A novel attention-based enhancement framework for face mask detection in complicated scenarios☆

In the context of COVID-19 pandemic prevention and control, it is of vital significance to realize accurate face mask detection via computer vision technique. In this paper, a novel attention improved Yolo (AI-Yolo) model is proposed, which can handle existing challenges in the complicated real-world scenarios with dense distribution, small-size object detection and interference of similar occlusions In particular, a selective kernel (SK) module is set to achieve convolution domain soft attention mechanism with split, fusion and selection operations;a spatial pyramid pooling (SPP) module is applied to enhance the expression of local and global features, which enriches the receptive field information;and a feature fusion (FF) module is utilized to promote sufficient fusions of multi-scale features from each resolution branch, which adopts basic convolution operators without excessive computational complexity. In addition, the complete intersection over union (CIoU) loss function is adopted in the training stage for accurate positioning. Experiments are carried out on two challenging public face mask detection datasets, and the results demonstrate the superiority of the proposed AI-Yolo against other seven state-of-the-art object detection algorithms, which achieves the best results in terms of mean average precision and F1 score on both datasets. Furthermore, effectiveness of the meticulously designed modules in AI-Yolo is validated through extensive ablation studies. In a word, the proposed AI-Yolo is competent to accomplish face mask detection tasks under extremely complex situations with precise localization and accurate classification.

Progressive attention integration-based multi-scale efficient network for medical imaging analysis with application to COVID-19 diagnosis.

In this paper, a novel deep learning-based medical imaging analysis framework is developed, which aims to deal with the insufficient feature learning caused by the imperfect property of imaging data. Named as multi-scale efficient network (MEN), the proposed method integrates different attention mechanisms to realize sufficient extraction of both detailed features and semantic information in a progressive learning manner. In particular, a fused-attention block is designed to extract fine-grained details from the input, where the squeeze-excitation (SE) attention mechanism is applied to make the model focus on potential lesion areas. A multi-scale low information loss (MSLIL)-attention block is proposed to compensate for potential global information loss and enhance the semantic correlations among features, where the efficient channel attention (ECA) mechanism is adopted. The proposed MEN is comprehensively evaluated on two COVID-19 diagnostic tasks, and the results show that as compared with some other advanced deep learning models, the proposed method is competitive in accurate COVID-19 recognition, which yields the best accuracy of 98.68% and 98.85%, respectively, and exhibits satisfactory generalization ability as well.

Insufficient epitope-specific T cell clones are responsible for impaired cellular immunity to inactivated SARS-CoV-2 vaccine in older adults.

Aging is a critical risk factor for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine efficacy. The immune responses to inactivated vaccine for older adults, and the underlying mechanisms of potential differences to young adults, are still unclear. Here we show that neutralizing antibody production by older adults took a longer time to reach similar levels in young adults after inactivated SARS-CoV-2 vaccination. We screened SARS-CoV-2 variant strains for epitopes that stimulate specific CD8 T cell response, and older adults exhibited weaker CD8 T-cell-mediated responses to these epitopes. Comparison of lymphocyte transcriptomes from pre-vaccinated and post-vaccinated donors suggested that the older adults had impaired antigen processing and presentation capability. Single-cell sequencing revealed that older adults had less T cell clone expansion specific to SARS-CoV-2, likely due to inadequate immune receptor repertoire size and diversity. Our study provides mechanistic insights for weaker response to inactivated vaccine by older adults and suggests the need for further vaccination optimization for the old population.

Engineering antiviral immune-like systems for autonomous virus detection and inhibition in mice.

