Deep Convolutional Neural Network Based Automatic COVID-19 Detection from Chest X-ray Images

To minimize the rate of death from COVID-19 and stop the disease from spreading early detection is vital. The normal RT-PCR tests for COVID-19 detection take a long time to complete. In contrast to this test, Covid-19 can be quickly detected using various machine-learning technologies. Previous studies only had access to smaller datasets, as COVID-19 data was not readily available back then. Since COVID-19 is a dangerous virus, the model needs to be robust and trustworthy, and the model must be trained on a large and diverse dataset. To overcome that problem, this study combines six publicly available Chest X-ray datasets to produce a larger and more diverse balanced dataset with a total of 68,424 images. In this study, we develop a CNN model that primarily entails two steps: (a) feature extraction and (b) classification, which are used to identify COVID-19 positive cases from X-ray images. The accuracy of this proposed model is 97.58%, which is higher than most state-of-the-art models. © 2022 IEEE.

Prediction of COVID-19 Using the Artificial Neural Network (ANN) with K-Fold Cross-Validation

Background: COVID-19 is a disease that attacks the respiratory system and is highly contagious, so cases of the spread of COVID-19 are increasing every day. The increase in COVID-19 cases cannot be predicted accurately, resulting in a shortage of services, facilities and medical personnel. This number will always increase if the community is not vigilant and actively reduces the rate of adding confirmed cases. Therefore, public awareness and vigilance need to be increased by presenting information on predictions of confirmed cases, recovered cases, and cases of death of COVID-19 so that it can be used as a reference for the government in taking and establishing a policy to overcome the spread of COVID-19. Objective: This research predicts COVID-19 in confirmed cases, recovered cases, and death cases in Lampung Province Method: This study uses the ANN method to determine the best network architecture for predicting confirmed cases, recovered cases, and deaths from COVID-19 using the k-fold cross-validation method to measure predictive model performance. Results: The method used has a good predictive ability with an accuracy value of 98.22% for confirmed cases, 98.08% for cured cases, and 99.05% for death cases. Conclusion: The ANN method with k-fold cross-validation to predict confirmed cases, recovered cases, and COVID-19 deaths in Lampung Province decreased from October 27, 2021, to January 24, 2022. © 2023 The Authors. Published by Universitas Airlangga.

COVID-19 and Viral Pneumonia Classification Using Radiomic Features and Deep Learning

The new coronavirus has become the greatest challenge of the 21st century. But since the first cases, much is being discovered about the disease and its effects on the body. Medical imaging, such as X-Rays and CT is widely used to visualize and follow up the patient's clinical picture, especially the effects on the lungs. Although useful, the analysis of this type of image requires some expertise from the radiologist. In less developed countries, the amount of radiologists specialized in chest X-Rays is inadequate, which motivates the development of new technologies to assist clinicians to provide reliable diagnoses. Therefore, this paper addresses the development of a computer-based method to assist in COVID-19 detection among viral pneumonia and health patients through X-Rays images. The proposed method is based on extracting radiomic features and analyzing them using Deep Neural Networks. Experiments following K-Fold Cross-Validation achieved an overall accuracy of 94.98%, a sensibility of 94.89% and an AUC of 99.20%. A benchmark with traditional machine learning algorithms and a binary assessment are also provided. From a multiclass perspective, the analysis and differentiation of COVID-19 and other viral pneumonia reached great results and may assist radiologists in better diagnosing the disease worldwide. © 2022 IEEE.

Class prediction of the prevalent transmission mode of COVID-19 within a geographic area

This research developed a multinomial classification model that predicts the prevalent mode of transmission of the coronavirus from person to person within a geographic area, using data from the World Health Organization (WHO). The WHO defines four transmission modes of the coronavirus disease 2019 (COVID-19);namely, community transmission, pending (unknown), sporadic cases, and clusters of cases. The logistic regression was deployed on the COVID-19 dataset to construct a multinomial model that can predict the prevalent transmission mode of coronavirus within a geographic area. The k-fold cross validation was employed to test predictive accuracy of the model, which yielded 73% accuracy. This model can be adopted by local authorities such as regional, state, local government, and cities, to predict the prevalent transmission mode of the virus within their territories. The outcome of the prediction will determine the appropriate strategies to put in place or re-enforced to curtail further transmission.Copyright © 2023 Inderscience Enterprises Ltd.

