Event Detection from Social Media for Epidemic Prediction (preprint)

Social media is an easy-to-access platform providing timely updates about societal trends and events. Discussions regarding epidemic-related events such as infections, symptoms, and social interactions can be crucial for informing policymaking during epidemic outbreaks. In our work, we pioneer exploiting Event Detection (ED) for better preparedness and early warnings of any upcoming epidemic by developing a framework to extract and analyze epidemic-related events from social media posts. To this end, we curate an epidemic event ontology comprising seven disease-agnostic event types and construct a Twitter dataset SPEED with human-annotated events focused on the COVID-19 pandemic. Experimentation reveals how ED models trained on COVID-based SPEED can effectively detect epidemic events for three unseen epidemics of Monkeypox, Zika, and Dengue; while models trained on existing ED datasets fail miserably. Furthermore, we show that reporting sharp increases in the extracted events by our framework can provide warnings 4-9 weeks earlier than the WHO epidemic declaration for Monkeypox. This utility of our framework lays the foundations for better preparedness against emerging epidemics.

Efficacy of SARS-CoV-2 vaccines and the dose-response relationship with three major antibodies: a systematic review and meta-analysis of randomised controlled trials.

BACKGROUND: The efficacy of SARS-CoV-2 vaccines in preventing severe COVID-19 illness and death is uncertain due to the rarity of data in individual trials. How well the antibody concentrations can predict the efficacy is also uncertain. We aimed to assess the efficacy of these vaccines in preventing SARS-CoV-2 infections of different severities and the dose-response relationship between the antibody concentrations and efficacy. METHODS: We did a systematic review and meta-analysis of randomised controlled trials (RCTs). We searched PubMed, Embase, Scopus, Web of Science, Cochrane Library, WHO, bioRxiv, and medRxiv for papers published between Jan 1, 2020 and Sep 12, 2022. RCTs on the efficacy of SARS-CoV-2 vaccines were eligible. Risk of bias was assessed using the Cochrane tool. A frequentist, random-effects model was used to combine efficacy for common outcomes (ie, symptomatic and asymptomatic infections) and a Bayesian random-effects model was used for rare outcomes (ie, hospital admission, severe infection, and death). Potential sources of heterogeneity were investigated. The dose-response relationships of neutralising, spike-specific IgG and receptor binding domain-specific IgG antibody titres with efficacy in preventing SARS-CoV-2 symptomatic and severe infections were examined by meta-regression. This systematic review is registered with PROSPERO, CRD42021287238. FINDINGS: 28 RCTs (n=286 915 in vaccination groups and n=233 236 in placebo groups; median follow-up 1-6 months after last vaccination) across 32 publications were included in this review. The combined efficacy of full vaccination was 44·5% (95% CI 27·8-57·4) for preventing asymptomatic infections, 76·5% (69·8-81·7) for preventing symptomatic infections, 95·4% (95% credible interval 88·0-98·7) for preventing hospitalisation, 90·8% (85·5-95·1) for preventing severe infection, and 85·8% (68·7-94·6) for preventing death. There was heterogeneity in the efficacy of SARS-CoV-2 vaccines against asymptomatic and symptomatic infections but insufficient evidence to suggest whether the efficacy could differ according to the type of vaccine, age of the vaccinated individual, and between-dose interval (p>0·05 for all). Vaccine efficacy against symptomatic infection waned over time after full vaccination, with an average decrease of 13·6% (95% CI 5·5-22·3; p=0·0007) per month but can be enhanced by a booster. We found a significant non-linear relationship between each type of antibody and efficacy against symptomatic and severe infections (p<0·0001 for all), but there remained considerable heterogeneity in the efficacy, which cannot be explained by antibody concentrations. The risk of bias was low in most studies. INTERPRETATION: The efficacy of SARS-CoV-2 vaccines is higher for preventing severe infection and death than for preventing milder infection. Vaccine efficacy wanes over time but can be enhanced by a booster. Higher antibody titres are associated with higher estimates of efficacy but precise predictions are difficult due to large unexplained heterogeneity. These findings provide an important knowledge base for interpretation and application of future studies on these issues. FUNDING: Shenzhen Science and Technology Programs.

