Structural Attention Graph Neural Network for Diagnosis and Prediction of COVID-19 Severity.

With rapid worldwide spread of Coronavirus Disease 2019 (COVID-19), jointly identifying severe COVID-19 cases from mild ones and predicting the conversion time (from mild to severe) is essential to optimize the workflow and reduce the clinician's workload. In this study, we propose a novel framework for COVID-19 diagnosis, termed as Structural Attention Graph Neural Network (SAGNN), which can combine the multi-source information including features extracted from chest CT, latent lung structural distribution, and non-imaging patient information to conduct diagnosis of COVID-19 severity and predict the conversion time from mild to severe. Specifically, we first construct a graph to incorporate structural information of the lung and adopt graph attention network to iteratively update representations of lung segments. To distinguish different infection degrees of left and right lungs, we further introduce a structural attention mechanism. Finally, we introduce demographic information and develop a multi-task learning framework to jointly perform both tasks of classification and regression. Experiments are conducted on a real dataset with 1687 chest CT scans, which includes 1328 mild cases and 359 severe cases. Experimental results show that our method achieves the best classification (e.g., 86.86% in terms of Area Under Curve) and regression (e.g., 0.58 in terms of Correlation Coefficient) performance, compared with other comparison methods.

The impact of facemask on patients with COPD: A systematic review and meta-analysis.

Background: Since the emergence of COVID-19, mandatory facemask wearing has been implemented around the world to prevent viral transmission, however, the impact of wearing facemasks on patients with COPD was unclear. Methods: The current study undertakes a systematic review and meta-analysis of a comprehensive literature retrieval from six databases, based on the pre-determined eligibility criteria, irrespective of language. The risk of bias was assessed using an established instrument. We primarily focused on analyzing ETCO2, SpO2, and heart and respiratory rates, and also considered the impacts on physiological and exercise performance. A descriptive summary of the data and possible meta-analysis was performed. Forest plots were generated to pool estimates based on each of the study outcomes. Results: Of the 3,751 publications considered, six publications were selected for a systematic review and two publications were included for meta-analysis, however, the quality of these six studies was relatively low overall. In the case of inactivity, the facemask wearing COPD cohort had higher respiratory rates than that of the non-facemask wearing cohort (MD = 1.00 and 95% CI 0.47-1.53, P < 0.05). There was no significant difference in ETCO2 (MD = 0.10 and 95% CI -1.57-1.78, P > 0.05) and heart rate (MD = 0.40 and 95% CI -3.59-4.39, P > 0.05) nor SpO2 (MD = -0.40 and 95% CI -0.84-0.04, P > 0.05) between the COPD patients with and without facemasks. Furthermore, it was observed that the only significant differences between the COPD patients with and without facemasks undertaking different activities were FEV1 (%) (MD = 3.84 and 95% CI 0.14-7.54, P < 0.05), FEV1/FVC (%) (MD = 3.25 and 95% CI 0.71-5.79, P < 0.05), and blood lactate (MD = -0.90 and 95% CI -1.73 to -0.07, P < 0.05). Conclusion: Wearing facemasks decreased the exercise performance of patients with COPD, however, it had minimal impact on physiological indexes. Further investigations will be performed on the high-quality data from randomized control studies. Systematic review registration: https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=326265, identifier: CRD42022326265.

The impact of facemask on patients with COPD: A systematic review and meta-analysis

Background Since the emergence of COVID-19, mandatory facemask wearing has been implemented around the world to prevent viral transmission, however, the impact of wearing facemasks on patients with COPD was unclear. Methods The current study undertakes a systematic review and meta-analysis of a comprehensive literature retrieval from six databases, based on the pre-determined eligibility criteria, irrespective of language. The risk of bias was assessed using an established instrument. We primarily focused on analyzing ETCO2, SpO2, and heart and respiratory rates, and also considered the impacts on physiological and exercise performance. A descriptive summary of the data and possible meta-analysis was performed. Forest plots were generated to pool estimates based on each of the study outcomes. Results Of the 3,751 publications considered, six publications were selected for a systematic review and two publications were included for meta-analysis, however, the quality of these six studies was relatively low overall. In the case of inactivity, the facemask wearing COPD cohort had higher respiratory rates than that of the non-facemask wearing cohort (MD = 1.00 and 95% CI 0.47–1.53, P < 0.05). There was no significant difference in ETCO2 (MD = 0.10 and 95% CI −1.57–1.78, P > 0.05) and heart rate (MD = 0.40 and 95% CI −3.59–4.39, P > 0.05) nor SpO2 (MD = −0.40 and 95% CI −0.84–0.04, P > 0.05) between the COPD patients with and without facemasks. Furthermore, it was observed that the only significant differences between the COPD patients with and without facemasks undertaking different activities were FEV1 (%) (MD = 3.84 and 95% CI 0.14–7.54, P < 0.05), FEV1/FVC (%) (MD = 3.25 and 95% CI 0.71–5.79, P < 0.05), and blood lactate (MD = −0.90 and 95% CI −1.73 to −0.07, P < 0.05). Conclusion Wearing facemasks decreased the exercise performance of patients with COPD, however, it had minimal impact on physiological indexes. Further investigations will be performed on the high-quality data from randomized control studies. Systematic review registration https://www.crd.york.ac.uk/PROSPERO/display_record.php?RecordID=326265, identifier: CRD42022326265.

The Impact of COVID-19 on SARSCoV-2-Negative Elderly Patients with Hip Fractures: A Single-Center Retrospective Study from Shanghai, China.

