Adaptive Feature Selection Guided Deep Forest for COVID-19 Classification With Chest CT.

Chest computed tomography (CT) becomes an effective tool to assist the diagnosis of coronavirus disease-19 (COVID-19). Due to the outbreak of COVID-19 worldwide, using the computed-aided diagnosis technique for COVID-19 classification based on CT images could largely alleviate the burden of clinicians. In this paper, we propose an Adaptive Feature Selection guided Deep Forest (AFS-DF) for COVID-19 classification based on chest CT images. Specifically, we first extract location-specific features from CT images. Then, in order to capture the high-level representation of these features with the relatively small-scale data, we leverage a deep forest model to learn high-level representation of the features. Moreover, we propose a feature selection method based on the trained deep forest model to reduce the redundancy of features, where the feature selection could be adaptively incorporated with the COVID-19 classification model. We evaluated our proposed AFS-DF on COVID-19 dataset with 1495 patients of COVID-19 and 1027 patients of community acquired pneumonia (CAP). The accuracy (ACC), sensitivity (SEN), specificity (SPE), AUC, precision and F1-score achieved by our method are 91.79%, 93.05%, 89.95%, 96.35%, 93.10% and 93.07%, respectively. Experimental results on the COVID-19 dataset suggest that the proposed AFS-DF achieves superior performance in COVID-19 vs. CAP classification, compared with 4 widely used machine learning methods.

Intervention effects in the transmission of COVID-19 depending on the detection rate and extent of isolation.

OBJECTIVE: In 2020, the coronavirus disease 2019 (COVID-19) respiratory infection is spreading in Korea. In order to prevent the spread of an infectious disease, infected people must be quickly identified and isolated, and contact with the infected must be blocked early. This study attempted to verify the intervention effects on the spread of an infectious disease by using these measures in a mathematical model. METHODS: We used the susceptible-infectious-recovery (SIR) model for a virtual population group connected by a special structured network. In the model, the infected state (I) was divided into I in which the infection is undetected and Ix in which the infection is detected. The probability of transitioning from an I state to Ix can be viewed as the rate at which an infected person is found. We assumed that only those connected to each other in the network can cause infection. In addition, this study attempted to evaluate the effects of isolation by temporarily removing the connection among these people. RESULTS: In Scenario 1, only the infected are isolated; in Scenario 2, those who are connected to an infected person and are also found to be infected are isolated as well. In Scenario 3, everyone connected to an infected person are isolated. In Scenario 3, it was possible to effectively suppress the infectious disease even with a relatively slow rate of diagnosis and relatively high infection rate. CONCLUSION: During the epidemic, quick identification of the infected is helpful. In addition, it was possible to quantitatively show through a simulation evaluation that the management of infected individuals as well as those who are connected greatly helped to suppress the spread of infectious diseases.

Trends in the prevalence of COVID-19 infection in Rio Grande do Sul, Brazil: repeated serological surveys. / Evolução da prevalência de infecção por COVID-19 no Rio Grande do Sul, Brasil: inquéritos sorológicos seriados.

COVID-19, the disease produced by the virus SARS-CoV-2, has spread quickly throughout the world, leading the World Health Organization to first classify it as an international health emergency and, subsequently, declaring it pandemic. The number of confirmed cases, as April 11, surpassed 1,700,000, but this figure does not reflect the prevalence of COVID-19 in the population as, in many countries, tests are almost exclusively performed in people with symptoms, particularly severe cases. To properly assess the magnitude of the problem and to contribute to the design of evidence-based policies for fighting COVID-19, one must accurately estimate the population prevalence of infection. Our study is aimed at estimating the prevalence of infected individuals in the state of Rio Grande do Sul, Brazil, to document how fast the infection spreads, and to estimate the proportion of infected persons who present or presented symptoms, as well as the proportion of asymptomatic infections. Four repeated serological surveys will be conducted in probability samples of nine sentinel cities every two weeks. Tests will be performed in 4,500 participants in each survey, totaling18,000 interviews. Interviews and tests will be conducted at the participants' household. A rapid test for the detection of antibodies will be used; the test was validated prior to the beginning of the fieldwork.