Hospital Strain and COVID-19 Fatality - England, April 2020-March 2022.

What is already known about this topic?: During the coronavirus disease 2019 (COVID-19) pandemic, tremendous efforts have been made in countries to suppress epidemic peaks and strengthen hospital services to avoid hospital strain and ultimately reduce the risk of death from COVID-19. However, there is limited empirical evidence that hospital strain increases COVID-19 deaths. What is added by this report?: We found the risk of death from COVID-19 was linearly associated with the number of patients currently in hospitals, a measure of hospital strain, before the Omicron period. This risk could be increased by a maximum of 188.0%. What are the implications for public health practice?: These findings suggest that any (additional) effort to reduce hospital strain would be beneficial during early large COVID-19 outbreaks and possibly also others alike. During an Omicron outbreak, vigilance remains necessary to prevent excess deaths caused by hospital strain as happened in Hong Kong Special Administrative Region, China.

Characteristics and Containment of 74 Imported COVID-19 Outbreaks: Experiences, Lessons, and Implications - China, 2020-2021.

What is already known about this topic?: After the initial coronavirus disease 2019 (COVID-19) outbreak in Wuhan, China, the outbreaks during the dynamic-zero policy period in the mainland of China have not been systematically documented. What is added by this report?: We summarized the characteristics of 74 imported COVID-19 outbreaks between March 19, 2020 and December 31, 2021. All outbreaks of early severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants were successfully contained with the aid of nucleic acid testing, modern communication technologies, and non-pharmacological interventions. What are the implications for public health practice?: These findings provide us with confidence for the containment of future emerging infectious diseases alike at early stages to prevent pandemics or to win time to gain experience, develop vaccines and drugs, vaccinate people, and wait for the possible lessening of the virus' pathogenicity.

Constructing a Decision Model for Health Club Members to Purchase Coaching Programs during the COVID-19 Epidemic

The recent COVID-19 epidemic has affected the global sports industry to a certain extent, and health clubs are no exception. To avoid unsustainable operations, health clubs need to restructure their programs to suit members' needs. Therefore, this study constructs a two-stage framework model to evaluate health club members' purchase of coaching programs. The first stage is to construct a hierarchy of evaluation, using the modified Delphi method, to select suitable criteria and extended sub-criteria, and add and delete them through expert discussion. In the second stage, we use the pairwise comparison matrix to calculate the weight of each criterion and sub-criterion to influence each other. Next, we evaluate and compare physical, online and offline, and live-stream coaching programs, by using network hierarchy analysis to identify the best class purchase plan during the epidemic and provide relevant suggestions. The results of the study found that during the epidemic, the primary sales were for weight training among physical programs (0.314), and activity classes among online and offline programs (0.633) as well as live-stream coaching programs (0.280). These findings have implications for health clubs in deciding which mode they need to adopt for sustainable operations.

Awake Prone Position in COVID-19-Related Acute Respiratory Failure: A Meta-analysis of Randomized Controlled Trials (preprint)

