ABSTRACT
(Aim) COVID-19 has caused more than 2.28 million deaths till 4/Feb/2021 while it is still spreading across the world. This study proposed a novel artificial intelligence model to diagnose COVID-19 based on chest CT images. (Methods) First, the two-dimensional fractional Fourier entropy was used to extract features. Second, a custom deep stacked sparse autoencoder (DSSAE) model was created to serve as the classifier. Third, an improved multiple-way data augmentation was proposed to resist overfitting. (Results) Our DSSAE model obtains a micro-averaged F1 score of 92.32% in handling a four-class problem (COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis, and healthy control). (Conclusion) Our method outperforms 10 state-of-the-art approaches. © 2021 Copyright held by the owner/author(s). Publication rights licensed to ACM.
ABSTRACT
(Aim) To make a more accurate and precise COVID-19 diagnosis system, this study proposed a novel deep rank-based average pooling network (DRAPNet) model, i.e., deep rank-based average pooling network, for COVID-19 recognition. (Methods) 521 subjects yield 1164 slice images via the slice level selection method. All the 1164 slice images comprise four categories: COVID-19 positive;community-acquired pneumonia;second pulmonary tuberculosis;and healthy control. Our method firstly introduced an improved multiple-way data augmentation. Secondly, an n-cony rank-based average pooling module (NRAPM) was proposed in which rank-based pooling-particularly, rank-based average pooling (RAP)-was employed to avoid overfitting. Third, a novel DRAPNet was proposed based on NRAPM and inspired by the VGG network. Grad-CAM was used to generate heatmaps and gave our AI model an explainable analysis. (Results) Our DRAPNet achieved a micro-averaged F1 score of 95.49% by 10 runs over the test set. The sensitivities of the four classes were 95.44%, 96.07%, 94.41%, and 96.07%, respectively. The precisions of four classes were 96.45%, 95.22%, 95.05%, and 95.28%, respectively. The F1 scores of the four classes were 95.94%, 95.64%, 94.73%, and 95.67%, respectively. Besides, the confusion matrix was given. (Conclusions) The DRAPNet is effective in diagnosing COVID-19 and other chest infectious diseases. The RAP gives better results than four other methods: strided convolution, l(2)-norm pooling, average pooling, and max pooling.
ABSTRACT
Since 2019, the coronavirus disease-19 (COVID-19) has been spreading rapidly worldwide, posing an unignorable threat to the global economy and human health. It is a disease caused by severe acute respiratory syndrome coronavirus 2, a single-stranded RNA virus of the genus Betacoronavirus. This virus is highly infectious and relies on its angiotensin-converting enzyme 2-receptor to enter cells. With the increase in the number of confirmed COVID-19 diagnoses, the difficulty of diagnosis due to the lack of global healthcare resources becomes increasingly apparent. Deep learning-based computer-aided diagnosis models with high generalisability can effectively alleviate this pressure. Hyperparameter tuning is essential in training such models and significantly impacts their final performance and training speed. However, traditional hyperparameter tuning methods are usually time-consuming and unstable. To solve this issue, we introduce Particle Swarm Optimisation to build a PSO-guided Self-Tuning Convolution Neural Network (PSTCNN), allowing the model to tune hyperparameters automatically. Therefore, the proposed approach can reduce human involvement. Also, the optimisation algorithm can select the combination of hyperparameters in a targeted manner, thus stably achieving a solution closer to the global optimum. Experimentally, the PSTCNN can obtain quite excellent results, with a sensitivity of 93.65% ± 1.86%, a specificity of 94.32% ± 2.07%, a precision of 94.30% ± 2.04%, an accuracy of 93.99% ± 1.78%, an F1-score of 93.97% ± 1.78%, Matthews Correlation Coefficient of 87.99% ± 3.56%, and Fowlkes-Mallows Index of 93.97% ± 1.78%. Our experiments demonstrate that compared to traditional methods, hyperparameter tuning of the model using an optimisation algorithm is faster and more effective. © 2023 Centro Regional de Invest. Cientif. y Tecn.. All rights reserved.
ABSTRACT
This editorial presents the recent advances and challenges of deep learning. We reviewed four main challenges: heterogeneity, copious size, reproducibility crisis, and explainability. Finally, we present the prospect of deep learning in industrial applications. © 2023, Bentham Science Publishers. All rights reserved.
ABSTRACT
This editorial presents the recent advances and challenges of deep learning. We reviewed four main challenges: heterogeneity, copious size, reproducibility crisis, and explainability. Finally, we present the prospect of deep learning in industrial applications. © 2023, Bentham Science Publishers. All rights reserved.