Identify Covid-19 (Corona Virus Disease 20219) with Chest X-Ray Photos Using Artificial Neural Network

Identification of lung abnormalities indicated by Covid-19 (Corona Virus Disease 2019) requires thoroughness and accuracy in decision making. Because it is related to the determination of the next follow-up to take appropriate action or treatment for the patient being treated. The study was raised to identify the lungs seen from chest X-rays whether normal or affected by Covid-19. Identification is made consists of several processes, namely pre-processing, characteristic extraction, and identification. Identify by comparing GLCM (Gray-Level Co-occurrence Matrix) and Statistical feature extraction. Pre-processing is used to improve the quality of X-ray photos including in this study, namely resize and grayscale. The characteristic extraction process is used to obtain input data that will be used when used in the identification process. Extraction of features here using two methods, namely: GLCM and Statistics. The GLCM methods used are Contrast, Correlation, Energy, Homogenity. As for statistics the values sought include Mean, Deviation, Skewness, Energy, Entropy, Smoothness. Identification using the Artificial Neural Network Radial Basis Function (ANN-RBF). The latter process sought the accuracy levels of two characteristic extraction methods (GLCM and Statistics) identified with ANN-RBF. The level of accuracy sought by using K-Fold Cross Validation. The data used a total of 50 data, with details of 25 chest X-rays that have been identified as normal and 25 chest X-rays identified as Covid-19. The results showed that the extraction of traits using the GLCM method was better viewed from its accuracy rate when compared to statistical methods, with each accuracy rate for GLCM at 91.63% and Statistics at 87.08%. © 2023 International Journal of Artificial Intelligence.

Delayed hypersensitivity reaction to hyaluronic acid dermal filler following ChAdOx1-S recombinant vaccine against coronavirus: Case report

Case report Background: Delayed hypersensitivity reactions to hyaluronic acid fillers are usually self-limiting and uncommon, and spontaneous resolution is frequent. These are presumably T-lymphocyte- mediated reactions that can be caused by flu-like infections and vaccinations. A delayed hypersensitivity reaction after hyaluronic acid filler following the mRNA vaccine against coronavirus has already been described in the literature. We wish to present a case that followed the ChAdOx1-s recombinant COVID-19 vaccine produced by Oxford/ AstraZeneca. Case report: Female patient, 61 years old, submitted to filling of the nasojugal sulcus and nasolabial fold with 1 ml of cross-linked hyaluronic acid -15 mg/ml and after 5 days filling in the lips with 1 ml of cross-linked hyaluronic acid -12 mg/ml, dermatological office, under aseptic technique and using cannulas. It evolved with ecchymosis on the lips and nasolabial folds, with spontaneous resolution after about a week. Sixteen weeks after the procedure, she received the first dose of the ChAdOx1-s recombinant COVID-19 vaccine, and 11 weeks later, the second dose. After 30 days of the 2nd vaccine dose, asymptomatic nodules appeared distributed in the upper and lower portions of nasolabial folds, in melomentonian grooves, in the supralabial region, in the upper and lower lip in the right and left lateral portions and in the infralabial region in the left lateral portion, coinciding with the topographies of the populated areas. Ultrasonographic evaluation with Doppler confirmed these findings. At the time, she denied fever or previous infectious signs or symptoms. Previously, the patient had already been submitted to the filling of the nasojugal and nasolabial folds with a hyaluronic acid-based filler (1 ml), 29 months before the current procedure, without intercurrences. After the appearance of the nodules, she underwent treatment with prednisone 40mg for 15 days, with total weaning after another 15 days, in addition to the use of levoceritizine 5 mg daily for 20 days, with partial reduction in the size of the nodules. Due to aesthetic complaints on the part of the patient, an intralesional injection of hyaluronidase was scheduled, but it was not performed at the request of the patient herself, who opted for expectant management and clinical treatment. In October 2021, the patient reported that the nodules had involuted and in December 2021 she remained asymptomatic, referring to resorption of the entire filler.