A COVID-19 é uma doença produzida pelo vírus SARS-CoV-2. Esse vírus se espalhou rapidamente pelo mundo, o que levou a Organização Mundial da Saúde a classificar a COVID-19 como uma emergência de saúde internacional e, posteriormente, a declará-la uma pandemia. O número de casos confirmados, no dia 11 de abril de 2020, já passa de 1.700.000, porém esses dados não refletem a real prevalência de COVID-19 na população, visto que, em muitos países, os testes são quase que exclusivamente realizados em pessoas com sintomas, especialmente os mais graves. Para definir políticas de enfrentamento, é essencial dispor de dados sobre a prevalência real de infecção na população. Este estudo tem por objetivos avaliar a proporção de indivíduos já infectados pelo SARS-CoV-2 no Rio Grande do Sul, Brasil, analisar a velocidade de expansão da infecção e estimar o percentual de infectados com e sem sintomas. Serão realizados quatro inquéritos sorológicos repetidos a cada 15 dias, com amostragem probabilística de nove cidades sentinela, em todas as sub-regiões do Estado. As entrevistas e testes ocorrerão no âmbito domiciliar. Serão utilizados testes rápidos para detecção de anticorpos, validados previamente ao início da coleta de dados.

Disparities in case frequency and mortality of coronavirus disease 2019 (COVID-19) among various states in the United States.

OBJECTIVE: To utilize publicly reported, state-level data to identify factors associated with the frequency of cases, tests, and mortality in the USA. MATERIALS AND METHODS: Retrospective study using publicly reported data collected included the number of COVID-19 cases, tests and mortality from March 14th through April 30th. Publicly available state-level data was collected which included: demographics comorbidities, state characteristics and environmental factors. Univariate and multivariate regression analyses were performed to identify the significantly associated factors with percent mortality, case and testing frequency. All analyses were state-level analyses and not patient-level analyses. RESULTS: A total of 1,090,500 COVID-19 cases were reported during the study period. The calculated case and testing frequency were 3332 and 19,193 per 1,000,000 patients. There were 63,642 deaths during this period which resulted in a mortality of 5.8%. Factors including to but not limited to population density (beta coefficient 7.5, p < .01), transportation volume (beta coefficient 0.1, p < .01), tourism index (beta coefficient -0.1, p = .02) and older age (beta coefficient 0.2, p = .01) are associated with case frequency and percent mortality. CONCLUSIONS: There were wide variations in testing and case frequencies of COVID-19 among different states in the US. States with higher population density had a higher case and testing rate. States with larger population of elderly and higher tourism had a higher mortality. Key messages There were wide variations in testing and case frequencies of COVID-19 among different states in the USA. States with higher population density had a higher case and testing rate. States with larger population of elderly and higher tourism had a higher mortality.

How are family doctors serving the Hong Kong community during the COVID-19 outbreak? A survey of HKCFP members.

INTRODUCTION: This study evaluated the preparedness of family doctors during the early phase of the coronavirus disease 2019 (COVID-19) outbreak in Hong Kong. METHODS: All members of the Hong Kong College of Family Physicians were invited to participate in a cross-sectional online survey using a 20-item questionnaire to collect information on practice preparedness for the COVID-19 outbreak through an email followed by a reminder SMS message between 31 January 2020 and 3 February 2020. RESULTS: Of 1589 family doctors invited, 491 (31%) participated in the survey, including 242 (49%) from private sector. In all, 98% surveyed doctors continued to provide clinical services during the survey period, but reduced clinic service demands were observed in 45% private practices and 24% public clinics. Almost all wore masks during consultation and washed hands between or before patient contact. Significantly more private than public doctors (80% vs 26%, P<0.001) experienced difficulties in stocking personal protective equipment (PPE); more public doctors used guidelines to manage suspected patients. The main concern of the respondents was PPE shortage. Respondents appealed for effective public health interventions including border control, quarantine measures, designated clinic setup, and public education. CONCLUSION: Family doctors from public and private sectors demonstrated preparedness to serve the community from the early phase of the COVID-19 outbreak with heightened infection control measures and use of guidelines. However, there is a need for support from local health authorities to secure PPE supply and institute public health interventions.