ABSTRACT
In this paper, we proposed COVID-19 lung CT (computed tomography) images recognition with superscalar winograd circuit based on VGG19. We adopt the VGG-19 machine learning architecture to recognize lung CT images and speed up neural network operations through Superscalar Winograd Circuit. After a series of experiments, our proposed method has a high pneumonia recognition rate and high computational efficiency. © 2022 IEEE.
ABSTRACT
Background. The COVID-19 pandemic has caused dramatic changes in the epidemiology of many diseases globally due to various changes in exposure to different pathogens, social restrictions, and demographic shifts. A university student health center located in the U.S. Midwest detected an increase in latent tuberculosis infection (LTBI) rate among its incoming international students (INTS) from 5.7% to 8.1% in the fall semesters of 2019 and 2021, respectively. We describe our approach to investigating the increase in LTBI rate at a university campus in a low-endemicity area. Methods. Factors that may affect LTBI rates were evaluated. LTBI testing policy and methods were reviewed. Medical and lab staff were interviewed regarding the consistency of specimen collection, transport, and processing. LTBI risks in the general population such as older age, male gender, and country of origin (COO) were also considered. Factors that were expected to be uncommon in the INTS (homelessness, incarceration, and illicit drug abuse) were not evaluated. Results. No changes in the INTS screening policy were noted. All incoming INTS were screened for LTBI during initial health screening, regardless of COO. The same manufacturer's QuantiFERON-TB Plus test was utilized. Compared to previous years, no inconsistencies in the testing logistics were reported. A total of 1,016 INTS were screened in 2019 and 1,179 in 2021. There were no significant differences in average age in years (23.1 vs. 23.3) or male gender (59.6% vs. 56.8%) between 2019 and 2021, respectively. Most INTS came from two countries (A and B). Country A was COO of 21.6% of INTS in 2019, which dropped to 8.4% in 2021. Country B was COO of 44.8% of INTS in 2019, which increased to 57.6% in 2021 (p< 0.001;Figure 1). Although LTBI rates within each country (A and B) remained similar before and during the COVID-19 pandemic (Figure 2), country B had consistently higher rates than country A (p< 0.001), which contributed to the overall increased rate of LTBI in 2021. (Table Presented) Conclusion. Evaluating changes in COO of INTS is essential in investigating trends in LTBI rates at otherwise low-endemicity universities. In our investigation, demographic changes in university admissions over two years relative to COVID-19 pandemic restrictions contributed to an increase in the overall LTBI rate.
ABSTRACT
Objective: To analyze the performance of emergency response to 2019 novel coronavirus (2019-nCoV) positive cases in an international test competition in an Winter Olympic Game venue and provide evidences for the COVID-19 prevention and control in similar competitions. Methods: A retrospective analysis on the epidemiological investigation and nucleic acid test results of the cases, the implementation of prevention and control measures, including the communication with sport teams and others, was conducted. Results: The positive cases of 2019-nCoV among entering people were detected before entry, at airport, hotel and venue. Two positive cases were reported before entry, 2 positive cases infected previously and 3 asymptomatic cases were reported after the entry. The venue public health team and local CDC conducted epidemiological investigation and contact assessment jointly in a timely and efficient manner. No local secondary transmission occurred, but the nucleic acid test results of positive persons fluctuated, posing serious challenges to the implementation of prevention and control measures. Conclusion: In large scale international competition, there is high risk of imported COVID-19. It is necessary to fully consider the fluctuation of nucleic acid test results, the criteria for determination and cancellation of positive results and give warm care to positive cases in the emergency response.
