Multiple Instance Learning for Cheating Detection and Localization in Online Examinations (preprint)

The spread of the Coronavirus disease-2019 epidemic has caused many courses and exams to be conducted online. The cheating behavior detection model in examination invigilation systems plays a pivotal role in guaranteeing the equality of long-distance examinations. However, cheating behavior is rare, and most researchers do not comprehensively take into account features such as head posture, gaze angle, body posture, and background information in the task of cheating behavior detection. In this paper, we develop and present CHEESE, a CHEating detection framework via multiplE inStancE learning. The framework consists of a label generator that implements weak supervision and a feature encoder to learn discriminative features. In addition, the framework combines body posture and background features extracted by 3D convolution with eye gaze, head posture and facial features captured by OpenFace 2.0. These features are fed into the spatio-temporal graph module by stitching to analyze the spatio-temporal changes in video clips to detect the cheating behaviors. Our experiments on three datasets, UCF-Crime, ShanghaiTech and Online Exam Proctoring (OEP), prove the effectiveness of our method as compared to the state-of-the-art approaches, and obtain the frame-level AUC score of 87.58% on the OEP dataset.

Discovering common pathogenetic processes between COVID-19 and sepsis by bioinformatics and system biology approach.

Corona Virus Disease 2019 (COVID-19), an acute respiratory infectious disease caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), has spread rapidly worldwide, resulting in a pandemic with a high mortality rate. In clinical practice, we have noted that many critically ill or critically ill patients with COVID-19 present with typical sepsis-related clinical manifestations, including multiple organ dysfunction syndrome, coagulopathy, and septic shock. In addition, it has been demonstrated that severe COVID-19 has some pathological similarities with sepsis, such as cytokine storm, hypercoagulable state after blood balance is disrupted and neutrophil dysfunction. Considering the parallels between COVID-19 and non-SARS-CoV-2 induced sepsis (hereafter referred to as sepsis), the aim of this study was to analyze the underlying molecular mechanisms between these two diseases by bioinformatics and a systems biology approach, providing new insights into the pathogenesis of COVID-19 and the development of new treatments. Specifically, the gene expression profiles of COVID-19 and sepsis patients were obtained from the Gene Expression Omnibus (GEO) database and compared to extract common differentially expressed genes (DEGs). Subsequently, common DEGs were used to investigate the genetic links between COVID-19 and sepsis. Based on enrichment analysis of common DEGs, many pathways closely related to inflammatory response were observed, such as Cytokine-cytokine receptor interaction pathway and NF-kappa B signaling pathway. In addition, protein-protein interaction networks and gene regulatory networks of common DEGs were constructed, and the analysis results showed that ITGAM may be a potential key biomarker base on regulatory analysis. Furthermore, a disease diagnostic model and risk prediction nomogram for COVID-19 were constructed using machine learning methods. Finally, potential therapeutic agents, including progesterone and emetine, were screened through drug-protein interaction networks and molecular docking simulations. We hope to provide new strategies for future research and treatment related to COVID-19 by elucidating the pathogenesis and genetic mechanisms between COVID-19 and sepsis.

Extended SARS-CoV-2 RBD booster vaccination induces humoral and cellular immune tolerance in mice.

The repetitive applications of vaccine boosters have been brought up in face of continuous emergence of SARS-CoV-2 variants with neutralization escape mutations, but their protective efficacy and potential adverse effects remain largely unknown. Here, we compared the humoral and cellular immune responses of an extended course of recombinant receptor binding domain (RBD) vaccine boosters with those from conventional immunization strategy in a Balb/c mice model. Multiple vaccine boosters after the conventional vaccination course significantly decreased RBD-specific antibody titers and serum neutralizing efficacy against the Delta and Omicron variants, and profoundly impaired CD4+ and CD8+T cell activation and increased PD-1 and LAG-3 expressions in these T cells. Mechanistically, we confirmed that extended vaccination with RBD boosters overturned the protective immune memories by promoting adaptive immune tolerance. Our findings demonstrate potential risks with the continuous use of SARS-CoV-2 vaccine boosters, providing immediate implications for the global COVID-19 vaccination enhancement strategies.

