Cardiovascular disease (CVD) is a leading cause of death worldwide, with millions dying each year. The identification and early diagnosis of CVD are critical in preventing adverse health outcomes. Hence, this study proposes a hybrid deep learning (DL) model that combines a convolutional neural network (CNN) and long short-term memory (LSTM) to identify CVD from the clinical data. This study utilizes CNN to extract the relevant features from the input data and the LSTM network to process sequential data and capture dependencies and patterns over time. This study provides insights into the potential of a hybrid DL model combined with feature engineering and explainable AI to improve the accuracy and interpretability of CVD prediction. We evaluated our model on a publicly available dataset where the proposed CNN-LSTM achieved a high accuracy of 73.52% and 74.15% with and without feature engineering, respectively, in identifying individuals with CVD, which is the best result compared to the current state-of-the-art model. The results of this study demonstrate the potential of DL models for the early diagnosis of CVD. Our proposed CNN-LSTM model also incorporates explainable AI to identify the top features responsible for CVD. They could be used to develop more effective screening tools in clinical practice.
Hyperspectral imaging (HSI) is a potent technique for capturing detailed spectral data across an extensive range of electromagnetic wavelengths with high precision. Furthermore, it is utilized in various domains like agriculture, surveillance of the environment, mineral investigation, medical studies, and defense due to its wide range of applications. It provides valuable insights by capturing and analyzing spectral information beyond the capabilities of traditional imaging techniques. However, storing, analyzing, and interpreting HSI data because of its great dimensionality presents difficulties. Therefore, specialized techniques are required to extract meaningful information from the vast amount of spectral data. In this paper, a segmentation-based approach for feature reduction in HSI data using Factor Analysis (FA) and feature selection techniques has been proposed. The Hyperspectral data was segmented based on spectral characteristics. Furthermore, FA was applied to extract the most significant spectral bands. Additionally, the extracted features were integrated, and feature selection methods based on Correlation, Mutual Information, and Maximum Relevance — Minimum Redundancy(mRMR) were applied. The proposed method aims to reduce dimensionality while preserving relevant information for accurate analysis and interpretation. The experiments on an Indian Pines dataset demonstrate that the spectral segmentation approach with mRMR-based feature selection achieves better classification results. The proposed approach surpasses traditional methods by requiring reduced computational time and memory. It enhances HSI data analysis through increased efficiency and accuracy, as evidenced by the improved classification results visualized in confusion matrices. This method marks a significant advancement in the field by optimizing both computational time and accuracy reducing memory requirements.
Guillain-Barré syndrome (GBS) is one of the most prominent and acute immune-mediated peripheral neuropathy, while autism spectrum disorders (ASD) are a group of heterogeneous neurodevelopmental disorders. The complete mechanism regarding the neuropathophysiology of these disorders is still ambiguous. Even after recent breakthroughs in molecular biology, the link between GBS and ASD remains a mystery. Therefore, we have implemented well-established bioinformatic techniques to identify potential biomarkers and drug candidates for GBS and ASD. 17 common differentially expressed genes (DEGs) were identified for these two disorders, which later guided the rest of the research. Common genes identified the protein-protein interaction (PPI) network and pathways associated with both disorders. Based on the PPI network, the constructed hub gene and module analysis network determined two common DEGs, namely CXCL9 and CXCL10, which are vital in predicting the top drug candidates. Furthermore, coregulatory networks of TF-gene and TF-miRNA were built to detect the regulatory biomolecules. Among drug candidates, imatinib had the highest docking and MM-GBSA score with the well-known chemokine receptor CXCR3 and remained stable during the 100 ns molecular dynamics simulation validated by the principal component analysis and the dynamic cross-correlation map. This study predicted the gene-based disease network for GBS and ASD and suggested