Convolutional neural network optimized by differential evolution for electrocardiogram classification

The Coronavirus disease 2019, or COVID-19, has shifted the medical paradigm from face-to-face to telehealth. Telehealth has become a vital resource to contain the virus spread and ensure the continued care of patients. In terms of preventing cardiovascular diseases, automating electrocardiogram (ECG) classification is a promising telehealth intervention. The healthcare service ensures that patient care is appropriate, comfortable, and accessible. Convolutional neural networks (CNNs) have demonstrated promising results in ECG categorization, which require high accuracy and short training time to ensure healthcare quality. This study proposes a one-dimensional-CNN (1D-CNN) arrhythmia classification based on the differential evolution (DE) algorithm to optimize the accuracy of ECG classification and training time. The performance of 1D-CNNs of different activation functions are optimized based on the standard DE algorithm. Finally, based on MIT-BIH and SCDH arrhythmia databases, the performances of optimized and unoptimized 1D-CNN are compared and analysed. Results show that the 1D-CNN optimized by the DE has higher accuracy in heartbeats classification. The optimized 1D-CNN improves from 97.6% to 99.5% on MIT-BIH and from 80.2% to 88.5% on SCDH. Therefore, the optimized 1D-CNN shows improvements of 1.9% and 8.3% in the two datasets, respectively. In addition, compared with the unoptimized 1D-CNN based on the same parameter settings, the optimized 1D-CNN has less training time. Under the conditions of ReLU function and 10 epochs, the training takes 9.22 s on MIT-BIH and 10.35 s on SCDH, reducing training time by 67.2% and 64.2%, respectively.

The Association Between the Albumin and Viral Negative Conversion Rate in Patients Infected with Novel Coronavirus Disease 2019 (COVID-19).

Purpose: The novel coronavirus disease 2019 (COVID-19) epidemic is the severe global pandemic with large numbers of infected cases and deaths in recent decades. The previous studies were all about the influence of albumin (ALB) for the severity and mortality of in-patients infected with COVID-19. But few studies exist about the influence factors to achieve viral negative conversion. Therefore, this study conducted an exploratory study to investigate the effect of albumin on negative conversion rate. Methods: Among the 190 hospitalized patients with moderate COVID-19 who had a course of disease longer than 30 days, 102 achieved viral negative conversion in 30-45 days and 88 not after 45 days. Taking other variables as concomitant variable, Cox proportional hazard regression model was applied to explore the influence of albumin to negative conversion rate under various factors. Results: By comparing patients who could and could not achieve the finally viral negative conversion, a possible nonlinear relationship between the continuous variables and clinical outcomes was examined by a restricted cubic spline regression model. An association was found between albumin levels and hazard ratio of viral negative conversion rate (P = 0.027). The increase of albumin was accompanied with decreases of hazard ratio of viral negative conversion rate (the value of albumin <38 g/L). But when the value of albumin was higher than 38 g/L, the hazard ratio of viral negative conversion rate approached 1, it means that albumin is not a risk factor for the viral negative conversion rate of COVID-19 disease. Conclusion: For patients with COVID-19, albumin is a common and observed laboratory parameter. It is associated with final viral negative conversion rate although its underlying mechanism and relationship with the viral negative conversion rate still need to be clarified.

Alcohol Consumption, Loneliness, Quality of Life, Social Media Usage and General Anxiety before and during the COVID-19 Pandemic in Singapore.

This cross-sectional study aims to identify factors associated with anxiety levels of adults living in Singapore before and during the COVID-19 pandemic. Data were collected using a web-based survey conducted from July to November 2020, accruing 264 eligible participants. Ordered logistic regression was used to assess factors associated with Generalized Anxiety Disorder-7 (GAD-7), ranked as minimal (0-4), mild (5-9), moderate (10-14), and severe (15-21) before and during the pandemic. About 74% of participants were female, 50% were aged 25-34, and 50% were married. The GAD-7 level went up from the pre-pandemic for moderate (12.5% to 16%) and severe GAD (2% to 11%). Alcohol consumption (AOR 1.79, 95% CI 1.04-3.06), loneliness (AOR 1.28, 95% CI 1.05-1.54), and difficulty in switching off social media (AOR 2.21, 95% CI 1.29-3.79) predicted increased GAD-7 levels. The quality of life (AOR 0.84, 95% CI 0.79-0.90) was significantly associated with decreased GAD-7 levels. The results heighten the awareness that early initiation of mental health support is crucial for the population in addition to the various financial support measures provided by the government as they are adapting to live with the COVID-19 pandemic.