A rare case of central post-stroke pain in a patient with migraine and lacunar infarcts in the medulla oblongata

Studies about headaches associated with acute ischemic stroke in patients suffering from migraine were limited, and therefore we present a clinical case of central post-stroke pain (CPSP) in a 47-year-old woman with migraine and lacunar infarcts in the medulla oblongata and also possible mechanisms of CPSP in patients with migraine. Magnetic resonance imaging of the brain revealed lacunar infarction in the medulla oblongata on the right (vertebral artery basin) and a single focus of gliosis in the parietal lobe on the right. Magnetic resonance angiography of cerebral vessels showed the fetal type of structure of both posterior cerebral arteries. This clinical case is a complex clinical situation of a combination of secondary headaches (post-stroke) in a patient with a primary headache (migraine), which was successfully treated by the combined administration of first-line drugs for the treatment of neuropathic pain in a patient with lacunar infarcts in the medulla oblongata. The treatment of CPSP is a difficult task due to the insufficiently unexplored mechanisms of development, the most effective approaches are those aimed at reducing the increased excitability of neurons.Copyright © 2023, Indian Association of Biomedical Scientists. All rights reserved.

Predicting Airflow from Measures Sensitive to Mid-cord Glottal Gap During the COVID-19 Pandemic.

OBJECTIVES: To determine if trans-laryngeal airflow, important in assessing vocal function in paresis/paralysis and presbylarynges patients with mid-cord glottal gaps, could be predicted by other measures sensitive to mid-cord glottal gap size but with smaller risks of spreading COVID-19, and if any patient factors need consideration. METHODS: Four populations were: unilateral vocal fold paresis/paralysis (UVFP, 148), aging and UVFP (UVFP plus aging, 22), bilateral vocal fold paresis/paralysis without airway obstruction (BVFP, 49), and presbylarynges (66). Five measures were selected from the initial clinic visit: mean airflow from repeated /pi/ syllables, longer of 2 /s/ and 2 /z/ productions, higher of 2 cepstral peak prominence smoothed for vowel /a/ (CPPSa), and Glottal Function Index (GFI). S/Z ratios were computed. Stepwise regression models used 3 measures and 5 patient factors (age, sex, etiology, diagnosis, and potentially impaired power source for voicing) to predict airflow. RESULTS: Log-transformations were required to normalize distributions of airflow and S/Z ratio. The final model revealed age, sex, impaired power source, log-transformed S/Z ratio, and GFI predicted log-transformed airflow (R2 = .275, F[5,278] = 21.1; P < .001). CONCLUSIONS: The amount of variance explained by the model was not high, suggesting adding other predictive variables to the model might increase the variance explained.

What is to Hand: The Fold as Landscape Within Textile Imagination

This article explores the fold and textile imagination within art by using as main case study the author's project Imaginary Landscapes. This work consists of a series of photographs taken during the first COVID-19 lockdown in the UK in 2020 and was motivated by a longing for spaces and places at a time of confinement. It provided an opportunity to work with "material to hand”, pointing to Martin Heidegger and Barbara Bolt's discussion of his theory regarding "handling.” The cloth as arranged or folded allows for light to enhance form whilst suggesting landscapes such as shorelines, mountains, forests, deserts or volcanoes. The discussion refers to Gilles Deleuze's reading of Leibniz, Christine Buci-Gluckmann's observations on the Baroque, and to various theoretical and artistic positions concerning the fold, drapery, and textile imagination within different visual contexts, including Giuliana Bruno's observations on the fold in relation to the screen. Imaginary Landscapes is explored with particular attention to contemporary artists Christo and Jeanne Claude, Christian Boltanski and Angela de la Cruz. The argument concludes that the fold as visual and conceptual process allows us to engage in spatio-temporal relations where the appreciation of materiality through handling/folding informs ideas of movement within and across media.

Platelet-Rich Plasma for Vocal Fold Scar: A Preliminary Report of Concept.