Subject(s)
COVID-19 , Humans , SARS-CoV-2 , Retrospective Studies , SeasonsABSTRACT
Purpose: COVID-19 has led to 4 million deaths worldwide since 2019. COVID-19 patients with cancers likely express biomarker changes in circulation. While many biomarker studies focused on COVID-19 diagnosis and prognosis, the panel of biomarkers used in SARS-CoV-2 infected cancer patients for COVID-19 severity and prognosis are largely unclear. Therefore, this systematic review aims to determine what biomarkers have been measured in cancer patients with COVID-19 and their prognostic utility. Methods: A systematic literature review in PubMed, Embase, and Scopus was performed on June 16th, 2021. The search keywords coronavirus, neoplasm, biomarkers, and disease progression were used to filter out 17 eligible studies, which were then carefully evaluated. Results: A total of 4,168 patients from 17 eligible articles were included in this study. Sixteen types of cancer and 60 biomarkers were identified. The majority of changed biomarkers in the cancer patients with COVID-19 compared to the healthy group and non-cancer patients with COVID-19 were biochemical and inflammatory markers. The up-regulated markers, including CRP, D-dimer, ferritin, IL-2R, IL-6, LDH, and PCT, were identified in eligible studies. Albumin and hemoglobin were significantly down-regulated in cancer patients with COVID-19. Additionally, we observed that the SARS-CoV-2 infected cancer patients with lower levels of CRP, ferritin, and LDH successfully survived from antiviral drug and immunotherapy for COVID-19 treatments. Conclusion: Several important clinical biomarkers, such as CRP, ferritin, and LDH, may serve as the prognostic markers to predict the outcomes following COVID-19 treatment and monitor the deterioration of COVID-19 in cancer patients.
ABSTRACT
(Aim) COVID-19 has caused more than 2.28 million deaths till 4/Feb/2021 while it is still spreading across the world. This study proposed a novel artificial intelligence model to diagnose COVID-19 based on chest CT images. (Methods) First, the two-dimensional fractional Fourier entropy was used to extract features. Second, a custom deep stacked sparse autoencoder (DSSAE) model was created to serve as the classifier. Third, an improved multiple-way data augmentation was proposed to resist overfitting. (Results) Our DSSAE model obtains a micro-averaged F1 score of 92.32% in handling a four-class problem (COVID-19, community-acquired pneumonia, secondary pulmonary tuberculosis, and healthy control). (Conclusion) Our method outperforms 10 state-of-the-art approaches.
ABSTRACT
Although reduced working time and furlough policy initiatives are widely regarded as important for economic and business reasons, little is known about their impacts on workers' mental health at the onset of COVID-19 pandemic. Using data from the UK Household Longitudinal Panel Study data from 2018 to February 2020 and April 2020 and change score analysis, this study aims to compare mental health changes between those who worked reduced hours, were furloughed and left/lost paid work. The results suggest that at the onset of COVID-19 reduced working time and furlough can protect workers' mental health, but only for men not for women. The gender differences remain significant even after controlling for housework and childcare responsibilities at the onset of COVID-19. These results highlight the importance of distributing paid work more equitably and formulating gender-sensitive labour market policies in protection of workers' mental health.
ABSTRACT
The rapid global spread of COVID-19 disease poses a huge threat to human health and the global economy. The rapid increase in the number of patients diagnosed has strained already scarce healthcare resources to track and treat Covid-19 patients in a timely and effective manner. The search for a fast and accurate way to diagnose Covid-19 has attracted the attention of many researchers. In our study, a deep learning framework for the Covid-19 diagnosis task was constructed using wavelet entropy as a feature extraction method and a feedforward neural network classifier, which was trained using an adaptive particle swarm algorithm. The model achieved an average sensitivity of 85.14% ± 2.74%, specificity of 86.76% ± 1.75%, precision of 86.57% ± 1.36%, accuracy of 85.95% ± 1.14%, and F1 score of 85.82% ± 1.30%, Matthews correlation coefficient of 71.95 ± 2.26%, and Fowlkes-Mallows Index of 85.83% ± 1.30%. Our experiments validate the usability of wavelet entropy-based feature extraction methods in the medical image domain and show the non-negligible impact of different optimisation algorithms on the models by comparing them with other models. © 2022, Springer Nature Switzerland AG.
ABSTRACT
Coronavirus disease 2019 (COVID-19) generated a global public health emergency since December 2019, causing huge economic losses. To help radiologists strengthen their recognition of COVID-19 cases, we developed a computer-aided diagnosis system based on deep learning to automatically classify chest computed tomography-based COVID-19, Tuberculosis, and healthy control subjects. Our novel classification model AdaD-FNN sequentially transfers the trained knowledge of an FNN estimator to the next FNN estimator while updating the weights of the samples in the training set with a decaying learning rate. This model inhibits the network from remembering the noisy information and improves the learning of complex patterns in the hard-to-identify samples. Moreover, we designed a novel image preprocessing model F-U2MNet-C by enhancing the image features using fuzzy stacking and eliminating the interference factors using U2MNet segmentation. Extensive experiments are conducted on four publicly available datasets namely, TLDCA, UCSD-Al4H, SARS-CoV-2, TCIA, and the obtained classification accuracies are 99.52%, 92.96%, 97.86%, 91.97%. Our novel system gives out compelling performance for assisting COVID-19 detection when compared with 22 state-of-the-art methods. We hope to help link together biomedical research and artificial intelligence and to assist the diagnosis of doctors, radiologists, and inspectors at each epidemic prevention site in the real world.