Background We aimed to investigate the effects of awake prone positioning (APP) in acute hypoxemic, non-intubated patients due to COVID-19 infections.Methods A collection of databases including PubMed, EMBASE, Web of Science and Cochrane Central Register were searched up to June 1, 2022. All randomized trials investigating the effects of the APP were included in the present meta-analysis. The primary outcome was intubation rate, and the secondary outcome included the length of intensive care unit (ICU) and hospital stay, and mortality. Prescribed sub-group analysis was also conducted.Results A total of 10 randomized trials enrolling 2324 patients were finally included in the present study. The results indicated APP was associated with a significant reduction of intubation rate (OR 0.77, 95% CI 0.63 to 0.93, P = 0.007). However, no differences could be observed in length of ICU stay or hospitalization, or mortality. Sub-group analysis suggested that patients in the ICU settings (OR 0.74, 95% CI 0.60 to 0.91, P = 0.004), patients whose median APP time was more than 4 hours (OR 0.77, 95% CI 0.63 to 0.93, P = 0.008), and patients with an average baseline SpO2 to FiO2 ratio less than 200 mmHg (OR 0.75, 95% CI 0.61 to 0.92) were more likely to benefit from APP with significant reduced intubation rate.Conclusion Based on the current evidence, non-intubated patients with hypoxemic respiratory failure due to COVID-19 infection conducted with APP was associated with significantly induced intubation rate. However, no differences in ICU or hospital length of stay, or mortality could be observed between APP and usual care.

Neural network training with highly incomplete medical datasets

Neural network training and validation rely on the availability of large high-quality datasets. However, in many cases only incomplete datasets are available, particularly in health care applications, where each patient typically undergoes different clinical procedures or can drop out of a study. Since the data to train the neural networks need to be complete, most studies discard the incomplete datapoints, which reduces the size of the training data, or impute the missing features, which can lead to artifacts. Alas, both approaches are inadequate when a large portion of the data is missing. Here, we introduce GapNet, an alternative deep-learning training approach that can use highly incomplete datasets without overfitting or introducing artefacts. First, the dataset is split into subsets of samples containing all values for a certain cluster of features. Then, these subsets are used to train individual neural networks. Finally, this ensemble of neural networks is combined into a single neural network whose training is fine-tuned using all complete datapoints. Using two highly incomplete real-world medical datasets, we show that GapNet improves the identification of patients with underlying Alzheimer’s disease pathology and of patients at risk of hospitalization due to Covid-19. Compared to commonly used imputation methods, this improvement suggests that GapNet can become a general tool to handle incomplete medical datasets.

DisinfoMeme: A Multimodal Dataset for Detecting Meme Intentionally Spreading Out Disinformation (preprint)

Disinformation has become a serious problem on social media. In particular, given their short format, visual attraction, and humorous nature, memes have a significant advantage in dissemination among online communities, making them an effective vehicle for the spread of disinformation. We present DisinfoMeme to help detect disinformation memes. The dataset contains memes mined from Reddit covering three current topics: the COVID-19 pandemic, the Black Lives Matter movement, and veganism/vegetarianism. The dataset poses multiple unique challenges: limited data and label imbalance, reliance on external knowledge, multimodal reasoning, layout dependency, and noise from OCR. We test multiple widely-used unimodal and multimodal models on this dataset. The experiments show that the room for improvement is still huge for current models.

Omicron-specific mRNA vaccine induced potent neutralizing antibody against Omicron but not other SARS-CoV-2 variants (preprint)

The emerging SARS-CoV-2 variants of concern (VOC) harbor mutations associated with increasing transmission and immune escape, hence undermine the effectiveness of current COVID-19 vaccines. In late November of 2021, the Omicron (B.1.1.529) variant was identified in South Africa and rapidly spread across the globe. It was shown to exhibit significant resistance to neutralization by serum not only from convalescent patients, but also from individuals recieving currently used COVID-19 vaccines with multiple booster shots. Therefore, there is an urgent need to develop next generation vaccines against VOCs like Omicron. In this study, we develop a panel of mRNA-LNP-based vaccines using the receptor binding domain (RBD) of Omicron and Delta variants, which are dominant in the current wave of COVID-19. In addition to the Omicron- and Delta-specific vaccines, the panel also includes a Hybrid vaccine that uses the RBD containing all 16 point-mutations shown in Omicron and Delta RBD, as well as a bivalent vaccine composed of both Omicron and Delta RBD-LNP in half dose. Interestingly, both Omicron-specific and Hybrid RBD-LNP elicited extremely high titer of neutralizing antibody against Omicron itself, but few to none neutralizing antibody against other SARS-CoV-2 variants. The bivalent RBD-LNP, on the other hand, generated antibody with broadly neutralizing activity against the wild-type virus and all variants. Surprisingly, similar cross-protection was also shown by the Delta-specifc RBD-LNP. Taken together, our data demonstrated that Omicron-specific mRNA vaccine can induce potent neutralizing antibody response against Omicron, but the inclusion of epitopes from other variants may be required for eliciting cross-protection. This study would lay a foundation for rational development of the next generation vaccines against SARS-CoV-2 VOCs.