Alcohol Consumption, Loneliness, Quality of Life, Social Media Usage, and General Anxiety Before and During the COVID-19 Pandemic in Singapore (preprint)

This study aims to identify factors associated with anxiety levels of adults living in Singapore before the pandemic and during the COVID-19 outbreak. Data were collected using a cross-sectional web-based survey conducted from July to November 2020 accruing 264 eligible participants. Ordered logistic regression was used to assess Generalized Anxiety Disorder-7 (GAD-7), ranked as minimal (0-4), mild (5-9), moderate (10-14), and severe (15-21) before the pandemic and during the pandemic. About 74% of participants were female, 50% were aged 25-34, and 50% were married. The GAD-7 level went up from pre-pandemic for both moderate (from 12.5% to 16%) and severe GAD (from 2% to 11%). Alcohol consumption (AOR 1.79, 95% CI 1.04-3.06), loneliness (AOR 1.28, 95% CI 1.05-1.54), and difficulty in switching off social media (AOR 2.21, 95% CI 1.29-3.79) predicted increased GAD-7 levels. The quality of life (AOR 0.84, 95% CI 0.79-0.90) was significantly associated with decreased GAD-7 levels. The results heighten the awareness that early initiation of mental health support is crucial for the population in addition to the various financial support measures provided by the government as they are adapting to live with the COVID-19 pandemic.

Assisting scalable diagnosis automatically via CT images in the combat against COVID-19.

The pandemic of Coronavirus Disease 2019 (COVID-19) is causing enormous loss of life globally. Prompt case identification is critical. The reference method is the real-time reverse transcription PCR (RT-PCR) assay, whose limitations may curb its prompt large-scale application. COVID-19 manifests with chest computed tomography (CT) abnormalities, some even before the onset of symptoms. We tested the hypothesis that the application of deep learning (DL) to 3D CT images could help identify COVID-19 infections. Using data from 920 COVID-19 and 1,073 non-COVID-19 pneumonia patients, we developed a modified DenseNet-264 model, COVIDNet, to classify CT images to either class. When tested on an independent set of 233 COVID-19 and 289 non-COVID-19 pneumonia patients, COVIDNet achieved an accuracy rate of 94.3% and an area under the curve of 0.98. As of March 23, 2020, the COVIDNet system had been used 11,966 times with a sensitivity of 91.12% and a specificity of 88.50% in six hospitals with PCR confirmation. Application of DL to CT images may improve both efficiency and capacity of case detection and long-term surveillance.

Temporal Dynamics in Clinical and Laboratory Manifestations During COVID-19 Progression (preprint)

BAckground Severe COVID-19 patients account for most of the mortality of this disease. Early detection of severe cases of the disease remains a major challenge. Here, we performed clinical and laboratory profiling of COVID-19 to explore the early warning indicators of severe cases.Methods An analysis of the evolution during the hospitalization of clinical and laboratory findings from 78 confirmed COVID-19 patients and the associated risk factors.Results Of the 78 patients who were classified as un-severe at admission, 60 patients(stable group) were stable as mild cases until discharge, and the remaining 18 patients progressed to severe cases(exacerbated group) during hospitalization. Compared with stable patients, exacerbated patients exhibited older, higher BMI values and higher proportion of smokers. In the exacerbated patients, the median time from onset to deterioration was 7.5 days. Before the time point(days 0–7 from onset), we observed higher-levels of White blood cells(WBC), neutrophil, Neutrophi-Lymphocyte-Ratio(NLR), Lactose-dehydrogenase(LDH), D-dimer, and lower-levels of albumin in the exacerbated group, compared with the stable group. In the second week after the time point, the exacerbated patients displayed lower numbers of lymphocytes, CD3+, and CD8+T-cells, and higher-levels of C-reactive protein(CRP), erythrocyte-sedimentation-rate(ESR), Alanine-aminotransferase(ALT),Aspartate-aminotransferase(AST), and Interleukin-6. In the third week, the highest temperature and the proportion of febrile patients declined. All of the laboratory indicators gradually improved.Conclusions Advanced age and smoking history could be risk factors for COVID-19 progression. In the early stage, high-levels of WBC and neutrophils, with noticeably increased LDH and D-dimer, could be early indicators of the disease’s conversion from mild to severe, followed by elevated inflammatory markers, liver enzymes, and decreased T-lymphocytes in the next week.

