Multi-branch sustainable convolutional neural network for disease classification

Pandemic and natural disasters are growing more often, imposing even more pressure on life care services and users. There are knowledge gaps regarding how to prevent disasters and pandemics. In recent years, after heart disease, corona virus disease-19 (COVID-19), brain stroke, and cancer are at their peak. Different machine learning and deep learning-based techniques are presented to detect these diseases. Existing technique uses two branches that have been used for detection and prediction of disease accurately such as brain hemorrhage. However, existing techniques have been focused on the detection of specific diseases with double-branches convolutional neural networks (CNNs). There is a need to develop a model to detect multiple diseases at the same time using computerized tomography (CT) scan images. We proposed a model that consists of 12 branches of CNN to detect the different types of diseases with their subtypes using CT scan images and classify them more accurately. We proposed multi-branch sustainable CNN model with deep learning architecture trained on the brain CT hemorrhage, COVID-19 lung CT scans and chest CT scans with subtypes of lung cancers. Feature extracted automatically from preprocessed input data and passed to classifiers for classification in the form of concatenated feature vectors. Six classifiers support vector machine (SVM), decision tree (DT), K-nearest neighbor (K-NN), artificial neural network (ANN), naïve Bayes (NB), linear regression (LR) classifiers, and three ensembles the random forest (RF), AdaBoost, gradient boosting ensembles were tested on our model for classification and prediction. Our model achieved the best results on RF on each dataset. Respectively, on brain CT hemorrhage achieved (99.79%) accuracy, on COVID-19 lung CT scans achieved (97.61%), and on chest CT scans dataset achieved (98.77%). © 2023 Wiley Periodicals LLC.

A Comparative Study of COVID-19 CT Image Synthesis using GAN and CycleGAN

Generative adversarial networks (GANs) have been very successful in many applications of medical image synthesis, which hold great clinical value in diagnosis and analysis tasks, especially when data is scarce. This study compares the two most adopted generative modelling algorithms in recent medical image synthesis tasks, namely the traditional Generative Adversarial Network (GAN) and Cycle-consistency Generative Adversarial Network (CycleGAN) for COVID-19 CT image synthesis. Experiments show that very plausible synthetic COVID-19 images with a clear vision of artificially generated ground glass opacity (GGO) can be generated with CycleGAN when trained using an identity loss constant at 0.5. Moreover, it is found that the synthesis of the synthetic GGO features is generalized across images with different chest and lung structures, which suggests that diverse patterns of GGO can be synthesized using a conventional Image-to- Image translation setting without additional auxiliary conditions or visual annotations. In addition, similar experiment setting achieves encouraging perceptual quality with a Fréchet Inception Distance score of 0.347, which outperforms GAN at 0.383 and CycleGAN at 0.380 with an identity loss constant of 0.005. The experiment outcomes postulate a negative correlation between the strength of the identity loss and the significance of the synthetic instances manifested on the generated images, which highlights an interesting research path to improve the quality of generated images without compromising the significance of synthetic instances upon the image translation. © 2022 IEEE.

Role of biological markers and CT severity score in predicting mortality in patients with COVID­19: An observational retrospective study.

COVID-19 pandemic is a continuing ongoing emergency of public concern. Early identification of markers associated with disease severity and mortality can lead to a prompter therapeutic approach. The present study conducted a multivariate analysis of different markers associated with mortality in order to establish their predictive role. Confirmed cases of 697 patients were examined. Demographic data, clinical symptoms and comorbidities were evaluated. Laboratory and imaging severity scores were reviewed. A total of 133 (19.1%) out of 697 patients succumbed during hospitalization. Obesity was the most common comorbidity, followed by hypertension, diabetes, coronary heart disease and chronic kidney disease. Compared with the survivor patients, non-survivors had a higher prevalence of diabetes, chronic kidney disease and coronary heart disease, as well as higher values of laboratory markers such as neutrophil-lymphocyte ratio (NLR), D-dimer, procalcitonin, IL-6 and C Reactive protein (CRP) and respectively high values of imaging severity scores. Multivariate regression analysis showed that high values of the proposed markers and chest computerized tomography (CT) severity imaging score were predictive for in hospital death: NLR [hazard ratio (HR): 3.127 confidence interval (CI) 95: 2.137-4.576]; D-dimer [HR: 6.223 (CI 95:3.809-10.167)]; procalcitonin [HR: 4.414 (CI 95:2.804-6.948)]; IL-6 [HR: 3.344 (CI 95:1.423-7.855)]; CRP [HR:2.997 (CI 95:1.940-4.630)]; and CT severity score [HR: 3.068 (CI 95:1.777-5.299)]. Laboratory markers and imaging severity scores could be used to stratify mortality risk in COVID-19 patients.

Data of atrial arrhythmias in hospitalized COVID-19 and influenza patients.