Fully Automated Deep Learning-enabled Detection for Hepatic Steatosis on Computed Tomography: A Multicenter International Validation Study (preprint)

Despite high global prevalence of hepatic steatosis, no automated diagnostics demonstrated generalizability in detecting steatosis on multiple international datasets. Traditionally, hepatic steatosis detection relies on clinicians selecting the region of interest (ROI) on computed tomography (CT) to measure liver attenuation. ROI selection demands time and expertise, and therefore is not routinely performed in populations. To automate the process, we validated an existing artificial intelligence (AI) system for 3D liver segmentation and used it to purpose a novel method: AI-ROI, which could automatically select the ROI for attenuation measurements. AI segmentation and AI-ROI method were evaluated on 1,014 non-contrast enhanced chest CT images from eight international datasets: LIDC-IDRI, NSCLC-Lung1, RIDER, VESSEL12, RICORD-1A, RICORD-1B, COVID-19-Italy, and COVID-19-China. AI segmentation achieved a mean dice coefficient of 0.957. Attenuations measured by AI-ROI showed no significant differences (p = 0.545) and a reduction of 71% time compared to expert measurements. The area under the curve (AUC) of the steatosis classification of AI-ROI is 0.921 (95% CI: 0.883 - 0.959). If performed as a routine screening method, our AI protocol could potentially allow early non-invasive, non-pharmacological preventative interventions for hepatic steatosis. 1,014 expert-annotated liver segmentations of patients with hepatic steatosis annotations can be downloaded here: https://drive.google.com/drive/folders/1-g_zJeAaZXYXGqL1OeF6pUjr6KB0igJX.

Discovering common pathogenetic processes between COVID-19 and sepsis by bioinformatics and system biology approach

Corona Virus Disease 2019 (COVID-19), an acute respiratory infectious disease caused by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), has spread rapidly worldwide, resulting in a pandemic with a high mortality rate. In clinical practice, we have noted that many critically ill or critically ill patients with COVID-19 present with typical sepsis-related clinical manifestations, including multiple organ dysfunction syndrome, coagulopathy, and septic shock. In addition, it has been demonstrated that severe COVID-19 has some pathological similarities with sepsis, such as cytokine storm, hypercoagulable state after blood balance is disrupted and neutrophil dysfunction. Considering the parallels between COVID-19 and non-SARS-CoV-2 induced sepsis (hereafter referred to as sepsis), the aim of this study was to analyze the underlying molecular mechanisms between these two diseases by bioinformatics and a systems biology approach, providing new insights into the pathogenesis of COVID-19 and the development of new treatments. Specifically, the gene expression profiles of COVID-19 and sepsis patients were obtained from the Gene Expression Omnibus (GEO) database and compared to extract common differentially expressed genes (DEGs). Subsequently, common DEGs were used to investigate the genetic links between COVID-19 and sepsis. Based on enrichment analysis of common DEGs, many pathways closely related to inflammatory response were observed, such as Cytokine-cytokine receptor interaction pathway and NF-kappa B signaling pathway. In addition, protein-protein interaction networks and gene regulatory networks of common DEGs were constructed, and the analysis results showed that ITGAM may be a potential key biomarker base on regulatory analysis. Furthermore, a disease diagnostic model and risk prediction nomogram for COVID-19 were constructed using machine learning methods. Finally, potential therapeutic agents, including progesterone and emetine, were screened through drug-protein interaction networks and molecular docking simulations. We hope to provide new strategies for future research and treatment related to COVID-19 by elucidating the pathogenesis and genetic mechanisms between COVID-19 and sepsis.

ACE2-containing defensosomes serve as decoys to inhibit SARS-CoV-2 infection.

