COVID-19 infection localization and severity grading from chest X-ray images

Perioperative Cardiac Research Considerations During the Coronavirus Disease 2019 (COVID-19) Pandemic

Coronavirus disease-19 (COVID-19) has rapidly spread all over the world. It is found that the low sensitivity of reverse transcription-polymerase chain reaction (RT-PCR) examinations during the early stage of COVID-19 disease. Thus, efficient models are desirable for early-stage testing of COVID-19 infected patients. Chest X-ray (CXR) images of COVID-19 infected patients have shown some bilateral changes. In this paper, deep transfer learning and a deep forest-based model are proposed to diagnose COVID-19 infection from CXR images. Initially, features of X-ray images are extracted using the well-known deep transfer learning model (i.e., ResNet101), which does not require tuning many parameters compared to the deep convolutional neural network (CNN). After that, the deep forest model is utilised to predict COVID-19 infected patients. The deep forest is based upon ensemble learning and requires a small number of hyper-parameters. Additionally, the proposed model is trained on a multi-class dataset that contains four different classes as COVID-19 (+), pneumonia, tuberculosis, and healthy patients. The comparisons are drawn among the proposed deep transfer learning and deep forest-based models, the competitive models. The obtained results show that the proposed model effectively diagnoses COVID-19 infection with an accuracy of 99.4%.

Post-mortem molecular investigations of SARS-CoV-2 in an unexpected death of a recent kidney transplant recipient.

The coronavirus disease 2019 (COVID-19) led to a dramatic increase in the number of cases of patients with pneumonia worldwide. In this study, we aimed to develop an AI-assisted multistrategy image enhancement technique for chest X-ray (CXR) images to improve the accuracy of COVID-19 classification. Our new classification strategy consisted of 3 parts. First, the improved U-Net model with a variational encoder segmented the lung region in the CXR images processed by histogram equalization. Second, the residual net (ResNet) model with multidilated-rate convolution layers was used to suppress the bone signals in the 217 lung-only CXR images. A total of 80% of the available data were allocated for training and validation. The other 20% of the remaining data were used for testing. The enhanced CXR images containing only soft tissue information were obtained. Third, the neural network model with a residual cascade was used for the super-resolution reconstruction of low-resolution bone-suppressed CXR images. The training and testing data consisted of 1,200 and 100 CXR images, respectively. To evaluate the new strategy, improved visual geometry group (VGG)-16 and ResNet-18 models were used for the COVID-19 classification task of 2,767 CXR images. The accuracy of the multistrategy enhanced CXR images was verified through comparative experiments with various enhancement images. In terms of quantitative verification, 8-fold cross-validation was performed on the bone suppression model. In terms of evaluating the COVID-19 classification, the CXR images obtained by the improved method were used to train 2 classification models. Compared with other methods, the CXR images obtained based on the proposed model had better performance in the metrics of peak signal-to-noise ratio and root mean square error. The super-resolution CXR images of bone suppression obtained based on the neural network model were also anatomically close to the real CXR images. Compared with the initial CXR images, the classification accuracy rates of the internal and external testing data on the VGG-16 model increased by 5.09% and 12.81%, respectively, while the values increased by 3.51% and 18.20%, respectively, for the ResNet-18 model. The numerical results were better than those of the single-enhancement, double-enhancement, and no-enhancement CXR images. The multistrategy enhanced CXR images can help to classify COVID-19 more accurately than the other existing methods.