Comparison of the clinical features, viral shedding and immune response in vaccine breakthrough infection by the Omicron and Delta variants (preprint)

BackgroundOn 26 November 2021, the World Health Organization designated the B.1.1.529 lineage of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) as the fifth variant of concern, Omicron. Infections have quickly spread worldwide, but understanding of the viral dynamics and the cytokine and cellular immunological response during infection remain limited.MethodsDetailed patient-level data from 174 age-matched patients with sequence confirmed Omicron or Delta infection admitted to the National Centre for Infectious Diseases, Singapore were analyzed in an observational cohort study. Peripheral blood samples for measurement of SARS-CoV-2 immunological parameters were obtained from a subset. Respiratory samples were collected for viral cultures and correlated to corresponding PCR cycle threshold (Ct) values. ResultsOmicron and Delta variant infections in this hospitalized cohort were mild with only 3 (3%) and 14 (16%) developing pneumonia respectively. Omicron infections were more likely to present with sore throat (46.0 vs x23.0%, p=0.005). Neutrophil counts and C-reactive protein (CRP) were significantly lower among the Omicron cohort (Median neutrophil 2.95 [IQR 2.16 – 3.96] vs 4.60 [IQR 3.76 – 6.10] x 109/L , p<0.001; Median CRP 5.7 [IQR 2.0 – 10.0] vs 12.0 [IQR 6.1 – 22.0] mg/L, p<0.001). Trough polymerase chain reaction (PCR) cycle threshold (Ct) values were significantly higher with Omicron infection (17.6 [IQR 16.3 – 19.3] vs 14.9 [IQR 13.9 – 19.0], p=0.001). The pattern and rate of rise in Ct values was similar between Omicron and Delta. At the time of infection, Omicron infected patients had lower levels of pro-inflammatory cytokines Vaccine breakthrough infections with the Omicron variant had a low concentration of proinflammatory cytokines, chemokines, and growth factors at the acute phase of infection, but a more robust IFN-γ response. Less dysregulated immune cell profiles were also observed, including a lower immature neutrophil cell count in Omicron breakthrough casesConclusionsOmicron infections resulted in mild vaccine breakthrough illness in the majority of patients. Compared with Delta, Omicron infections were more frequently associated with upper respiratory tract infections, had lower viral loads, lower levels of pro-inflammatory cytokines and less dysregulated immune cell profiles.

Chinese Consensus Report on Family-Based Helicobacter pylori Infection Control and Management (2021 Edition).

OBJECTIVE: Helicobacter pylori infection is mostly a family-based infectious disease. To facilitate its prevention and management, a national consensus meeting was held to review current evidence and propose strategies for population-wide and family-based H. pylori infection control and management to reduce the related disease burden. METHODS: Fifty-seven experts from 41 major universities and institutions in 20 provinces/regions of mainland China were invited to review evidence and modify statements using Delphi process and grading of recommendations assessment, development and evaluation system. The consensus level was defined as ≥80% for agreement on the proposed statements. RESULTS: Experts discussed and modified the original 23 statements on family-based H. pylori infection transmission, control and management, and reached consensus on 16 statements. The final report consists of three parts: (1) H. pylori infection and transmission among family members, (2) prevention and management of H. pylori infection in children and elderly people within households, and (3) strategies for prevention and management of H. pylori infection for family members. In addition to the 'test-and-treat' and 'screen-and-treat' strategies, this consensus also introduced a novel third 'family-based H. pylori infection control and management' strategy to prevent its intrafamilial transmission and development of related diseases. CONCLUSION: H. pylori is transmissible from person to person, and among family members. A family-based H. pylori prevention and eradication strategy would be a suitable approach to prevent its intra-familial transmission and related diseases. The notion and practice would be beneficial not only for Chinese residents but also valuable as a reference for other highly infected areas.

