID-19 and other pneumonia in CT scans make it difficult to learn their distinguishing features. To address these issues, we propose a novel weakly-supervised deep learning model, named Multi-DeepNet, that can be well trained to perform fine-grained classification on small and imbalanced datasets. Specifically, a multi-task pre-training module is introduced to better extract distinguishing features between COVID-19 and other similar pneumonia. Furthermore, a multi-view-oriented classifier is proposed to extract complimentary information from the axial, coronal and sagittal planes. Experimental results demonstrate that our Multi-DeepNet achieves superior sensitivities, specificity, and accuracies compared to state-of-the-art methods. © 2021 IEEE.

Evaluation of pulmonary ventilation in COVID-19 patients using oxygen-enhanced three-dimensional ultrashort echo time MRI: a preliminary study.

AIM: To evaluate the lung function of coronavirus disease 2019 (COVID-19) patients using oxygen-enhanced (OE) ultrashort echo time (UTE) MRI. MATERIALS AND METHODS: Forty-nine patients with COVID-19 were included in the study. The OE-MRI was based on a respiratory-gated three-dimensional (3D) radial UTE sequence. For each patient, the percent signal enhancement (PSE) map was calculated using the expression PSE = (S100% - S21%)/S21%, where S21% and S100% are signals acquired during room air and 100% oxygen inhalation, respectively. Agreement of lesion detectability between UTE-MRI and computed tomography (CT) was performed using the kappa test. The Mann-Whitney U-test was used to evaluate the difference in the mean PSE between mild-type COVID-19 and common-type COVID-19. Spearman's test was used to assess the relationship between lesion mean PSE and lesion size. Furthermore, the Mann-Whitney U-test was used to evaluate the difference in region of interest (ROI) mean PSE between normal pulmonary parenchyma and lesions. The Kruskal-Wallis test was applied to test the difference in the mean PSE between different lesion types. RESULTS: CT and UTE-MRI reached good agreement in lesion detectability. Ventilation measures in mild-type patients (5.3 ± 5.5%) were significantly different from those in common-type patients (3 ± 3.9%). Besides, there was no significant correlation between lesion mean PSE and lesion size. The mean PSE of COVID-19 lesions (3.2 ± 4.9%) was significantly lower than that of the pulmonary parenchyma (5.4 ± 3.9%). No significant difference was found among different lesion types. CONCLUSION: OE-UTE-MRI could serve as a promising method for the assessment of lung function or treatment management of COVID-19 patients.

Video analysis method of basketball training assistant based on deep learning theory during COVID-19 spread

During covid-19, basketball training was stopped. Instead, the basketball video analysis is used. In this paper, literature, theoretical analysis, numerical simulation, experimental research and other research methods are used. The ant colony algorithm model of deep learning optimization for basketball technical and tactical decision-making is established to solve the optimization problem of actual technical and tactical decision-making. In this paper, video image correlation algorithm is used. In the video of players' free throw basket, there are many independent frames. The real frame set of free throw basket includes the whole process of jumping, arm lifting, squatting and stretching. The shooting frame set and shooting information of the ball are obtained. In this paper, a shot frame detection algorithm is proposed by analyzing multiple samples of multi shot video. The mathematical model of the shooting frame is established, which can locate the shooting frame quickly and accurately and determine the penalty frame set. Further obtain the basketball release status information for preparation. The reliability and robustness of the algorithm are verified by experiments on several samples. It provides a new method for basketball training during covid-19. © 2020 - IOS Press and the authors. All rights reserved.