Vaccination against coronavirus disease 2019 (COVID-19) is effective in protecting patients from severe COVID-19 infection. Disease flare-up following immunization in children with rheumatic disorders may result in patient reluctance to receive the vaccine. Underlying rheumatic diseases or the use of immunosuppressive drugs may influence the outcomes of COVID-19 vaccination and infection. We aimed to describe outcomes in children with rheumatic diseases following COVID-19 immunization and infection. This retrospective study was performed at two large academic centers in Thailand. During the COVID-19 pandemic, all patients were routinely queried about COVID-19-related conditions. We included patients with rheumatic diseases aged <18 years who received at least one dose of a COVID-19 vaccine or had a history of COVID-19 infection with more than 6 months of recorded follow-up after the last vaccine dose or COVID-19 illness. Demographic information and data on clinical symptoms, disease activity, treatment, outcomes, and COVID-19 vaccination and infection were collected. A total of 479 patients were included. Most (229; 47.81%) patients had juvenile idiopathic arthritis, followed by connective tissue diseases (189; 39.46%), vasculitis syndromes (42; 8.76%), and other rheumatic diseases (19; 3.97%). Approximately 90% of patients received at least one dose of COVID-19 vaccination, and half of the patients had COVID-19 infection. Among patients, 10.72% and 3.27% developed a flare after COVID-19 vaccination and COVID-19 illness, respectively. Flare severity after COVID immunization and infection was mainly mild to moderate. The predictor of flare after COVID-19 vaccination was the use of prednisolone ≥10 mg/day before vaccination (hazard ratio: 2.04, 95% confidence interval: 1.05-3.97, p = 0.037). Inactive disease before receiving the COVID-19 vaccination was a predictor of inactive status after a flare (hazard ratio: 2.95, 95% confidence interval: 1.04-8.40; p = 0.043). Overall, 3.36% and 1.61% of patients experienced a new onset of rheumatic disease after receiving the COVID-19 vaccine and after COVID-19 infection, respectively. The COVID-19 vaccine is recommended for children with rheumatic disease, particularly those who are in stable condition. After COVID-19 vaccination, patients-especially those with active disease before vaccination or those receiving concurrent prednisolone doses of ≥10 mg/day-should be closely monitored.