Durable T cell responses contrast with faster antibody waning in BNT162b2-vaccinated elderly at 6 month (preprint)

Efficient COVID-19 vaccines have been developed in record time. Here, we present findings from a comprehensive and integrated analysis of multiple compartments of the memory immune response in 312 individuals vaccinated with the BNT162b2 mRNA vaccine. Two vaccine doses induced high antibody and T cell responses in most individuals. However, antibody recognition of the Spike protein of delta variant was less efficient than that of the Wuhan strain. Age stratified analyses identified a group of low antibody responders where individuals ≥ 60 years were overrepresented. Waning of the antibody and cellular responses was observed in 30% of the vaccinees after six months. However, age did not influence the waning of these responses. Taken together, while individuals ≥ 60 years old took longer to acquire vaccine-induced immunity, they develop more sustained acquired immunity at six months post-vaccination. However, the higher proportion of older individuals in the group of antibody low responders and the lower antibody reactivity the Delta variant call for a booster immunization to increase immune responses and protection.

The Association Between Ventilatory Ratio and Mortality of Adult Patients with ARDS: Systematic Review and Meta-Analysis (preprint)

Background: Ventilatory ratio is a simple bedside index of impaired efficiency of ventilation and correlates well with physiological dead space fraction in patients with ARDS. So it was regarded as a dead-space marker associated with mortality in mechanically ventilated adults with ARDS. However, the association between VR and outcome of patients with ARDS remains largely unknown. Methods: : We searched articles in three electronic databases including PubMed, EMBASE and Web of Science. All the English publications up to 1 st Oct. 2021 will be searched without any restriction of countries. All the observational study that investigated the association between ventilatory ratio and the mortality of ARDS patients were identified in this meta-analysis. The main outcome was mortality. Summary estimates of effect using odds ratio (OR) for dichotomous outcomes with accompanying 95% confidence interval (CI) were expressed. Results: : A total of 9 trials enrolling 5638 patients were finally included in this meta-analysis. The results revealed that the use of ventilatory ratio could be significantly related to the mortality in adult ARDS (OR=1.27; 95% CI 1.10 to 1.47; P=0.001). Ventilatory ratio may have the capability of predicting the mortality of NON- COVID-related patients (OR 1.39, 95% CI 1.12 to 1.73 P = 0.003) while it has no predictable significance in patients with COVID (OR 1.18, 95% CI 0.94 to 1.48 P = 0.16). Importantly, the dynamic changes of VR adds more predictable value (OR 1.21 vs 1.19). Conclusion: Our study suggests that ventilatory ratio can be regarded as a valuable marker to predict the mortality of adult patients with ARDS. Compared to patients with COVID, ventilatory ratio is more predictable in patients with NON-COVID. What’s more, the dynamic changes of VR may have the potential to improve the prognostic value.