Co-infection of SARS-COV-2 and influenza A virus: a case series and fast review (preprint)

Background: COVID-19 has become a global pandemic. Studies about SARS-CoV-2 co-infection with influenza A virus (IAV) in the influenza season will contribute to the antiviral interventions of co-infected patients.Methods: A cohort of 145 COVID-19 patients in Wuhan union hospital were reviewed and we found 2 patients were co-infected with both SARS-CoV-2 and IAV. Then we searched from PubMed, Web of Science and CNKI with combinations of the following key words: “COVID-19, SARS-COV-2, influenza A and co-infection” from January 1 up to May 1, and 6 studies were included in this descriptive analysis. Results: Of the 13 co-infected patients, 2 were from Wuhan union hospital, another 11 were collected from the reports published on PubMed, Web of Science and CNKI. Of the 13 patients, the median age was 50 years (IQR, 40.5-67.5). Among the 13 patients, 7 (53.8%) were severe types. The most common symptoms among the 13 patients were cough (100%), fever (92.3%) and dyspnea (76.9%). 8 patients had lymphocytopenia on admission and all the 13 patients had abnormal radiological changes. The median time from symptom onset to hospital admission was 4.5 days (IQR, 2.75-5.5), and the median time of hospital stay was 17 days (IQR,15-20). Conclusion: Patients with both SARS-COV-2 and IAV infection showed similar changes in symptoms and radiological images with patients infected with SARS-COV-2 only.  SARS-COV-2 co-infection with IAV can lead to more severe clinical condition but had similar hospital stay compared with patients infected with SARS-COV-2 only in the fast review.

Field-deployable, rapid diagnostic testing of saliva samples for SARS-CoV-2. (preprint)

Abstract Rapid, scalable, point-of-need, COVID-19 diagnostic testing is necessary to safely re-open economies and prevent future outbreaks. We developed an assay that detects single copies of SARS-CoV-2 virus directly from saliva and swab samples in 30 min using a simple, one-step protocol that utilizes only a heat block and microcentrifuge tube prefilled with a mixture containing the necessary reagents and has a sensitivity and specificity of 97% and 100%, respectively.

Direct diagnostic testing of SARS-CoV-2 without the need for prior RNA extraction. (preprint)

The COVID-19 pandemic has resulted in an urgent global need for rapid, point-of-care diagnostic testing. Existing methods for nucleic acid amplification testing (NAAT) require an RNA extraction step prior to amplification of the viral RNA. This step necessitates the use of a centralized laboratory or complex and costly proprietary cartridges and equipment, and thereby prevents low-cost, scalable, point-of-care testing. We report the development of a highly sensitive and robust, easy-to-implement, SARS-CoV-2 test that utilizes isothermal amplification and can be run directly on viral transport media following a nasopharyngeal swab without the need for prior RNA extraction. Our assay provides visual results in 30 min with 85% sensitivity, 100% specificity, and a limit of detection (LoD) of 2.5 copies/l, and can be run using a simple heat block.

Interferon-a2b treatment for COVID-19 (preprint)

Summary Background The global pandemic of COVID-19 cases caused by infection with SARS-CoV-2 is ongoing. We describe here the clinical course of COVID-19 in a cohort of confirmed cases in Wuhan, China, treated with the repurposed potential experimental therapeutics IFN-2b, arbidol or a combination of IFN-2b plus arbidol. Methods 77 adults with confirmed COVID-19 were treated with either nebulized IFN-2b (5mU,b.i.d.), arbidol (200mg dispersible tablet, t.i.d.) or a combination of IFN-2b plus arbidol. Serial SARS-CoV-2 testing along with hematological measurements, including cell counts and blood biochemistry, serum cytokine levels, temperature and blood oxygen saturation levels were recorded for each patient during their hospital stay. Findings Treatment with IFN-2b with or without arbidol significantly reduced the duration of detectable virus in the upper respiratory tract and in parallel reduced duration of elevated blood levels for the inflammatory markers IL-6 and CRP. Interpretation IFN-2b should be investigated as therapy in COVID-19 cases.