To the Editor: Chronic subdural hematomas (CSDHs) are encapsulated collections of blood breakdown products and fluid between the dura mater and the arachnoid.1 Their global incidence is estimated to be about 5/100 000/yr in patients younger than 65 yr,2 increasing to 58/100 000/yr in patients over 65 yr.1 Surgery usually allows a good outcome even in elderly. The clinical outcome of CSDH is conditioned by comorbidities including pulmonary diseases. Nowadays, we are facing a new possible condition related to coronavirus disease 2019 (COVID-19) that could affect the outcome. The spread of this infection started in 2019, December 31st when China alerted WHO about cases of unusual amount of interstitial pneumonia (IP) in Wuhan. After 7 d, on January 7th, scientists announced they had identified a new virus named COVID-19 similar to SARS-CoV. Italy is dealing with pandemic spread of COVID-19. As a matter of fact, on 19th March, it became the country with the highest number of confirmed deaths in the world. On March 23rd 2020 Italy was the world's centre of the epidemic with 69.176 confirmed cases, 6.820 deaths, and 21.137 recoveries.3 Clinically COVID-19 could lead to a severe IP and acute respiratory distress syndrome (ARDS), especially in older patients with multiple comorbidities.4 No studies about surgery in COVID-19 patients have been reported so far. We aim to describe our surgical experience with CSDH in a COVID-19 neurosurgical center. We admitted 5 patients with CSDH to the Neurosurgical Department of Brescia University Hospital between February 21, 2020 (first COVID-19 case reported in Italy) and March 23, 2020. The patients were positive for COVID-19 real-time polymerase chain reaction obtained by nasopharyngeal swab.5,6 We gathered the following data for each patient: age and sex, Cumulative Illness Rating Scale,7 antithrombotic therapy and perioperative mortality (within 30 d). All the patients had signed a consent form approved by the Local Ethic Committee regarding the use of their clinical data. We compared the mortality rate of this new cohort with the data extracted from our pre-COVID-19 CSDH case series. All the patients underwent pre-operative chest X-Ray (CXR) or computed tomography (CT) and routine blood test. An anesthesiological evaluation was performed for each patient. We cohortized our neurosurgical ward to separate COVID-19 positive from negative cases. COVID-19 positive patients started anti-retroviral therapy with lopinavir/ritonavir and hydroxychloroquine. Daily arterial blood gas analysis and CXR were performed. Surgery or endovascular treatment were indicated when the CSDH was symptomatic (presence of focal neurological deficits or mental status changes) and the maximum thickness was greater than 1 cm (Figures 1-4). One case did not present severe neurological impairment and was treated conservatively (Figure 5). No preoperative respiratory impairment was observed.FIGURE 1.: A and B, Head CT scan, axial view: pre- and postoperative CDSH. C, CXR showing bilateral and diffuse IP.FIGURE 2.: A, Head CT scan, axial view: bilateral CSDH. B, CXR showing bilateral and diffuse IP. C and D, MMA embolization procedure.FIGURE 3.: A, Head CT scan, axial view: preoperative CSDH. B, Head CT scan, axial view: postoperative CSDH with signs of recent rebleeding. C, CXR showing bilateral and IP. D, Chest CT scan, axial view, showing ground glass opacity.FIGURE 4.: A, Head CT scan, axial view: preoperative left CSDH. B, Head CT scan, axial view: postoperative CSDH with signs of recent rebleeding. C, CXR showing bilateral IP.FIGURE 5.: A, Head CT scan, axial view: left CSDH. B, normal CXR.We treated all the CSDH patients under general anesthesia. Three patients were operated (2 through craniotomy and 1 with a burr hole). Endovascular occlusion of the middle meningeal artery (MMA) was performed in 1 case (Figure 2). In the surgical cases we placed a subdural drainage for 48 h. After its removal patients underwent a CT head scan (Figures 1, 3, and 4). Afterwards the patients started low-molecular-weight heparin at prophylactic dose before mobilization. The cases are reported in the supplementary material section (Supplemental Digital Content). CSDHs mostly occur in elderly, with an average age of 63.1 Head trauma is the major risk factor, identified in less than 50% (often minor head trauma).8 Other risk factors are described like alcohol abuse, seizures, CSF shunts and coagulopathies, including therapeutic anticoagulation. CSDHs are bilateral in 20% to 25% of cases.9 At present there is no scientific experience regarding surgical outcome of COVID-19 patients. We compared the recent cohort of COVID-19 CSDH patients with our historical series. Between May 2018 and September 2019, we operated 142 patients for CSDH and we observe 5 death at our Institution. Our mortality rate was 3.7% according to the literature.10 In our recent experience, we observed 4 death in 5 COVID-19 patients suffering from CSDH. Therefore, we observed a mortality rate of 80% about 21,6 times greater than our control data. HEMATOLOGICAL DISORDERS We observed 2 cases of mild thrombocytopenia (Table). In these cases, we observed a re-bleeding (Figures 3 and 4) that led from a rapid neurological worsening to death within 5 d. This condition could be related to COVID-19 infection, as reported in literature.11 Independently from the hemorrhagic risk, Lippi et al12 showed that low platelet count is associated with increased risk of severe disease and mortality in patients with COVID-19, and this should serve as clinical indicator of worsening illness during hospitalization.12 In our series, we suppose that thrombocytopenia led to re-bleeding and was associated with poor outcome. TABLE. - Clinical Features of our Cohort Sex Age CIRS AT Side COVID-19 IP TP* Surgery Time to death(d) M 82 13 Yes Left + + No Burr-hole 14 M 86 18 No Bilateral + + No MMA embolization 10 M 77 20 No Right + + Yes Craniotomy 5 M 85 22 Yes Left + + Yes Burr-hole 5 M 78 19 Yes Left + – No NO Alive *TP (Thrombocytopenia): 100.000<PLTs<150.000All patients developed IP after surgery. Thrombocytopenia was observed in 2 cases (40%). The patients suffered rebleeding and showed a shorter time to death. CIRS (Cumulative Illness Rating Scale), AT (antithrombotic drugs), IP (Interstitial Pneumonia), COVID-19 (Coronavirus Disease 2019). INTERSTITIAL PNEUMONIA Dyspnoea and fatigue occurred in all our patients within 48 hours from surgery. CXR revealed severe IP (Figures 1-4). Ground glass opacity and bilateral patchy shadowing were observed on the chest CT (Figure 3).5,13 All our patients had COVID-19 infection without pre-operative respiratory symptoms. Immune system could be impaired following surgical procedures and this may have unmasked a subclinical infection.14,15 On the other hand no respiratory failure was observed in the conservative-treated case (Figure 5). In our experience IP occurred in all cases after treatment and worsened the outcome. This evenience represents a possible complication, often fatal, of COVID-19 infection.5 HEALTH EMERGENCY Nowadays, Italian health care system is the most afflicted in the pandemic scenario.16 The overload of the intensive care units in Lombardy, despite the efforts made, has necessarily influenced the resuscitation possibilities of elderly patients.17 Therefore, we cannot exclude that our results could be affected by the health emergency status. Seung et al. reported good bleeding control from CSDH membrane with MMA embolization, prevention of further growth of hematoma and even spontaneous resolution without surgery.18 Anyway our endovascular treated patient showed the same poor outcome of the surgical cases. Although statistical analysis of these preliminary data is not possible, COVID-19 patients appear to suffer from a negative surgical outcome. It is our belief that conservative treatment should be preferred whenever surgery could be postponed. Disclosures The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.