Development and Validation of a Multivariable Risk Prediction Model for COVID-19 Mortality in the Southern United States.

OBJECTIVE: To evaluate clinical characteristics of patients admitted to the hospital with coronavirus disease 2019 (COVID-19) in Southern United States and development as well as validation of a mortality risk prediction model. PATIENTS AND METHODS: Southern Louisiana was an early hotspot during the pandemic, which provided a large collection of clinical data on inpatients with COVID-19. We designed a risk stratification model to assess the mortality risk for patients admitted to the hospital with COVID-19. Data from 1673 consecutive patients diagnosed with COVID-19 infection and hospitalized between March 1, 2020, and April 30, 2020, was used to create an 11-factor mortality risk model based on baseline comorbidity, organ injury, and laboratory results. The risk model was validated using a subsequent cohort of 2067 consecutive hospitalized patients admitted between June 1, 2020, and December 31, 2020. RESULTS: The resultant model has an area under the curve of 0.783 (95% CI, 0.76 to 0.81), with an optimal sensitivity of 0.74 and specificity of 0.69 for predicting mortality. Validation of this model in a subsequent cohort of 2067 consecutively hospitalized patients yielded comparable prognostic performance. CONCLUSION: We have developed an easy-to-use, robust model for systematically evaluating patients presenting to acute care settings with COVID-19 infection.

Fluctuation of Public Interest in COVID-19 in the United States: Retrospective Analysis of Google Trends Search Data.

BACKGROUND: In the absence of vaccines and established treatments, nonpharmaceutical interventions (NPIs) are fundamental tools to control coronavirus disease (COVID-19) transmission. NPIs require public interest to be successful. In the United States, there is a lack of published research on the factors that influence public interest in COVID-19. Using Google Trends, we examined the US level of public interest in COVID-19 and how it correlated to testing and with other countries. OBJECTIVE: The aim of this study was to determine how public interest in COVID-19 in the United States changed over time and the key factors that drove this change, such as testing. US public interest in COVID-19 was compared to that in countries that have been more successful in their containment and mitigation strategies. METHODS: In this retrospective study, Google Trends was used to analyze the volume of internet searches within the United States relating to COVID-19, focusing on dates between December 31, 2019, and March 24, 2020. The volume of internet searches related to COVID-19 was compared to that in other countries. RESULTS: Throughout January and February 2020, there was limited search interest in COVID-19 within the United States. Interest declined for the first 21 days of February. A similar decline was seen in geographical regions that were later found to be experiencing undetected community transmission in February. Between March 9 and March 12, 2020, there was a rapid rise in search interest. This rise in search interest was positively correlated with the rise of positive tests for SARS-CoV-2 (6.3, 95% CI -2.9 to 9.7; P<.001). Within the United States, it took 52 days for search interest to rise substantially after the first positive case; in countries with more successful outbreak control, search interest rose in less than 15 days. CONCLUSIONS: Containment and mitigation strategies require public interest to be successful. The initial level of COVID-19 public interest in the United States was limited and even decreased during a time when containment and mitigation strategies were being established. A lack of public interest in COVID-19 existed in the United States when containment and mitigation policies were in place. Based on our analysis, it is clear that US policy makers need to develop novel methods of communicating COVID-19 public health initiatives.

Early prediction and identification for severe patients during the pandemic of COVID-19: A severe COVID-19 risk model constructed by multivariate logistic regression analysis.