The ongoing COVID-19 pandemic has demonstrated that viral diseases represent an enormous public health and economic threat to mankind and that individuals with compromised immune systems are at greater risk of complications and death from viral diseases. The development of broad-spectrum antivirals is an important part of pandemic preparedness. Here, we have engineer a series of designer cells which we term autonomous, intelligent, virus-inducible immune-like (ALICE) cells as sense-and-destroy antiviral system. After developing a destabilized STING-based sensor to detect viruses from seven different genera, we have used a synthetic signal transduction system to link viral detection to the expression of multiple antiviral effector molecules, including antiviral cytokines, a CRISPR-Cas9 module for viral degradation and the secretion of a neutralizing antibody. We perform a proof-of-concept study using multiple iterations of our ALICE system in vitro, followed by in vivo functionality testing in mice. We show that dual output ALICESaCas9+Ab system delivered by an AAV-vector inhibited viral infection in herpetic simplex keratitis (HSK) mouse model. Our work demonstrates that viral detection and antiviral countermeasures can be paired for intelligent sense-and-destroy applications as a flexible and innovative method against virus infection.

The Mpro structure-based modifications of ebselen derivatives for improved antiviral activity against SARS-CoV-2 virus.

The main protease (Mpro or 3CLpro) of SARS-CoV-2 virus is a cysteine enzyme critical for viral replication and transcription, thus indicating a potential target for antiviral therapy. A recent repurposing effort has identified ebselen, a multifunctional drug candidate as an inhibitor of Mpro. Our docking of ebselen to the binding pocket of Mpro crystal structure suggests a noncovalent interaction for improvement of potency, antiviral activity and selectivity. To test this hypothesis, we designed and synthesized ebselen derivatives aimed at enhancing their non-covalent bonds within Mpro. The inhibition of Mpro by ebselen derivatives (0.3 µM) was screened in both HPLC and FRET assays. Nine ebselen derivatives (EBs) exhibited stronger inhibitory effect on Mpro with IC50 of 0.07-0.38 µM. Further evaluation of three derivatives showed that EB2-7 exhibited the most potent inhibition of SARS-CoV-2 viral replication with an IC50 value of 4.08 µM in HPAepiC cells, as compared to the prototype ebselen at 24.61 µM. Mechanistically, EB2-7 functions as a noncovalent Mpro inhibitor in LC-MS/MS assay. Taken together, our identification of ebselen derivatives with improved antiviral activity may lead to developmental potential for treatment of COVID-19 and SARS-CoV-2 infection.

GIANA allows computationally-efficient TCR clustering and multi-disease repertoire classification by isometric transformation.

Similarity in T-cell receptor (TCR) sequences implies shared antigen specificity between receptors, and could be used to discover novel therapeutic targets. However, existing methods that cluster T-cell receptor sequences by similarity are computationally inefficient, making them impractical to use on the ever-expanding datasets of the immune repertoire. Here, we developed GIANA (Geometric Isometry-based TCR AligNment Algorithm) a computationally efficient tool for this task that provides the same level of clustering specificity as TCRdist at 600 times its speed, and without sacrificing accuracy. GIANA also allows the rapid query of large reference cohorts within minutes. Using GIANA to cluster large-scale TCR datasets provides candidate disease-specific receptors, and provides a new solution to repertoire classification. Querying unseen TCR-seq samples against an existing reference differentiates samples from patients across various cohorts associated with cancer, infectious and autoimmune disease. Our results demonstrate how GIANA could be used as the basis for a TCR-based non-invasive multi-disease diagnostic platform.

Streamlining SARS-CoV-2 confirmatory testing to reduce false positive results.