The research community has recently shown significant interest in designing automated systems to detect coronavirus disease 2019 (COVID-19) using deep learning approaches and chest radiography images. However, state-of-the-art deep learning techniques, especially convolutional neural networks (CNNs), demand more learnable parameters and memory. Therefore, they may not be suitable for real-time diagnosis. Thus, the design of a lightweight CNN model for fast and accurate COVID-19 detection is an urgent need. In this paper, a lightweight CNN model called LW-CORONet is proposed that comprises a sequence of convolution, rectified linear unit (ReLU), and pooling layers followed by two fully connected layers. The proposed model facilitates extracting meaningful features from the chest X-ray (CXR) images with only five learnable layers. The proposed model is evaluated using two larger CXR datasets (Dataset-1: 2250 images and Dataset-2: 15,999 images) and the classification accuracy obtained are 98.67% and 99.00% on Dataset-1 and 95.67% and 96.25% on Dataset-2 for multi-class and binary classification cases, respectively. The results are compared with four contemporary pre-trained CNN models as well as state-of-the-art models. The effect of several hyperparameters: different optimization techniques, batch size, and learning rate have also been investigated. The proposed model demands fewer parameters and requires less memory space. Hence, it is effective for COVID-19 detection and can be utilized as a supplementary tool to assist radiologists in their diagnosis.

ObjectivesIn the midst of the coronavirus disease 2019 (COVID-19) outbreak, chest X-ray (CXR) imaging is playing an important role in diagnosis and monitoring of patients with COVID-19. We propose a deep learning model for detection of COVID-19 from CXRs, as well as a tool for retrieving similar patients according to the model’s results on their CXRs. For training and evaluating our model, we collected CXRs from inpatients hospitalized in four different hospitals.MethodsIn this retrospective study, 1384 frontal CXRs, of COVID-19 confirmed patients imaged between March and August 2020, and 1024 matching CXRs of non-COVID patients imaged before the pandemic, were collected and used to build a deep learning classifier for detecting patients positive for COVID-19. The classifier consists of an ensemble of pre-trained deep neural networks (DNNS), specifically, ReNet34, ReNet50¸ ReNet152, and vgg16, and is enhanced by data augmentation and lung segmentation. We further implemented a nearest-neighbors algorithm that uses DNN-based image embeddings to retrieve the images most similar to a given image.ResultsOur model achieved accuracy of 90.3%, (95% CI: 86.3–93.7%) specificity of 90% (95% CI: 84.3–94%), and sensitivity of 90.5% (95% CI: 85–94%) on a test dataset comprising 15% (350/2326) of the original images. The AUC of the ROC curve is 0.96 (95% CI: 0.93–0.97).ConclusionWe provide deep learning models, trained and evaluated on CXRs that can assist medical efforts and reduce medical staff workload in handling COVID-19.Key Points• A machine learning model was able to detect chest X-ray (CXR) images of patients tested positive for COVID-19 with accuracy and detection rate above 90%.• A tool was created for finding existing CXR images with imaging characteristics most similar to a given CXR, according to the model’s image embeddings.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00330-021-08050-1.