Extracellular vesicles of endosomal origin, exosomes, mediate intercellular communication by transporting substrates with a variety of functions related to tissue homeostasis and disease. Their diagnostic and therapeutic potential has been recognized for diseases such as cancer in which signaling defects are prominent. However, it is unclear to what extent exosomes and their cargo inform the progression of infectious diseases. We recently defined a subset of exosomes termed defensosomes that are mobilized during bacterial infection in a manner dependent on autophagy proteins. Through incorporating protein receptors on their surface, defensosomes mediated host defense by binding and inhibiting pore-forming toxins secreted by bacterial pathogens. Given this capacity to serve as decoys that interfere with surface protein interactions, we investigated the role of defensosomes during infection by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), the etiological agent of Coronavirus Disease 2019 (COVID-19). Consistent with a protective function, exosomes containing high levels of the viral receptor ACE2 in bronchoalveolar lavage fluid (BALF) from critically ill COVID-19 patients was associated with reduced intensive care unit (ICU) and hospitalization times. We found ACE2+ exosomes were induced by SARS-CoV-2 infection and activation of viral sensors in cell culture, which required the autophagy protein ATG16L1, defining these as defensosomes. We further demonstrate that ACE2+ defensosomes directly bind and block viral entry. These findings suggest that defensosomes may contribute to the antiviral response against SARS-CoV-2 and expand our knowledge on the regulation and effects of extracellular vesicles during infection.

[Clinical observation on filiform-fire needling of "Biaoben acupoint combination" for 33 cases of sequelae of coronavirus disease 2019 during recovery period].

OBJECTIVE: To observe the clinical therapeutic effect of filiform-fire needling of "Biaoben acupoint combination" on the sequelae of patients with coronavirus disease 2019 (COVID-19) during the recovery period. METHODS: A total of 33 patients with COVID-19 during the recovery period were treated with filiform-fire needling at the acupoints of Mingmen (GV 4), Shenzhu (GV 12), Gaohuang (BL 43), Zusanli (ST 36) and Shangjuxu (ST 37), etc., once every other day, 3 times a week, and 3 times was one course of treatment and totally 2 courses of treatment were required. The TCM symptom, Hamilton anxiety scale (HAMA) and Hamilton depression scale (HAMD) scores, pulmonary function indexes (forced vital capacity [FVC], forced expiratory volume in one second [FEV1], peak expiratory flow [PEF]) and chest CT imaging change were observed before and after treatment, and the therapeutic effect was evaluated. RESULTS: After treatment, the scores of TCM symptom, HAMA and HAMD were decreased compared with those before treatment (P<0.05), and the levels of FVC, FEV1 and PEF were increased compared with those before treatment (P<0.05), and the recovery rate of 22 patients with pulmonary ventilation dysfunction was 86.4% (19/22). After treatment, the lung shadow area was smaller than that before treatment (P<0.05). The effective rate of 25 patients with lung CT abnormalities was 84.0% (21/25). After treatment, 23 cases were cured, 5 cases were markedly effective, 4 cases were effective, 1 case was ineffective, the cured and markedly effective rate was 84.8%. CONCLUSION: The filiform-fire needling of "Biaoben acupoint combination" could significantly reduce the sequelae of cough, fatigue, chest tightness, etc. and mental symptoms such as anxiety and depression in patients with COVID-19 during the recovery period, and promote inflammatory exudation absorption of pulmonary lesion and improve lung ventilation function.

Artificial intelligence-based CT metrics used in predicting clinical outcome of COVID-19 in young and middle-aged adults.