Atrial arrhythmias (AA) are common in hospitalized COVID-19 patients with limited data on their association with COVID-19 infection, clinical and imaging outcomes. In the related research article using retrospective research data from one quaternary care and five community hospitals, patients aged 18 years and above with positive SARS-CoV-2 polymerase chain reaction test were included. 6927 patients met the inclusion criteria. The data in this article provides demographics, home medications, in-hospital events and COVID-19 treatments, multivariable generalized linear regression regression models using a log link with a Poisson distribution (multi-parameter regression [MPR]) to determine predictors of new-onset AA and mortality in COVID-19 patients, computerized tomography chest scan findings, echocardiographic findings, and International Classification of Diseases-Tenth Revision codes. The clinical outcomes were compared to a propensity-matched cohort of influenza patients. For influenza, data is reported on baseline demographics, comorbid conditions, and in-hospital events. Generalized linear regression models were built for COVID-19 patients using demographic characteristics, comorbid conditions, and presenting labs which were significantly different between the groups, and hypoxia in the emergency room. Statistical analysis was performed using R programming language (version 4, ggplot2 package). Multivariable generalized linear regression model showed that, relative to normal sinus rhythm, history of AA (adjusted relative risk [RR]: 1.38; 95% CI: 1.11-1.71; p = 0.003) and newly-detected AA (adjusted RR: 2.02 95% CI: 1.68-2.43; p < 0.001) were independently associated with higher in-hospital mortality. Age in increments of 10 years, male sex, White race, prior history of coronary artery disease, congestive heart failure, end-stage renal disease, presenting leukocytosis, hypermagnesemia, and hypomagnesemia were found to be independent predictors of new-onset AA in the MPR model. The dataset reported is related to the research article entitled "Incidence, Mortality, and Imaging Outcomes of Atrial Arrhythmias in COVID-19" [Jehangir et al. Incidence, Mortality, and Imaging Outcomes of Atrial Arrhythmias in COVID-19, American Journal of Cardiology] [1].

COVID-19 DETECTION USING CHEST COMPUTED TOMOGRAPHY SCANS on ECUADORIAN PATIENTS WHO LIVE in the HIGHLAND REGION

The early detection of COVID-19 is one of the current challenges in developing effective diagnosis and treatment mechanisms for patients who are at a high risk for community contagion. Computed Tomography (CT) is an essential support for detecting the infection pattern that causes this disease. CT scans provide relevant information on the morphological appearance of the infected parenchymal tissue, known as ground-glass opacities. Artificial Intelligence (AI) can assist in the quick evaluation of CT scans to differentiate COVID-19 findings in suggestive clinical cases. In this context, AI in the form of, Convolutional Neural Networks (CNN), has achieved successful results in the analysis and classification of medical images. A deep CNN architecture is proposed in this study to diagnose COVID-19 based on the classification of Chest Computed Tomography (CCT) images. In this study 8,624 CCTs of Ecuadorian patients affected by COVID-19 in the first quarter of 2021, were examined. The initial review of CCTs was performed by medical experts to discriminate the CCTs against other chronic lung diseases not associated with COVID-19. The CCTs were pre-processed by techniques such as morphological segmentation, erosion, dilation, and adjustment. After training the model reached an overall F1-score of 97%. © 2021 ACM.

An Adaptive Data Processing Framework for Cost-Effective COVID-19 and Pneumonia Detection

Medical imaging modalities have been showing great potentials for faster and efficient disease transmission control and containment. In the paper, we propose a cost-effective COVID-19 and pneumonia detection framework using CT scans acquired from several hospitals. To this end, we incorporate a novel data processing framework that utilizes 3D and 2D CT scans to diversify the trainable inputs in a resource-limited setting. Moreover, we empirically demonstrate the significance of several data processing schemes for our COVID-19 and pneumonia detection network. Experiment results show that our proposed pneumonia detection network is comparable to other pneumonia detection tasks integrated with imaging modalities, with 93% mean AUC and 85.22% mean accuracy scores on generalized datasets. Additionally, our proposed data processing framework can be easily adapted to other applications of CT modality, especially for cost-effective and resource-limited scenarios, such as breast cancer detection, pulmonary nodules diagnosis, etc. © 2021 IEEE

Comparison of Residual Pulmonary Abnormalities 3 Months After Discharge in Patients Who Recovered From COVID-19 of Different Severity.