Although the range of immune responses to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is variable, cytokine storm is observed in a subset of symptomatic individuals. To further understand the disease pathogenesis and, consequently, to develop an additional tool for clinicians to evaluate patients for presumptive intervention, we sought to compare plasma cytokine levels between a range of donor and patient samples grouped by a COVID-19 Severity Score (CSS) based on the need for hospitalization and oxygen requirement. Here we utilize a mutual information algorithm that classifies the information gain for CSS prediction provided by cytokine expression levels and clinical variables. Using this methodology, we found that a small number of clinical and cytokine expression variables are predictive of presenting COVID-19 disease severity, raising questions about the mechanism by which COVID-19 creates severe illness. The variables that were the most predictive of CSS included clinical variables such as age and abnormal chest x-ray as well as cytokines such as macrophage colony-stimulating factor, interferon-inducible protein 10, and interleukin-1 receptor antagonist. Our results suggest that SARS-CoV-2 infection causes a plethora of changes in cytokine profiles and that particularly in severely ill patients, these changes are consistent with the presence of macrophage activation syndrome and could furthermore be used as a biomarker to predict disease severity.

Multisystem inflammatory syndrome in children associated with SARS-CoV-2 in a solid organ transplant recipient.

Coronavirus disease 2019 (COVID-19) has been spreading since late 2019, leading the world into a serious health crisis. To control the spread rate of infection, identifying patients accurately and quickly is the most crucial step. Computed tomography (CT) images of the chest are an important basis for diagnosing COVID-19. They also allow doctors to understand the details of the lung infection. However, manual segmentation of infected areas in CT images is time-consuming and laborious. With its excellent feature extraction capabilities, deep learning-based method has been widely used for automatic lesion segmentation of COVID-19 CT images. But, the segmentation accuracy of these methods is still limited. To effectively quantify the severity of lung infections, we propose a Sobel operator combined with Multi-Attention networks for COVID-19 lesion segmentation (SMA-Net). In our SMA-Net, an edge feature fusion module uses Sobel operator to add edge detail information to the input image. To guide the network to focus on key regions, the SMA-Net introduces a self-attentive channel attention mechanism and a spatial linear attention mechanism. In addition, Tversky loss function is adopted for the segmentation network for small size of lesions. Comparative experiments on COVID-19 public datasets show that the average Dice similarity coefficient (DSC) and joint intersection over Union (IOU) of proposed SMA-Net are 86.1% and 77.8%, respectively, which are better than most existing neural networks used for COVID-19 lesion segmentation.

COVID-19 infection in patients with severe aplastic anaemia

Electrocardiographic features of patients with COVID-19: One year of unexpected manifestations

Plasma exchange and COVID 19

The coronavirus disease 2019 (COVID-19) pandemic constitutes a global health emergency. Currently, there are no completely effective therapeutic medications for the management of this outbreak. The cytokine storm is a hyperinflammatory medical condition due to excessive and uncontrolled release of pro-inflammatory cytokines in patients suffering from severe COVID-19, leading to the development of acute respiratory distress syndrome (ARDS) and multiple organ dysfunction syndrome (MODS) and even mortality. Understanding the pathophysiology of COVID-19 can be helpful for the treatment of patients. Evidence suggests that the levels of tumor necrosis factor alpha (TNF-α) and interleukin (IL)-1 and IL-6 are dramatically different between mild and severe patients, so they may be important contributors to the cytokine storm. Several serum markers can be predictors for the cytokine storm. This review discusses the cytokines involved in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, focusing on interferons (IFNs) and ILs, and whether they can be used in COVID-19 treatment. Moreover, we highlight several microRNAs that are involved in these cytokines and their role in the cytokine storm caused by COVID-19.