BACKGROUND: As an emergent and fulminant infectious disease, Corona Virus Disease 2019 (COVID-19) has caused a worldwide pandemic. The early identification and timely treatment of severe patients are crucial to reducing the mortality of COVID-19. This study aimed to investigate the clinical characteristics and early predictors for severe COVID-19, and to establish a prediction model for the identification and triage of severe patients. METHODS: All confirmed patients with COVID-19 admitted by the Second Affiliated Hospital of Air Force Medical University were enrolled in this retrospective non-interventional study. The patients were divided into a mild group and a severe group, and the clinical data were compared between the two groups. Univariate and multivariate analysis were used to identify the independent early predictors for severe COVID-19, and the prediction model was constructed by multivariate logistic regression analysis. Receiver operating characteristic (ROC) curve was used to evaluate the predictive value of the prediction model and each early predictor. RESULTS: A total of 40 patients were enrolled in this study, of whom 19 were mild and 21 were severe. The proportions of patients with venerable age (≥60 years old), comorbidities, and hypertension in severe patients were higher than that of the mild (P < 0.05). The duration of fever and respiratory symptoms, and the interval from illness onset to viral clearance were longer in severe patients (P < 0.05). Most patients received at least one form of oxygen treatments, while severe patients required more mechanical ventilation (P < 0.05). Univariate and multivariate analysis showed that venerable age, hypertension, lymphopenia, hypoalbuminemia and elevated neutrophil lymphocyte ratio (NLR) were the independent high-risk factors for severe COVID-19. ROC curves demonstrated significant predictive value of age, lymphocyte count, albumin and NLR for severe COVID-19. The sensitivity and specificity of the newly constructed prediction model for predicting severe COVID-19 was 90.5% and 84.2%, respectively, and whose positive predictive value, negative predictive value and crude agreement were all over 85%. CONCLUSIONS: The severe COVID-19 risk model might help clinicians quickly identify severe patients at an early stage and timely take optimal therapeutic schedule for them.

'Self-collected upper respiratory tract swabs for COVID-19 test': A feasible way to increase overall testing rate and conserve resources in South Africa.

Disparity in the testing rate of SARS-CoV-2 amongst different countries and regions is a very big challenge in understanding the COVID-19 pandemic. Although some developed countries have a very high testing rate and subsequently a high number of confirmed cases, less developed countries have a low testing rate and an illusive positivity rate. Collection of the upper respiratory specimen is not often comfortable. The discomfort could be accompanied with epistaxis and headache in some patients. The trained personnel taking the swab is forced to protect self with personal protective equipment (PPE) to avoid infections that may result from the patient due to provoked cough, sneezing and spitting. This study looks into an efficient means of increasing the testing rate for COVID 19 without compromising the quality. A literature review was conducted on the different modalities of collecting upper respiratory specimens and assessing the efficacy of samples collected using different methods in terms of the laboratory yield of different pathogens. Self-collection of upper respiratory tract specimen for diagnostic purposes is not new. Studies have demonstrated that trained staff-collected nasal swabs are not in any way superior to self-collected or parent-assisted swabs. The laboratory yield of different specimens is not determined by who took the sample but by the anatomical site from where the specimen was collected. Self collection of the upper respiratory swabs will not only increase the testing rate but also preserve the scarce PPE and reduces health care worker's COVID 19 infection rate in South Africa.

Risk Factors Associated With Mortality Among Patients With COVID-19 in Intensive Care Units in Lombardy, Italy.