INTRODUCTION: Vocal fold atrophy and scar can lead to loss of normal superficial lamina propria, negatively affecting the vibratory function of the vocal fold. These changes can lead to dysphonia, vocal fatigue, decreased volume, and altered pitch. Treatment options for these conditions are limited. Platelet-rich plasma (PRP) consists of platelets, growth factors, and cytokines derived from the patient's own blood and is believed to activate tissue regeneration. The purpose of this study was to review the technical aspects of collecting PRP and injecting it into the vocal fold injection - based on our initial experience with this procedure. CASE: A patient with vocal fold scar was identified and enrolled in an ongoing prospective clinical trial study of a series of 4 monthly subepithelial vocal fold PRP injections, which was temporarily halted due to the COVID-19 pandemic. Patient underwent a single injection of autologous PRP into the left vocal fold. There were no adverse events during the study period. Subjective improvement in voice was noted at 1 month after injection with subsequent return to baseline over the next 4 months. Videostroboscopy performed on postinjection day 1 and day 7 and demonstrated no concerning exam changes. Compared to the preinjection baseline, the patient-reported voice-handicap index-10 (VHI-10) and voice catastrophization index were similar at 4 months following injection (20 to 20 and 4 to 3, respectively). Independent perceptual analysis of voice showed improvement at 4 months postinjection, compared to baseline consensus auditory-perceptual evaluation of voice 60 to 44. CONCLUSIONS: This preliminary report was part of a prospective trial investigating the use of PRP to treat vocal fold atrophy and scar. This work highlights the technical considerations for injecting PRP into the vocal fold. Planned prospective enrollment in this study will help to validate the safety and efficacy of PRP injections.

Artificial Neural Networks for the Prediction of Monkeypox Outbreak.

While the world is still struggling to recover from the harm caused by the widespread COVID-19 pandemic, the monkeypox virus now poses a new threat of becoming a pandemic. Although it is not as dangerous or infectious as COVID-19, new cases of the disease are nevertheless being reported daily from many countries. In this study, we have used public datasets provided by the European Centre for Disease Prevention and Control for developing a prediction model for the spread of the monkeypox outbreak to and throughout the USA, Germany, the UK, France and Canada. We have used certain effective neural network models for this purpose. The novelty of this study is that a neural network model for a time series monkeypox dataset is developed and compared with LSTM and GRU models using an adaptive moment estimation (ADAM) optimizer. The Levenberg-Marquardt (LM) learning technique is used to develop and validate a single hidden layer artificial neural network (ANN) model. Different ANN model architectures with varying numbers of hidden layer neurons were trained, and the K-fold cross-validation early stopping validation approach was employed to identify the optimum structure with the best generalization potential. In the regression analysis, our ANN model gives a good R-value of almost 99%, the LSTM model gives almost 98% and the GRU model gives almost 98%. These three model fits demonstrated that there was a good agreement between the experimental data and the forecasted values. The results of our experiments show that the ANN model performs better than the other methods on the collected monkeypox dataset in all five countries. To the best of the authors' knowledge, this is the first report that has used ANN, LSTM and GRU to predict a monkeypox outbreak in all five countries.

Methyltransferases of Riboviria.

Many viruses from the realm Riboviria infecting eukaryotic hosts encode protein domains with sequence similarity to S-adenosylmethionine-dependent methyltransferases. These protein domains are thought to be involved in methylation of the 5'-terminal cap structures in virus mRNAs. Some methyltransferase-like domains of Riboviria are homologous to the widespread cellular FtsJ/RrmJ-like methyltransferases involved in modification of cellular RNAs; other methyltransferases, found in a subset of positive-strand RNA viruses, have been assigned to a separate "Sindbis-like" family; and coronavirus-specific Nsp13/14-like methyltransferases appeared to be different from both those classes. The representative structures of proteins from all three groups belong to a specific variety of the Rossmann fold with a seven-stranded ß-sheet, but it was unclear whether this structural similarity extends to the level of conserved sequence signatures. Here I survey methyltransferases in Riboviria and derive a joint sequence alignment model that covers all groups of virus methyltransferases and subsumes the previously defined conserved sequence motifs. Analysis of the spatial structures indicates that two highly conserved residues, a lysine and an aspartate, frequently contact a water molecule, which is located in the enzyme active center next to the methyl group of S-adenosylmethionine cofactor and could play a key role in the catalytic mechanism of the enzyme. Phylogenetic evidence indicates a likely origin of all methyltransferases of Riboviria from cellular RrmJ-like enzymes and their rapid divergence with infrequent horizontal transfer between distantly related viruses.