BACKGROUND: Confirmatory testing of SARS-CoV-2 results is essential to reduce false positives, but comes at a cost of significant extra workload for laboratories and increased turnaround time. A balance must be sought. We analysed our confirmatory testing pathway to produce a more refined approach in preparation for rising case numbers. METHODS: Over a 10-week low prevalence period we performed confirmatory testing on all newly positive results. Turnaround time was measured and results were analysed to identify a threshold that could be applied as a cut-off for future confirmatory testing and reduce overall workload for the laboratory. RESULTS: Between 22/06/20 and 31/08/20 confirmatory testing was performed on 108 newly positive samples, identifying 32 false positive results (30 %). Turnaround time doubled, increasing by an extra 17 h. There was a highly statistically significant difference between initial Relative Light Unit (RLU) of results that confirmed compared to those that did not, 1176 vs 721 (P < 0.00001). RLU = 1000 was identified as a suitable threshold for confirmatory testing in our laboratory: with RLU ≥ 1000, 55/56 (98 %) confirmed as positive, whereas with RLU < 1000 only 12/38 (32 %) confirmed. CONCLUSIONS: False positive SARS-CoV-2 tests can be identified by confirmatory testing, yet this may significantly delay results. Establishing a threshold for confirmatory testing streamlines this process to focus only on samples where it is most required. We advise all laboratories to follow a similar process to identify thresholds that trigger confirmatory testing for their own assays, increasing accuracy while maintaining efficiency for when case numbers are high.

Point of Care Nucleic Acid Testing for SARS-CoV-2 in Hospitalized Patients: A Clinical Validation Trial and Implementation Study.

There is an urgent need for rapid SARS-CoV-2 testing in hospitals to limit nosocomial spread. We report an evaluation of point of care (POC) nucleic acid amplification testing (NAAT) in 149 participants with parallel combined nasal and throat swabbing for POC versus standard lab RT-PCR testing. Median time to result is 2.6 (IQR 2.3-4.8) versus 26.4 h (IQR 21.4-31.4, p < 0.001), with 32 (21.5%) positive and 117 (78.5%) negative. Cohen's κ correlation between tests is 0.96 (95% CI 0.91-1.00). When comparing nearly 1,000 tests pre- and post-implementation, the median time to definitive bed placement from admission is 23.4 (8.6-41.9) versus 17.1 h (9.0-28.8), p = 0.02. Mean length of stay on COVID-19 "holding" wards is 58.5 versus 29.9 h (p < 0.001). POC testing increases isolation room availability, avoids bed closures, allows discharge to care homes, and expedites access to hospital procedures. POC testing could mitigate the impact of COVID-19 on hospital systems.

Performance Evaluation of the SAMBA II SARS-CoV-2 Test for Point-of-Care Detection of SARS-CoV-2.

Nucleic acid amplification for the detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA in respiratory samples is the standard method for diagnosis. The majority of this testing is centralized and therefore has turnaround times of several days. Point-of-care (POC) testing with rapid turnaround times would allow more effective triage in settings where patient management and infection control decisions need to be made rapidly. The inclusivity and specificity of the Simple AMplification-Based Assay (SAMBA) II SARS-CoV-2 test were determined by both in silico analyses of the primers and probes and wet testing. The SAMBA II SARS-CoV-2 test was evaluated for performance characteristics. Clinical performance was evaluated in residual combined throat/nose swabs and compared to that of the Public Health England real-time PCR assay targeting the RdRp gene. The SAMBA II SARS-CoV-2 test has an analytical sensitivity of 250 copies/ml for detecting two regions of the genome (open reading frame 1ab [ORF1ab] and nucleocapsid protein [N]). The clinical performance was evaluated in 172 residual combined nose/throat swabs provided by the Clinical Microbiology and Public Health Laboratory, Addenbrooke's Hospital, Cambridge (CMPHL), which showed an estimated positive percent agreement of 98.9% (95% confidence interval [CI], 93.83 to 99.97) and negative percent agreement of 96.4% (95% CI, 89.92 to 99.26) compared to testing by the CMPHL. The data show that the SAMBA II SARS-CoV-2 test performs equivalently to the centralized testing methods, but with a shorter turnaround time of 86 to 101 min. Point-of-care tests such as SAMBA should enable rapid patient management and effective implementation of infection control measures.

Posttraumatic stress disorder symptoms in healthcare workers after the peak of the COVID-19 outbreak: A survey of a large tertiary care hospital in Wuhan.