Sludge Disinfection, Reduction and Energy Recovery by Anaerobic Digestion with Thermal Hydrolysis Pretreatment

The disinfection of sewage sludge has become a hot focus due to the outbreak of COVID-19, which belongs to one aspect of sludge harmless treatment. However, the threat of sludge to the environment is not only caused by harmful microbes. It is necessary to explore energy recovery and resource utilization of sludge treatment that is based on harmlessness, volume reduction and stabilization. Various advanced technologies were integrated to build an advanced anaerobic digestion system in Beijing urban area, which has achieved a good sludge treatment effect in the past three years since its construction and operation. Thermal hydrolysis can effectively inactivate harmful microorganisms, the hygienic properties of advanced anaerobic digestion products can meet the standard, and its heavy metals contents are significantly lower than the related standard limit, which have changed to more stable forms. The unit treatment capacity of digester increased to 3.6 times that of conventional anaerobic digestion due to the modification effect of thermal hydrolysis on sludge properties, saving about 68% investment in digester construction. The degradation rate of organics increased from 40% to 52%, indicating that the stability of the digestion products was further improved, the residual solids were further reduced with the volume reduction rate of sludge increased from 28% to 66%. The biogas production rate was greatly increased with the improvement of organic conversion efficiency. Even operated below capacity, the advanced anaerobic digestion system still had the capacity to output 4 089×10~4 kW·h power under the condition that deducting the heat and power consumption of the system itself from the energy recovered. The treatment and disposal process of sludge were successfully connected by the resource utilization of advanced anaerobic digestion products. The large-scale production application in Beijing has created a successful and mature engineering reference for sludge advanced anaerobic digestion.

Spectral Image Contrast Based Flow Digital Nanoplasmon-metry For Ultrasensitive Antibody Detection (preprint)

Background: Gold nanoparticles (AuNPs) have been widely used as local surface plasmon resonance (LSPR) immunoassay for biomolecule sensings. It is primarily based on two conventional methods - absorption spectra and colorimetric. While the low figure of merit (FoM) of the LSPR and high-concentration AuNPs requirement restrict their limit of detection (LOD), which is around ng to μg per mL in antibody detection if there is no other signal or analytes amplification. By now, the improvement of sensitivity has bogged down for years. It also reveals a great challenge of the current LSPR immunoassay in biosensing - pushing the boundary of the current LOD. Results: : In this work, we developed a spectral image contrast-based flow digital nanoplasmon-metry(FDNM) to push the LOD boundary. Comparing the scattering image brightness of AuNPs in two neighboring wavelength bands near the LSPR peak, the signal of peak shift is extremely amplified and quickly detected. Introducing the digital analysis, the FDNM provides an ultra-high signal-to-noise ratio and less sample volume requirement. Compared to conventional analog LSPR immunoassay, FDNM has a LOD down to 1 pg mL -1 within only a 15-minute detection time and 500 μL sample volume. Antibody against spike proteins of SARS-CoV-2 in artificial saliva that contained various proteins was also conducted to validate the detection of FDNM in complicated samples. Of the detection, FDNM shows significant discrimination with a LOD of 10 pg mL -1 and a broad dynamic detecting range of five orders of magnitude. Conclusion: Together with the quick readout time and simple operation, this work outstandingly demonstrated the high sensitivity and selectivity of the developed FDNM in rapid antibody detection. The spectral image contrast and digital analysis further provide a new generation LSPR immunoassay of AuNPs.

Research progress on antiviral pharmacology and clinical research of chloroquine and hydroxychloroquine

Viral infectious disease have the characteristics of high pathogenicity, strong infectivity and great damage, which pose great threats to human health, and even to social development and stability. Therefore, discovering preventive and therapeutic drugs has always been the focus of society. Chloroquine and hydroxychloroquine are common drugs approved by the US Food and Drug Administration for the treatment of malaria and rheumatic immune disease, with low-budget as well as high safety. This review analyzed studies about the antiviral activity of chloroquine and hydroxychloroquine, including laboratory experiments and clinical trials about human coronavirus, HIV, Zika virus, and enterovirus E-7. The effects of many types of viruses suggestted that chloroquine and hydroxychloroquine possess great clinical application potential.