A Transfer Learning Approach to Detect Face Mask in COVID-19 Pandemic

COVID-19 is tumultuous creating our life so unpredictable. There has no solution of this contagious disease rather than vaccination and prevention. The first and foremost preventative step is using face masks. Face mask can hindrance its droplet from one to another. So this paper has focused the detection of facial mask from image processing using Transfer Learning. For this purpose, total 1376 images have been collected where 690 images of with mask and 686 images of without a mask. Here transfer learning is chosen for the reason of its capability to produce best accurate regardless the limited size of the image dataset. Here, multifarious transfer learning models have been trained to find out the best fitting model. Finally, We have found the VGG16 model with the best accuracy where training accuracy is 98.25% and testing accuracy is 96.38%. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Distinguishing features of fold-switching proteins.

Though many folded proteins assume one stable structure that performs one function, a small-but-increasing number remodel their secondary and tertiary structures and change their functions in response to cellular stimuli. These fold-switching proteins regulate biological processes and are associated with autoimmune dysfunction, severe acute respiratory syndrome coronavirus-2 infection, and more. Despite their biological importance, it is difficult to computationally predict fold switching. With the aim of advancing computational prediction and experimental characterization of fold switchers, this review discusses several features that distinguish fold-switching proteins from their single-fold and intrinsically disordered counterparts. First, the isolated structures of fold switchers are less stable and more heterogeneous than single folders but more stable and less heterogeneous than intrinsically disordered proteins (IDPs). Second, the sequences of single fold, fold switching, and intrinsically disordered proteins can evolve at distinct rates. Third, proteins from these three classes are best predicted using different computational techniques. Finally, late-breaking results suggest that single folders, fold switchers, and IDPs have distinct patterns of residue-residue coevolution. The review closes by discussing high-throughput and medium-throughput experimental approaches that might be used to identify new fold-switching proteins.

Structural characterization of SARS-CoV-2 dimeric ORF9b reveals potential fold-switching trigger mechanism.

The constant emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants indicates the evolution and adaptation of the virus. Enhanced innate immune evasion through increased expression of viral antagonist proteins, including ORF9b, contributes to the improved transmission of the Alpha variant; hence, more attention should be paid to these viral proteins. ORF9b is an accessory protein that suppresses innate immunity via a monomer conformation by binding to Tom70. Here, we solved the dimeric structure of SARS-CoV-2 ORF9b with a long hydrophobic tunnel containing a lipid molecule that is crucial for the dimeric conformation and determined the specific lipid ligands as monoglycerides by conducting a liquid chromatography with tandem mass spectrometry analysis, suggesting an important role in the viral life cycle. Notably, a long intertwined loop accessible for host factor binding was observed in the structure. Eight phosphorylated residues in ORF9b were identified, and residues S50 and S53 were found to contribute to the stabilization of dimeric ORF9b. Additionally, we proposed a model of multifunctional ORF9b with a distinct conformation, suggesting that ORF9b is a fold-switching protein, while both lipids and phosphorylation contribute to the switching. Specifically, the ORF9b monomer interacts with Tom70 to suppress the innate immune response, whereas the ORF9b dimer binds to the membrane involving mature virion assembly. Our results provide a better understanding of the multiple functions of ORF9b.

An efficient lung disease classification from X-ray images using hybrid Mask-RCNN and BiDLSTM

Lung diseases mainly affect the inner lining of the lungs causing complications in breathing, airway obstruction, and exhalation. Identifying lung diseases such as COVID-19, pneumonia, fibrosis, and tuberculosis at the earlier stage is a great challenge due to the availability of insufficient laboratory kits and image modalities. The rapid progression of the lung disease can be easily identified via Chest X-rays and this serves as a major boon for the terminally ill patients admitted to Intensive Care Units (ICU). To enhance the decision-making capability of the clinicians, a novel lung disease prediction framework is proposed using a hybrid bidirectional Long-Short-Term-Memory (BiDLSTM)-Mask Region-Based Convolutional Neural Network (Mask-RCNN) model. The Crystal algorithm is used to optimize the scalability and convergence issues in the Mask-RCNN model by hyperparameter tuning. The long-range dependencies for lung disease prediction are done using the BiDLSTM architecture which is connected to the fully connected layer of the Mask RCNN model. The efficiency of the proposed methodology is evaluated using three publicly accessible lung disease datasets namely the COVID-19 radiography dataset, Tuberculosis (TB) Chest X-ray Database, and National Institute of Health Chest X-ray Dataset which consists of the images of infected lung disease patients. The efficiency of the proposed technique is evaluated using different performance metrics such as Accuracy, Precision, Recall, F-measure, Specificity, confusion matrix, and sensitivity. The high accuracy obtained when comparing the proposed methodology with conventional techniques shows its efficiency of it in improving lung disease diagnosis. Copyright © 2022 Elsevier Ltd

A hybrid CNN-KNN approach for identification of COVID-19 with 5-fold cross validation.