Background and Objectives: To investigate whether coronavirus disease 2019 (COVID-19) survivors who had different disease severities have different levels of pulmonary sequelae at 3 months post-discharge. Methods: COVID-19 patients discharged from four hospitals 3 months previously, recovered asymptomatic patients from an isolation hotel, and uninfected healthy controls (HCs) from the community were prospectively recruited. Participants were recruited at Wuhan Union Hospital and underwent examinations, including quality-of-life evaluation (St. George Respiratory Questionnaire [SGRQ]), laboratory examination, chest computed tomography (CT) imaging, and pulmonary function tests. Results: A total of 216 participants were recruited, including 95 patients who had recovered from severe/critical COVID-19 (SPs), 51 who had recovered from mild/moderate disease (MPs), 28 who had recovered from asymptomatic disease (APs), and 42 HCs. In total, 154 out of 174 (88.5%) recovered COVID-19 patients tested positive for serum SARS-COV-2 IgG, but only 19 (10.9%) were still positive for IgM. The SGRQ scores were highest in the SPs, while APs had slightly higher SGRQ scores than those of HCs; 85.1% of SPs and 68.0% of MPs still had residual CT abnormalities, mainly ground-glass opacity (GGO) followed by strip-like fibrosis at 3 months after discharge, but the pneumonic lesions were largely absorbed in the recovered SPs or MPs relative to findings in the acute phase. Pulmonary function showed that the frequency of lung diffusion capacity for carbon monoxide abnormalities were comparable in SPs and MPs (47.1 vs. 41.7%), while abnormal total lung capacity (TLC) and residual volume (RV) were more frequent in SPs than in MPs (TLC, 18.8 vs. 8.3%; RV, 11.8 vs. 0%). Conclusions: Pulmonary abnormalities remained after recovery from COVID-19 and were more frequent and conspicuous in SPs at 3 months after discharge.

Cross-sectional analysis of follow-up chest MRI and chest CT scans in patients previously affected by COVID-19.

PURPOSE: The aim of the study was to prospectively evaluate the agreement between chest magnetic resonance imaging (MRI) and computed tomography (CT) and to assess the diagnostic performance of chest MRI relative to that of CT during the follow-up of patients recovered from coronavirus disease 2019. MATERIALS AND METHODS: Fifty-two patients underwent both follow-up chest CT and MRI scans, evaluated for ground-glass opacities (GGOs), consolidation, interlobular septal thickening, fibrosis, pleural indentation, vessel enlargement, bronchiolar ectasia, and changes compared to prior CT scans. DWI/ADC was evaluated for signal abnormalities suspicious for inflammation. Agreement between CT and MRI was assessed with Cohen's k and weighted k. Measures of diagnostic accuracy of MRI were calculated. RESULTS: The agreement between CT and MRI was almost perfect for consolidation (k = 1.00) and change from prior CT (k = 0.857); substantial for predominant pattern (k = 0.764) and interlobular septal thickening (k = 0.734); and poor for GGOs (k = 0.339), fibrosis (k = 0.224), pleural indentation (k = 0.231), and vessel enlargement (k = 0.339). Meanwhile, the sensitivity of MRI was high for GGOs (1.00), interlobular septal thickening (1.00), and consolidation (1.00) but poor for fibrotic changes (0.18), pleural indentation (0.23), and vessel enlargement (0.50) and the specificity was overall high. DWI was positive in 46.0% of cases. CONCLUSIONS: The agreement between MRI and CT was overall good. MRI was very sensitive for GGOs, consolidation and interlobular septal thickening and overall specific for most findings. DWI could be a reputable imaging biomarker of inflammatory activity.

Current state of diagnostic, screening and surveillance testing methods for COVID-19 from an analytical chemistry point of view.

Since December 2019, we have been in the battlefield with a new threat to the humanity known as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). In this review, we describe the four main methods used for diagnosis, screening and/or surveillance of SARS-CoV-2: Real-time reverse transcription polymerase chain reaction (RT-PCR); chest computed tomography (CT); and different complementary alternatives developed in order to obtain rapid results, antigen and antibody detection. All of them compare the highlighting advantages and disadvantages from an analytical point of view. The gold standard method in terms of sensitivity and specificity is the RT-PCR. The different modifications propose to make it more rapid and applicable at point of care (POC) are also presented and discussed. CT images are limited to central hospitals. However, being combined with RT-PCR is the most robust and accurate way to confirm COVID-19 infection. Antibody tests, although unable to provide reliable results on the status of the infection, are suitable for carrying out maximum screening of the population in order to know the immune capacity. More recently, antigen tests, less sensitive than RT-PCR, have been authorized to determine in a quicker way whether the patient is infected at the time of analysis and without the need of specific instruments.

The Role of Chest Computed Tomography in Asymptomatic Patients of Positive Coronavirus Disease 2019: A Case and Literature Review.

Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is an ongoing global health emergency. At present, patients are the primary source of infection. A randomly diagnosed confirmed case of COVID-19 highlights the importance of computerized tomography of thorax in diagnosing asymptomatic patients. In the early phase of COVID-19, routine screenings miss patients who are virus carriers, and tracking travel history is of paramount importance to early detection and isolation of SARS-CoV-2 cases.

Radiological approaches to COVID-19 pneumonia

COVID-19 pneumonia has high mortality rates. The symptoms are undiagnostic, the results of viral nucleic acid detection method (PCR) can delay, so that chest computerized tomography is often key diagnostic test in patients with possible COVID-19 pneumonia. In this review, we discussed the main radiological findings of this infection.