Neural Network Training with Highly Incomplete Datasets (preprint)

Neural network training and validation rely on the availability of large high-quality datasets. However, in many cases only incomplete datasets are available, particularly in health care applications, where each patient typically undergoes different clinical procedures or can drop out of a study. Since the data to train the neural networks need to be complete, most studies discard the incomplete datapoints, which reduces the size of the training data, or impute the missing features, which can lead to artefacts. Alas, both approaches are inadequate when a large portion of the data is missing. Here, we introduce GapNet, an alternative deep-learning training approach that can use highly incomplete datasets. First, the dataset is split into subsets of samples containing all values for a certain cluster of features. Then, these subsets are used to train individual neural networks. Finally, this ensemble of neural networks is combined into a single neural network whose training is fine-tuned using all complete datapoints. Using two highly incomplete real-world medical datasets, we show that GapNet improves the identification of patients with underlying Alzheimer's disease pathology and of patients at risk of hospitalization due to Covid-19. By distilling the information available in incomplete datasets without having to reduce their size or to impute missing values, GapNet will permit to extract valuable information from a wide range of datasets, benefiting diverse fields from medicine to engineering.

Air-writing recognition using reverse time ordered stroke context

Air-writing is a new human and smart device communication approach, permits users to write inputs in a natural and relentless way. This touch-less way can prevent users fromvirus infection such as COVID-19. Compared with othermethods, air writing ismore challenging due to its unique characteristics such as redundant lifting strokes, multiplicity (different writing styles from various users), and confusion (different character types written in air are similar). Without the need of any starting trigger, a novel reverse time-ordered algorithm is proposed in this paper toefficiently filter out unnecessary lifting strokes, and thus simplifies the matching procedure. As to the second and third issues, a tiered arrangement structure is proposed by sampling the air-writing results with various sampling rates to solvethe multiplicity and confusion problems. Analyzed with other recently proposed air writing algorithms, the proposed approach reaches satisfactory recognition accuracy (above 94%) without any starting triggers.

A glimpse into the diverse cellular immunity against SARS-CoV-2 (preprint)

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-specific cellular immune response may prove to be essential for long-term immune protection against the novel coronavirus disease 2019 (COVID-19). To assess COVID-19-specific immunity in the population, we synthesized selected peptide pools of SARS-CoV-2 structural and functional proteins, including Spike (S), Membrane (M), Envelope (E), Nucleocapsid (N) and Protease (P) as target antigens. Survey of the T cell precursur frequencies in healthy individuals specific to these viral antigens demonstrated a diverse cellular immunity, including high, medium, low and no responders. This was further confirmed by in vitro induction of anti-SARS-CoV-2 T cell immune responses using dendritic cell (DC)/T cell coculture, which supported the corresponding T cell precursor frequencies in each of the individuals tested. In general, the combination of all five viral antigen pools induced the strongest cellular immune response, yet individual donors responded differently to different viral antigens. Importantly, in vitro restimulation of the T cells with the DC-peptides induced increased anti-viral immune responses in all individuals even in the no responders, suggesting that repeated antigen stimulation could elicit a broad protection in immune naïve population. Our analysis recapitulates the critical role of cellular immunity in fighting COVID-19 and the importance of analyzing anti-SARS-CoV-2 T cell response in addition to antibody response in the population. Importance Facing the rapid evolving SARS-CoV-2 variants in the world, current emphasis on antibody-producing vaccines needs a quick revisit. The virus-specific cellular immunity may prove to be essential for long-term protection against COVID-19. This study designed a series of antigenic peptides encompassing the conserved and/or essential domains of Spike (S), Membrane (M), envelope (E), Nucleocapsid (N) and Protease (P) as targets to assess Covid-19-specific immunity in the population. The results demonstrated a diverse cellular immunity, including high, medium, low and no responders. This was verified by in vitro generation of anti-SARS-CoV-2 T-cells from these subjects. The study suggested that individuals responded differently to the different viral antigens, and importantly, repeated stimulation could produce virus specific T cells in all individuals, including the no responders. This study illustrates the needs for assessing anti-viral cellular immunity in addition to antibody response in the general population.