BACKGROUND: Currently, most researchers mainly analyzed coronavirus disease 2019 (COVID-19) pneumonia visually or qualitatively, probably somewhat time-consuming and not precise enough. PURPOSE: This study aimed to excavate more information, such as differences in distribution, density, and severity of pneumonia lesions between males and females in a specific age group using artificial intelligence (AI)-based computed tomography (CT) metrics. Besides, these metrics were incorporated into a clinical regression model to predict the short-term outcome. MATERIALS AND METHODS: The clinical, laboratory information and a series of HRCT images from 49 patients, aged from 20 to 50 years and confirmed with COVID-19, were collected. The volumes and percentages of infection (POIs) among bilateral lungs and each bronchopulmonary segment were extracted using uAI-Discover-NCP software (version R001). The POI in three HU ranges (i.e., <-300, -300-49, and ≥50 HU representing ground-glass opacity [GGO], mixed opacity, and consolidation) were also extracted. Hospital stay was predicted with several POI after adjusting days from illness onset to admission, leucocytes, lymphocytes, C-reactive protein, age, and gender using a multiple linear regression model. A total of 91 patients aged 20-50 from public database were selected. RESULTS: Right lower lobes had the highest POI, followed by left lower lobes, right upper lobes, middle lobes, and left upper lobes. The distributions in lung lobes and segments were different between the sexes. Men had a higher total POI and GGO of the lungs, but less consolidation than women in initial CT (all p < 0.05). The total POI, percentage of consolidation on initial CT, and changed POI were positively correlated with hospital stay in the model. A total of 91 patients aged 20-50 years in the public database were selected, and AI segmentation was performed. The POI of the lower lobes was obviously higher than that in the upper lobes; the POI of each segment of the right upper lobe in the males was higher than that in the females, which was consistent with the result of the 49 patients previously. CONCLUSION: Both men and women had characteristic distributions in lung lobes and bronchopulmonary segments. AI-based CT quantitative metrics can provide more precise information regarding lesion distribution and severity to predict clinical outcome.

Data-Driven Decision Making in COVID-19 Response: A Survey (preprint)

COVID-19 has spread all over the world, having an enormous effect on our daily life and work. In response to the epidemic, a lot of important decisions need to be taken to save communities and economies worldwide. Data clearly plays a vital role in effective decision making. Data-driven decision making uses data related evidence and insights to guide the decision making process and to verify the plan of action before it is committed. To better handle the epidemic, governments and policy making institutes have investigated abundant data originating from COVID-19. These data include those related to medicine, knowledge, media, etc. Based on these data, many prevention and control policies are made. In this survey paper, we summarize the progress of data-driven decision making in the response to COVID-19, including COVID-19 prevention and control, psychological counselling, financial aid, work resumption, and school re-opening. We also propose some current challenges and open issues in data-driven decision making, including data collection and quality, complex data analysis, and fairness in decision making. This survey paper sheds light on current policy making driven by data, which also provides a feasible direction for further scientific research.

ACE2-containing defensosomes serve as decoys to inhibit SARS-CoV-2 infection (preprint)

Extracellular vesicles of endosomal origin, exosomes, mediate intercellular communication by transporting substrates with a variety of functions related to tissue homeostasis and disease. Their diagnostic and therapeutic potential has been recognized for diseases such as cancer in which signaling defects are prominent. However, it is unclear to what extent exosomes and their cargo inform the progression of infectious diseases. We recently defined a subset of exosomes termed defensosomes that are mobilized during bacterial infection in a manner dependent on autophagy proteins. Through incorporating protein receptors on their surface, defensosomes mediated host defense by binding and inhibiting pore-forming toxins secreted by bacterial pathogens. Given this capacity to serve as decoys that interfere with surface protein interactions, we investigated the role of defensosomes during infection by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the etiological agent of COVID-19. Consistent with a protective function, exosomes containing high levels of the viral receptor ACE2 in bronchioalveolar lavage fluid from critically ill COVID-19 patients was associated with reduced ICU and hospitalization times. We found ACE2+ exosomes were induced by SARS-CoV-2 infection and activation of viral sensors in cell culture, which required the autophagy protein ATG16L1, defining these as defensosomes. We further demonstrate that ACE2+ defensosomes directly bind and block viral entry. These findings suggest that defensosomes may contribute to the antiviral response against SARS-CoV-2 and expand our knowledge on the regulation and effects of extracellular vesicles during infection.