Purpose: Coronavirus disease 2019 (COVID-19) has brought an unprecedented change in wellbeing globally. The spread of the pandemic reportedly reduced the incidence of activity-related trauma, while that of fragility fractures remained stable. Here, we aimed to identify the risk factors associated with the prognosis of SARS-CoV-2 negative elderly patients with hip fractures. Patients and Methods: This retrospective study included elderly patients with hip fractures between 1st January and 9th May during the COVID-19 pandemic (Experiment group) and the same period from 2017 to 2019 (Control group). Perioperative mortality, complications, and functional recovery were compared between two groups of different time frame in the total cohort and patients who received surgical treatment. Multiple linear regression was carried out to identify the risk factors influencing the prognosis of COVID-negative elderly patients with hip fractures. Results: The proportion of patients with admission time less than 24 hours and the 6-month postoperative Parker score were significantly decreased during COVID-19 compared with the pre-COVID-19 period (p < 0.001). Multiple linear regression demonstrated that TTA (defined as time from injury to admission), rehabilitation after discharge and outpatient follow-up were associated with the 6-month Parker score in the total population (p < 0.001) and in patients who received surgical treatment (p < 0.001). Conclusion: Elderly patients with hip fractures had a poorer prognosis in epidemic period despite being COVID-19 negative. Factors including timely admission, postoperative follow-up, and rehabilitation could optimize safety and significantly improve the prognosis of elderly COVID-19 negative patients with hip fractures, even during a pandemic.

Neuropilin-1-Mediated SARS-CoV-2 Infection in Bone Marrow-Derived Macrophages Inhibits Osteoclast Differentiation.

In humans, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection can cause medical complications across various tissues and organs. Despite the advances to understanding the pathogenesis of SARS-CoV-2, its tissue tropism and interactions with host cells have not been fully understood. Existing clinical data have revealed disordered calcium and phosphorus metabolism in Coronavirus Disease 2019 (COVID-19) patients, suggesting possible infection or damage in the human skeleton system by SARS-CoV-2. Herein, SARS-CoV-2 infection in mouse models with wild-type and beta strain (B.1.351) viruses is investigated, and it is found that bone marrow-derived macrophages (BMMs) can be efficiently infected in vivo. Single-cell RNA sequencing (scRNA-Seq) analyses of infected BMMs identify distinct clusters of susceptible macrophages, including those related to osteoblast differentiation. Interestingly, SARS-CoV-2 entry on BMMs is dependent on the expression of neuropilin-1 (NRP1) rather than the widely recognized receptor angiotensin-converting enzyme 2 (ACE2). The loss of NRP1 expression during BMM-to-osteoclast differentiation or NRP1 neutralization and knockdown can significantly inhibit SARS-CoV-2 infection in BMMs. Importantly, it is found that authentic SARS-CoV-2 infection impedes BMM-to-osteoclast differentiation. Collectively, this study provides evidence for NRP1-mediated SARS-CoV-2 infection in BMMs and establishes a potential link between disturbed osteoclast differentiation and disordered skeleton metabolism in COVID-19 patients.

Neuropilin-1 Mediates SARS-CoV-2 Infection in Bone Marrow-derived Macrophages (preprint)

SARS-CoV-2 infection in human can cause medical complications across various tissues and organs. Despite of the advances to understanding the pathogenesis of SARS-CoV-2, its tissue tropism and interactions with host cells have not been fully understood. Existing clinical data have suggested possible SARS-CoV-2 infection in human skeleton system. In the present study, we found that authentic SARS-CoV-2 could efficiently infect human and mouse bone marrow-derived macrophages (BMMs) and alter the expression of macrophage chemotaxis and osteoclast-related genes. Importantly, in a mouse SARS-CoV-2 infection model that was enabled by the intranasal adenoviral (AdV) delivery of human angiotensin converting enzyme 2 (hACE2), SARS-CoV-2 was found to be present in femoral BMMs as determined by in situ immunofluorescence analysis. Using single-cell RNA sequencing (scRNA-Seq), we characterized SARS-CoV-2 infection in BMMs. Importantly, SARS-CoV-2 entry on BMMs appeared to be dependent on the expression of neuropilin-1 (NRP1) rather than the widely recognized receptor ACE2. It was also noted that unlike brain macrophages which displayed aging-dependent NRP1 expression, BMMs from neonatal and aged mice had constant NRP1 expression, making BMMs constantly vulnerable target cells for SARS-CoV-2. Furthermore, it was found that the abolished SARS-CoV-2 entry in BMM-derived osteoclasts was associated with the loss of NRP1 expression during BMM-to-osteoclast differentiation. Collectively, our study has suggested that NRP1 can mediate SARS-CoV-2 infection in BMMs, which precautions the potential impact of SARS-CoV-2 infection on human skeleton system.

Diagnosis of Coronavirus Disease 2019 (COVID-19) With Structured Latent Multi-View Representation Learning.

Recently, the outbreak of Coronavirus Disease 2019 (COVID-19) has spread rapidly across the world. Due to the large number of infected patients and heavy labor for doctors, computer-aided diagnosis with machine learning algorithm is urgently needed, and could largely reduce the efforts of clinicians and accelerate the diagnosis process. Chest computed tomography (CT) has been recognized as an informative tool for diagnosis of the disease. In this study, we propose to conduct the diagnosis of COVID-19 with a series of features extracted from CT images. To fully explore multiple features describing CT images from different views, a unified latent representation is learned which can completely encode information from different aspects of features and is endowed with promising class structure for separability. Specifically, the completeness is guaranteed with a group of backward neural networks (each for one type of features), while by using class labels the representation is enforced to be compact within COVID-19/community-acquired pneumonia (CAP) and also a large margin is guaranteed between different types of pneumonia. In this way, our model can well avoid overfitting compared to the case of directly projecting high-dimensional features into classes. Extensive experimental results show that the proposed method outperforms all comparison methods, and rather stable performances are observed when varying the number of training data.