This study examined the prevalence of posttraumatic stress disorder (PTSD) symptoms and assessed mental illness via an online survey among healthcare workers (HCWs) at the Central Hospital of Wuhan after the peak of the COVID-19 outbreak. PTSD symptoms were measured using the PTSD Checklist Civilian Version (PCL-C), with a cutoff score of 50. Among the 642 HCWs, the prevalence of probable PTSD was 20.87%. Additionally, 88.88%, 82.09%, 100%, and 95.52% of HCWs with probable PTSD reported varying degrees of anxiety, depression, somatic symptoms, and insomnia, respectively. HCWs with probable PTSD scored higher on the Hospital Anxiety and Depression Scale (HADS), Patient Health questionnaire-15 (PHQ-15), and Insomnia Severity Index (ISI) than non-PTSD HCWs (all p < 0.05). Multivariate regression analysis revealed that HCWs with negative COVID-19 tests (OR, 0.35; 95% CI, 0.21-0.58; p < 0.00), those with high Social Support Self-Rating Scale (SSRS) scores (OR, 0.30; 95% CI, 0.17-0.52; p < 0.00), and HCWs whose family members tested negative (OR, 0.64; 95% CI, 0.42-0.96; p = 0.03) were less likely to have probable PTSD. This study found a high prevalence of probable PTSD and severe mental illness among local HCWs. Our finding emphasizes the need to provide mental health support for HCWs.

Specificity and positive predictive value of SARS-CoV-2 nucleic acid amplification testing in a low-prevalence setting.

OBJECTIVES: When the prevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is low, many positive test results are false positives. Confirmatory testing reduces overdiagnosis and nosocomial infection and enables real-world estimates of test specificity and positive predictive value. This study estimates these parameters to evaluate the impact of confirmatory testing and to improve clinical diagnosis, epidemiological estimation and interpretation of vaccine trials. METHODS: Over 1 month we took all respiratory samples from our laboratory with a patient's first detection of SARS-CoV-2 RNA (Hologic Aptima SARS-CoV-2 assay or in-house RT-PCR platform), and repeated testing using two platforms. Samples were categorized by source, and by whether clinical details suggested COVID-19 or corroborative testing from another laboratory. We estimated specificity and positive predictive value using approaches based on maximum likelihood. RESULTS: Of 19 597 samples, SARS-CoV-2 RNA was detected in 107; 52 corresponded to first-time detection (0.27% of tests on samples without previous detection). Further testing detected SARS-CoV-2 RNA once or more ('confirmed') in 29 samples (56%), and failed to detect SARS-CoV-2 RNA ('not confirmed') in 23 (44%). Depending upon assumed parameters, point estimates for specificity and positive predictive value were 99.91-99.98% and 61.8-89.8% respectively using the Hologic Aptima SARS-CoV-2 assay, and 97.4-99.1% and 20.1-73.8% respectively using an in-house assay. CONCLUSIONS: Nucleic acid amplification testing for SARS-CoV-2 is highly specific. Nevertheless, when prevalence is low a significant proportion of initially positive results fail to confirm, and confirmatory testing substantially reduces the detection of false positives. Omitting additional testing in samples with higher prior detection probabilities focuses testing where it is clinically impactful and minimizes delay.

Combined Point-of-Care Nucleic Acid and Antibody Testing for SARS-CoV-2 following Emergence of D614G Spike Variant.

Rapid COVID-19 diagnosis in the hospital is essential, although this is complicated by 30%-50% of nose/throat swabs being negative by SARS-CoV-2 nucleic acid amplification testing (NAAT). Furthermore, the D614G spike mutant dominates the pandemic and it is unclear how serological tests designed to detect anti-spike antibodies perform against this variant. We assess the diagnostic accuracy of combined rapid antibody point of care (POC) and nucleic acid assays for suspected COVID-19 disease due to either wild-type or the D614G spike mutant SARS-CoV-2. The overall detection rate for COVID-19 is 79.2% (95% CI 57.8-92.9) by rapid NAAT alone. The combined point of care antibody test and rapid NAAT is not affected by D614G and results in very high sensitivity for COVID-19 diagnosis with very high specificity.