Automated COVID-19 detection from X-ray and CT images with stacked ensemble convolutional neural network.

Automatic and rapid screening of COVID-19 from the radiological (X-ray or CT scan) images has become an urgent need in the current pandemic situation of SARS-CoV-2 worldwide. However, accurate and reliable screening of patients is challenging due to the discrepancy between the radiological images of COVID-19 and other viral pneumonia. So, in this paper, we design a new stacked convolutional neural network model for the automatic diagnosis of COVID-19 disease from the chest X-ray and CT images. In the proposed approach, different sub-models have been obtained from the VGG19 and the Xception models during the training. Thereafter, obtained sub-models are stacked together using softmax classifier. The proposed stacked CNN model combines the discriminating power of the different CNN's sub-models and detects COVID-19 from the radiological images. In addition, we collect CT images to build a CT image dataset and also generate an X-ray images dataset by combining X-ray images from the three publicly available data repositories. The proposed stacked CNN model achieves a sensitivity of 97.62% for the multi-class classification of X-ray images into COVID-19, Normal and Pneumonia Classes and 98.31% sensitivity for binary classification of CT images into COVID-19 and no-Finding classes. Our proposed approach shows superiority over the existing methods for the detection of the COVID-19 cases from the X-ray radiological images.

Uncertainty-aware convolutional neural network for COVID-19 X-ray images classification.

Deep learning (DL) has shown great success in the field of medical image analysis. In the wake of the current pandemic situation of SARS-CoV-2, a few pioneering works based on DL have made significant progress in automated screening of COVID-19 disease from the chest X-ray (CXR) images. But these DL models have no inherent way of expressing uncertainty associated with the model's prediction, which is very important in medical image analysis. Therefore, in this paper, we develop an uncertainty-aware convolutional neural network model, named UA-ConvNet, for the automated detection of COVID-19 disease from CXR images, with an estimation of associated uncertainty in the model's predictions. The proposed approach utilizes the EfficientNet-B3 model and Monte Carlo (MC) dropout, where an EfficientNet-B3 model has been fine-tuned on the CXR images. During inference, MC dropout has been applied for M forward passes to obtain the posterior predictive distribution. After that mean and entropy have been calculated on the obtained predictive distribution to get the mean prediction and model uncertainty. The proposed method is evaluated on the three different datasets of chest X-ray images, namely the COVID19CXr, X-ray image, and Kaggle datasets. The proposed UA-ConvNet model achieves a G-mean of 98.02% (with a Confidence Interval (CI) of 97.99-98.07) and sensitivity of 98.15% for the multi-class classification task on the COVID19CXr dataset. For binary classification, the proposed model achieves a G-mean of 99.16% (with a CI of 98.81-99.19) and a sensitivity of 99.30% on the X-ray Image dataset. Our proposed approach shows its superiority over the existing methods for diagnosing the COVID-19 cases from the CXR images.

A novel frameshift deletion in NAGLU causing sanfilipo type III-B in an Indian family.

Mucopolysaccharidoses are group of inherited lysosomal storage disorder. Two siblings of a family manifested behavioral abnormalities; hepatosplenomegaly and hypotonia of infantile onset were found to have a novel homozygous frameshift variation, p.Leu280TrpfsTer19 in NAGLU. This variant was predicted to cause the loss of TIM-barrel and alpha-helical region of NAGLU protein.

A novel mutation in SGSH causing Sanfillipo type 3A Mucopolysaccharidoses in an Indian family.

Mucopolysaccharidoses (MPS) type III also termed as Sanfillipo syndrome, involves defect in enzymes required for degradation of heparan sulphate. We report a clinical case of MPS-III later followed by genetic investigation for MPS-III genes SGSH, NAGLU, HGSNAT and GNS. It allowed us to identify a novel and likely pathogenic variant p. G205R in SGSH. Protein based Inslico prediction and protein modelling suggests aberration of helical structure of SGSH protein and reduced binding affinity for its substrate.

Western Indian Rural Gut Microbial Diversity in Extreme Prakriti Endo-Phenotypes Reveals Signature Microbes.

Heterogeneity amidst healthy individuals at genomic level is being widely acknowledged. This, in turn, is modulated by differential response to environmental cues and treatment regimens, necessitating the need for stratified/personalized therapy. We intend to understand the molecular determinants of Ayurvedic way (ancient Indian system of medicine) of endo-phenotyping individuals into distinct constitution types termed "Prakriti," which forms the basis of personalized treatment. In this study, we explored and analyzed the healthy human gut microbiome structure within three predominant Prakriti groups from a genetically homogenous cohort to discover differentially abundant taxa, using 16S rRNA gene based microbial community profiling. We found Bacteroidetes and Firmicutes as major gut microbial components in varying composition, albeit with similar trend across Prakriti. Multiple species of the core microbiome showed differential abundance within Prakriti types, with gender specific signature taxons. Our study reveals that despite overall uniform composition of gut microbial community, healthy individuals belonging to different Prakriti groups have enrichment of specific bacteria. It highlights the importance of Prakriti based endo-phenotypes to explain the variability amongst healthy individuals in gut microbial flora that have important consequences for an individual's health, disease and treatment.

Metabolic enzyme considerations in cancer therapy.

The clinical application of new antineoplastic drugs has been limited because of low therapeutic index and lack of efficacy in humans. Thus, improvement in efficacy of old and new anticancer drugs has been attempted by manipulating their pharmacokinetic properties. Four inter-related factors, which determine the pharmacokinetic behavior of a drug include absorption, distribution, metabolism and excretion. The drug-metabolizing enzymes have been classified in two major groups: phase I and phase II enzymes. Phase I enzymes comprise the oxidases, dehydrogenases, deaminases, hydrolases. Phase II enzymes include primarily UDP-glucuronosyltransferases (UGTs), glutathionetransferases (GSTs), sulfotransferases (SULTs), N-acetyl transferases (NATs), methyltransferases and aminoacid transferases that conjugate products of phase I reactions and parent compounds with appropriate functional groups to generate more water soluble compounds which are more readily eliminated. The importance of these enzymes in the metabolism of specific drugs varies according to the chemical nature of the drug, Drug metabolism is modulated by factors that change among species and even among individuals in a population. Such factors can be environmental or genetic in origin, and influence how a drug is metabolized and to what extent. An awareness of these variables is invaluable when the safety and efficacy of new anticancer drugs are evaluated (1).