ABSTRACT
Objective: Overview the systematic review/Meta analysis of Lianhua Qingwen (连花清瘟) combined with conventional western medicine in the treatment of coronavirus disease 2019 (COVID-19). Methods: Systematic reviews/Meta-analysis of Lianhua Qingwen combined with western conventional in the treatment of COVID-19 from CNKI, Wanfang, CBM, VIP, PubMed, Embase, Cochrane Library, and Web of Sciencewere search, retrieved as of October 1, 2021. Two investigators screened the literature according to the inclusion and exclusion criteria, and determined the final inclusion of the literature. AMSTAR-2 scale, GRADE system, and PRISMA statements were used to evaluate the methodological quality and GRADE the evidence quality. Results: A total of eight systematic reviews/Meta analyses were included, including six in Chinese and four in English. The quality evaluation and evidence quality classification results show that the quality of the literature and the level of evidence were low. Conclusion: The existing evidence shows that Lianhua Qingwen combined with conventional western has a good effectin the treatment of COVID-19. However, due to the low methodological quality and evidence quality level of the systematic review/Meta analysis and the low level of evidence quality, more high-quality researchs are needed to obtain high-quality research results for verification.
ABSTRACT
Background. COVID-19 pneumonia can be indistinguishable from other infectious respiratory etiologies, so providers are challenged with deciding whether empiric antibiotics should be prescribed to hospitalized patients with SARS-CoV-2. This study aimed to evaluate predictors of respiratory bacterial co-infections (RBCI) in hospitalized patients with COVID-19. Methods. Retrospective study evaluating COVID-19 inpatients from Feb 1, 2020 to Sept 30, 2020 at a tertiary academic medical center. Patients with RBCI were matched with three COVID-19 inpatients lacking RBCI admitted within 7 days of each other. The primary objectives of this study were to determine the prevalence of and identify variables associated with RBCI in COVID-19 inpatients. Secondary outcomes included length of stay and mortality. Data collected included demographics;inflammatory markers;bacterial culture/antigen results;antibiotic exposure;and COVID-19 severity. Wilcoxon rank sum, Chi Square tests, or Fisher's exact tests were utilized as appropriate. A multivariable logistic regression (MLR) model was conducted to identify covariates associated with RBCI. Results. Seven hundred thirty-five patients were hospitalized with COVID-19 during the study period. Of these, 82 (11.2%) had RBCI. Fifty-seven of these patients met inclusion criteria and were matched to three patients lacking RBCI (N = 228 patients). Patients with RBCI were more likely to receive antibiotics [57 (100%) vs. 130 (76%), p < 0.0001] and for a longer cumulative duration [19 (13-33) vs. 8 (4-13) days, p < 0.0001] compared to patients lacking RBCI. The MLR model revealed risk factors of RBCI to be admission from SNF/LTAC/NH (AOR 6.8, 95% CI 2.6-18.2), severe COVID-19 (AOR 3.03, 95% CI 0.78-11.9), and leukocytosis (AOR 3.03, 95% CI 0.99-1.16). Conclusion. Although RBCI is rare in COVID-19 inpatients, antibiotic use is common. COVID-19 inpatients may be more likely to have RBCI if they are admitted from a SNF/LTAC/NH, have severe COVID-19, or present with leukocytosis. Early and prompt recognition of RBCI predictors in COVID-19 inpatients may facilitate timely antimicrobial therapy while improving antimicrobial stewardship among patients at low risk for co-infection.
ABSTRACT
The new coronavirus COVID-19 has been spreading all over the world in the last six months, and the death toll is still rising. The accurate diagnosis of COVID-19 is an emergent task as to stop the spreading of the virus. In this paper, we proposed to leverage image feature fusion for the diagnosis of COVID-19 in lung window computed tomography (CT). Initially, ResNet-18 and ResNet-50 were selected as the backbone deep networks to generate corresponding image representations from the CT images. Second, the representative information extracted from the two networks was fused by discriminant correlation analysis to obtain refined image features. Third, three randomized neural networks (RNNs): extreme learning machine, Schmidt neural network and random vector functional-link net, were trained using the refined features, and the predictions of the three RNNs were ensembled to get a more robust classification performance. Experiment results based on five-fold cross validation suggested that our method outperformed state-of-the-art algorithms in the diagnosis of COVID-19.