Heterogeneous network-based drug repurposing for COVID-19 (preprint)

The Corona Virus Disease 2019 (COVID-19) belongs to human coronaviruses (HCoVs), which spreads rapidly around the world. Compared with new drug development, drug repurposing may be the best shortcut for treating COVID-19. Therefore, we constructed a comprehensive heterogeneous network based on the HCoVs-related target proteins and use the previously proposed deepDTnet, to discover potential drug candidates for COVID-19. We obtain high performance in predicting the possible drugs effective for COVID-19 related proteins. In summary, this work utilizes a powerful heterogeneous network-based deep learning method, which may be beneficial to quickly identify candidate repurposable drugs toward future clinical trials for COVID-19. The code and data are available at https://github.com/stjin-XMU/HnDR-COVID.

Longitudinal Profiling of Antibody Response in Patients With COVID-19 in a Tertiary Care Hospital in Beijing, China.

The novel coronavirus named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) caused a global pandemic of the coronavirus disease 2019 (COVID-19), which elicits a wide variety of symptoms, ranging from mild to severe, with the potential to lead to death. Although used as the standard method to screen patients for SARS-CoV-2 infection, real-time PCR has challenges in dealing with asymptomatic patients and those with an undetectable viral load. Serological tests are therefore considered potent diagnostic tools to complement real-time PCR-based diagnosis and are used for surveillance of seroprevalence in populations. However, the dynamics of the antibody response against SARS-CoV-2 currently remain to be investigated. Here, through analysis of plasma samples from 84 patients with COVID-19, we observed that the response of virus-specific antibodies against three important antigens, RBD, N and S, dynamically changed over time and reached a peak 5-8 weeks after the onset of symptoms. The antibody responses were irrespective of sex. Severe cases were found to have higher levels of antibody response, larger numbers of inflammatory cells and C-reactive protein levels. Within the mild/moderate cases, pairwise comparison indicated moderate association between anti-RBD vs. anti-N, anti-RBD vs. anti-S1S2, and anti-N vs. anti-S1S2. Furthermore, the majority of cases could achieve IgM and IgG seroconversion at 2 weeks since the disease onset. Analysis of neutralizing antibodies indicated that these responses were able to last for more than 112 days but decline significantly after the peak. In summary, our findings demonstrate the longitudinally dynamic changes in antibody responses against SARS-CoV-2, which can contribute to the knowledge of humoral immune response after SARS-CoV-2 infection and are informative for future development of vaccine and antibody-based therapies.

Clinical characteristics of coronavirus disease 2019 (COVID-19) in patients out of Wuhan from China: a case control study.

BACKGROUND: A large-scale global outbreak of coronavirus disease-19 (COVID-19) out of Wuhan, from China, occurred in January 2020. To examine the clinical characteristics of COVID-19 in infected patients out of Wuhan, from China. METHODS: Thirteen patients were confirmed to be infected with novel coronavirus-2019 (2019-nCoV) between January 27 and February 8, 2020, in Baoji city, Shannxi, northwestern China. Epidemiological and clinical information, and computed to morphology imaging data from all COVID-19 patients were collected; cases were divided into two groups according to the severity of infection (mild or severe). RESULTS: Nine (9/13) COVID-19 patients exhibited mild disease severity, and defined as second-generation human-to-human transmission cases. Most patients (11/13) had a history of travel to or from Wuhan. There were no differences in sex and age between the mild and severe cases (all P > 0.05). A moderate degree of fever (11/13), cough (13/13), and fatigue (8/13) were common symptoms; however, there was no statistical difference between mild and severe cases in this regard (all P > 0.05). Oxyhemoglobin saturation and oxygenation index decreased, and C-reactive protein (CRP) and serum amyloid A (SAA) levels were elevated in all patients with COVID-19 infection, with statistically significant differences between those with severe disease and mild infection (all P < 0.05). Twelve of 13 COVID-19 patients exhibited changes in chest CT imaging features, and time course changes were different between mild and severe cases (all P < 0.05). CONCLUSION: Most cases of COVID-19 infection were second-generation human-to-human transmissions from Wuhan and were mild in severity. The clinical characteristics of COVID-19 varied. Oxyhemoglobin saturation, oxygenation index, CRP and SAA levels, and CT features were reliable parameters to evaluate the severity of COVID-19 infection. However, a few patients with mild COVID-19 disease lacked typical characteristics such as fever and changes in CT imaging features.