Importance: Many patients with coronavirus disease 2019 (COVID-19) are critically ill and require care in the intensive care unit (ICU). Objective: To evaluate the independent risk factors associated with mortality of patients with COVID-19 requiring treatment in ICUs in the Lombardy region of Italy. Design, Setting, and Participants: This retrospective, observational cohort study included 3988 consecutive critically ill patients with laboratory-confirmed COVID-19 referred for ICU admission to the coordinating center (Fondazione IRCCS [Istituto di Ricovero e Cura a Carattere Scientifico] Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy) of the COVID-19 Lombardy ICU Network from February 20 to April 22, 2020. Infection with severe acute respiratory syndrome coronavirus 2 was confirmed by real-time reverse transcriptase-polymerase chain reaction assay of nasopharyngeal swabs. Follow-up was completed on May 30, 2020. Exposures: Baseline characteristics, comorbidities, long-term medications, and ventilatory support at ICU admission. Main Outcomes and Measures: Time to death in days from ICU admission to hospital discharge. The independent risk factors associated with mortality were evaluated with a multivariable Cox proportional hazards regression. Results: Of the 3988 patients included in this cohort study, the median age was 63 (interquartile range [IQR] 56-69) years; 3188 (79.9%; 95% CI, 78.7%-81.1%) were men, and 1998 of 3300 (60.5%; 95% CI, 58.9%-62.2%) had at least 1 comorbidity. At ICU admission, 2929 patients (87.3%; 95% CI, 86.1%-88.4%) required invasive mechanical ventilation (IMV). The median follow-up was 44 (95% CI, 40-47; IQR, 11-69; range, 0-100) days; median time from symptoms onset to ICU admission was 10 (95% CI, 9-10; IQR, 6-14) days; median length of ICU stay was 12 (95% CI, 12-13; IQR, 6-21) days; and median length of IMV was 10 (95% CI, 10-11; IQR, 6-17) days. Cumulative observation time was 164 305 patient-days. Hospital and ICU mortality rates were 12 (95% CI, 11-12) and 27 (95% CI, 26-29) per 1000 patients-days, respectively. In the subgroup of the first 1715 patients, as of May 30, 2020, 865 (50.4%) had been discharged from the ICU, 836 (48.7%) had died in the ICU, and 14 (0.8%) were still in the ICU; overall, 915 patients (53.4%) died in the hospital. Independent risk factors associated with mortality included older age (hazard ratio [HR], 1.75; 95% CI, 1.60-1.92), male sex (HR, 1.57; 95% CI, 1.31-1.88), high fraction of inspired oxygen (Fio2) (HR, 1.14; 95% CI, 1.10-1.19), high positive end-expiratory pressure (HR, 1.04; 95% CI, 1.01-1.06) or low Pao2:Fio2 ratio (HR, 0.80; 95% CI, 0.74-0.87) on ICU admission, and history of chronic obstructive pulmonary disease (HR, 1.68; 95% CI, 1.28-2.19), hypercholesterolemia (HR, 1.25; 95% CI, 1.02-1.52), and type 2 diabetes (HR, 1.18; 95% CI, 1.01-1.39). No medication was independently associated with mortality (angiotensin-converting enzyme inhibitors HR, 1.17; 95% CI, 0.97-1.42; angiotensin receptor blockers HR, 1.05; 95% CI, 0.85-1.29). Conclusions and Relevance: In this retrospective cohort study of critically ill patients admitted to ICUs in Lombardy, Italy, with laboratory-confirmed COVID-19, most patients required IMV. The mortality rate and absolute mortality were high.

Impact of COVID-19 Pandemic on Global Poliovirus Surveillance.

On January 30, 2020, the World Health Organization (WHO) declared coronavirus disease 2019 (COVID-19) a Public Health Emergency of International Concern (1). On March 24, 2020, the Global Polio Eradication Initiative (GPEI) suspended all polio supplementary immunization activities and recommended the continuation of polio surveillance (2). In April 2020, GPEI shared revised polio surveillance guidelines in the context of the COVID-19 pandemic, which focused on reducing the risk for transmission of SARS-CoV-2, the virus that causes COVID-19, to health care workers and communities by modifying activities that required person-to-person contact, improving hand hygiene and personal protective equipment use practices, and overcoming challenges related to movement restrictions, while continuing essential polio surveillance functions (3). GPEI assessed the impact of the COVID-19 pandemic on polio surveillance by comparing data from January to September 2019 to the same period in 2020. Globally, the number of acute flaccid paralysis (AFP) cases reported declined 33% and the mean number of days between the second stool collected and receipt by the laboratory increased by 70%. Continued analysis of AFP case reporting and stool collection is critical to ensure timely detection and response to interruptions of polio surveillance.