ABSTRACT
SARS-CoV-2 exploits the host cellular machinery for virus replication leading to the acute syndrome of coronavirus disease 2019 (COVID-19). Growing evidence suggests SARS-CoV-2 also exacerbates many chronic diseases, including cancers. As mutations on the spike protein (S) emerged as dominant variants that reduce vaccine efficacy, little is known about the relation between SARS-CoV-2 virus variants and cancers. Compared to the SARS-CoV-2 wild-type, the Gamma variant contains two additional NXT/S glycosylation motifs on the S protein. The hyperglycosylated S of Gamma variant is more stable, resulting in more significant epithelial-mesenchymal transition (EMT) potential. SARS-CoV-2 infection promoted NF-κB signaling activation and p65 nuclear translocation, inducing Snail expression. Pharmacologic inhibition of NF-κB activity by nature food compound, I3C suppressed viral replication and Gamma variant-mediated breast cancer metastasis, indicating that NF-κB inhibition can reduce chronic disease in COVID-19 patients. Our study revealed that the Gamma variant of SARS-CoV-2 activates NF-κB and, in turn, triggers the pro-survival function for cancer progression.
ABSTRACT
Long-range pollution transport (LRT) events have a wide impact across East Asia, but are often difficult to track due to imprecise emission inventories and changing domain scales as the plume moves from source to receptor locations. This study adjusts a bottom-up emission inventory based on changes in remotely sensed NO2 column densities for a source region of East Asia, then with CMAQv5.2.1 simulates transport of LRT plumes to Taiwan. Adjustment of an emissions inventory based on satellite measurements during the COVID-19 lockdown in China led to a -59% reduction in emissions over the relevant source area in China compared to base emissions. As a result, PM2.5 mass concentrations were reproduced to match observations (mean fractional bias, MFB of -13.9% and 18.5% at a remote and urban station) as the plume passed through northern Taiwan. Furthermore, the OMI-adjusted emissions simulation brought all of the major PM2.5 components to within -50% of the measured values. Another LRT event from 2018 with more subtle OMI-adjustments to the emissions was also simulated and with improved overall PM2.5 mass concentration at the northern tip of Taiwan (MFB: -91.5%) compared to the base model (MFB: -102.1%), and an acceptable index of agreement (0.78). For the 2018 event, non sea-salt sulfate concentrations were consistently underpredicted (0.2-0.4), while nitrate concentrations were overpredicted by up to factor of 11. Copernicus Atmosphere Monitoring Service (CAMS) reanalysis of the PM(2.)5 concentrations shows high sulfate concentrations in eastern China in the areas associated with 72-h back-trajectories from northern Taiwan during both events, lending support for future model investigations of sulfate source area production and transport to Taiwan. In order to better track these LRT events out of East Asia and optimize OMI-adjustment methodology, it is recommended to explore other satellite-based products to map unaccounted for SO2 sources upstream of Taiwan.
ABSTRACT
(Aim) COVID-19 is an ongoing infectious disease. It has caused more than 107.45 m confirmed cases and 2.35 m deaths till 11/Feb/2021. Traditional computer vision methods have achieved promising results on the automatic smart diagnosis. (Method) This study aims to propose a novel deep learning method that can obtain better performance. We use the pseudo-Zernike moment (PZM), derived from Zernike moment, as the extracted features. Two settings are introducing: (i) image plane over unit circle;and (ii) image plane inside the unit circle. Afterward, we use a deep-stacked sparse autoencoder (DSSAE) as the classifier. Besides, multiple-way data augmentation is chosen to overcome overfitting. The multiple-way data augmentation is based on Gaussian noise, salt-and-pepper noise, speckle noise, horizontal and vertical shear, rotation, Gamma correction, random translation and scaling. (Results) 10 runs of 10-fold cross validation shows that our PZM-DSSAE method achieves a sensitivity of 92.06% +/- 1.54%, a specificity of 92.56% +/- 1.06%, a precision of 92.53% +/- 1.03%, and an accuracy of 92.31% +/- 1.08%. Its F1 score, MCC, and FMI arrive at 92.29% +/- 1.10%, 84.64% +/- 2.15%, and 92.29% +/- 1.10%, respectively. The AUC of our model is 0.9576. (Conclusion) We demonstrate "image plane over unit circle" can get better results than "image plane inside a unit circle." Besides, this proposed PZM-DSSAE model is better than eight state-of-the-art approaches.