The novel coronavirus is the new member of the SARS family, which can cause mild to severe infection in the lungs and other vital organs like the heart, kidney and liver. For detecting COVID-19 from images, traditional ANN can be employed. This method begins by extracting the features and then feeding the features into a suitable classifier. The classification rate is not so high as feature extraction is dependent on the experimenters' expertise. To solve this drawback, a hybrid CNN-KNN-based model with 5-fold cross-validation is proposed to classify covid-19 or non-covid19 from CT scans of patients. At first, some pre-processing steps like contrast enhancement, median filtering, data augmentation, and image resizing are performed. Secondly, the entire dataset is divided into five equal sections or folds for training and testing. By doing 5-fold cross-validation, the generalization of the dataset is ensured and the overfitting of the network is prevented. The proposed CNN model consists of four convolutional layers, four max-pooling layers, and two fully connected layers combined with 23 layers. The CNN architecture is used as a feature extractor in this case. The features are taken from the CNN model's fourth convolutional layer and finally, the features are classified using K Nearest Neighbor rather than softmax for better accuracy. The proposed method is conducted over an augmented dataset of 4085 CT scan images. The average accuracy, precision, recall and F1 score of the proposed method after performing a 5-fold cross-validation is 98.26%, 99.42%,97.2% and 98.19%, respectively. The proposed method's accuracy is comparable with the existing works described further, where the state of the art and the custom CNN models were used. Hence, this proposed method can diagnose the COVID-19 patients with higher efficiency.

Functional dynamics of SARS-CoV-2 3C-like protease as a member of clan PA.

SARS-CoV-2 3C-like protease (3CLpro), a potential therapeutic target for COVID-19, consists of a chymotrypsin fold and a C-terminal α-helical domain (domain III), the latter of which mediates dimerization required for catalytic activation. To gain further understanding of the functional dynamics of SARS-CoV-2 3CLpro, this review extends the scope to the comparative study of many crystal structures of proteases having the chymotrypsin fold (clan PA of the MEROPS database). First, the close correspondence between the zymogen-enzyme transformation in chymotrypsin and the allosteric dimerization activation in SARS-CoV-2 3CLpro is illustrated. Then, it is shown that the 3C-like proteases of family Coronaviridae (the protease family C30), which are closely related to SARS-CoV-2 3CLpro, have the same homodimeric structure and common activation mechanism via domain III mediated dimerization. The survey extended to order Nidovirales reveals that all 3C-like proteases belonging to Nidovirales have domain III, but with various chain lengths, and 3CLpro of family Mesoniviridae (family C107) has the same homodimeric structure as that of C30, even though they have no sequence similarity. As a reference, monomeric 3C proteases belonging to the more distant family Picornaviridae (family C3) lacking domain III are compared with C30, and it is shown that the 3C proteases are rigid enough to maintain their structures in the active state. Supplementary Information: The online version contains supplementary material available at 10.1007/s12551-022-01020-x.

A Transfer Learning Approach to Detect Face Mask in COVID-19 Pandemic

COVID-19 is tumultuous creating our life so unpredictable. There has no solution of this contagious disease rather than vaccination and prevention. The first and foremost preventative step is using face masks. Face mask can hindrance its droplet from one to another. So this paper has focused the detection of facial mask from image processing using Transfer Learning. For this purpose, total 1376 images have been collected where 690 images of with mask and 686 images of without a mask. Here transfer learning is chosen for the reason of its capability to produce best accurate regardless the limited size of the image dataset. Here, multifarious transfer learning models have been trained to find out the best fitting model. Finally, We have found the VGG16 model with the best accuracy where training accuracy is 98.25% and testing accuracy is 96.38%. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.