prospective drug candidates. However, more in-depth research is required for clinical validation. Communicated by Ramaswamy H. Sarma 17 common differentially expressed genes (DEGs) were identified from 693 DEGs of the GBS dataset (GSE72748) and 365 DEGs of the ASD dataset (GSE113834), which is the preliminary part of this investigation.From the PPI network analysis, a total of 10 hub genes were identified and two common DEGs named CXCL9 and CXCL10 were found in both the hub gene and essential module analysis.The identified leading pathways and GO pathways, TF-gene interaction, and TF-miRNAs network has made the process more relevant and appropriate for suggesting probable drug candidates.Among the drug candidates, imatinib was suggested as the main drug candidate due to its interaction with the hub gene CXCL9 and CXCL10 and lower p value than the other candidates. It showed the highest binding affinity score and remained stable with the CXCR3 chemokine receptor. 17 common differentially expressed genes (DEGs) were identified from 693 DEGs of the GBS dataset (GSE72748) and 365 DEGs of the ASD dataset (GSE113834), which is the preliminary part of this investigation. From the PPI network analysis, a total of 10 hub genes were identified and two common DEGs named CXCL9 and CXCL10 were found in both the hub gene and essential module analysis. The identified leading pathways and GO pathways, TF-gene interaction, and TF-miRNAs network has made the process more relevant and appropriate for suggesting probable drug candidates. Among the drug candidates, imatinib was suggested as the main drug candidate due to its interaction with the hub gene CXCL9 and CXCL10 and lower p value than the other candidates. It showed the highest binding affinity score and remained stable with the CXCR3 chemokine receptor.
Introduction: Brain metastasis represents the most common form of intracranial tumor and causes significant morbidity and mortality in cancer patients. This prospective study was carried out at Radiation Oncology Department, Combined Military Hospital Dhaka from January 2010 to December 2012. The concurrent chemotherapy and radiotherapy have shown better outcome and improved the quality of life. Objectives: To evaluate the efficacy and toxicity of concurrent treatment with whole brain radiotherapy (WBRT) and temozolomide (TMZ) in patients with brain metastasis. Methods: Sixty patients with multiple brain metastases were enrolled and received WBRT with 30 Gray (Gy) in ten fractions with concurrent TMZ (75mg/m2/day) for ten days. Results: Remarkable symptomatic relief occurred in eighteen (72%) patients of headache, nine (60%) patients of altered mental status, thirteen (76.5%) of vomiting, ten (71.4%) of seizure, eleven (68.7%) of altered sensation, seven (58.3%) of focal weakness and two (40%) of visual change. In relation to objective response four (6.7%) patients had complete response, twenty three (38.3%) patients had partial response while twenty one (35%) had stable disease and twelve (20%) had progressive disease. The overall response rate was 45%. The most frequent toxicities included anorexia in twenty one (35%), nausea in eighteen (30%), vomiting in ten (16.6%), lethargy in seventeen (28%), anemia in eight (13.3%) and neutropenia in thirteen (21.6%) cases. 64 JAFMC Bangladesh. Vol 10, No 1 (June) 2014 Conclusion: The concurrent treatment with whole brain radiotherapy (WBRT) and temozolomide (TMZ) in patients with brain metastasis is well tolerated with an encouraging response. DOI: http://dx.doi.org/10.3329/jafmc.v10i1.22927 Journal of Armed Forces Medical College Bangladesh Vol.10(1) 2014
Here, we develop a framework for the prediction and screening of native defects and functional impurities in a chemical space of Group IV, III-V, and II-VI zinc blende (ZB) semiconductors, powered by crystal Graph-based Neural Networks (GNNs) trained on high-throughput density functional theory (DFT) data. Using an innovative approach of sampling partially optimized defect configurations from DFT calculations, we generate one of the largest computational defect datasets to date, containing many types of vacancies, self-interstitials, anti-site substitutions, impurity interstitials and substitutions, as well as some defect complexes. We applied three types of established GNN techniques, namely Crystal Graph Convolutional Neural Network (CGCNN), Materials Graph Network (MEGNET), and Atomistic Line Graph Neural Network (ALIGNN), to rigorously train models for predicting defect formation energy (DFE) in multiple charge states and chemical potential conditions. We find that ALIGNN yields the best DFE predictions with root mean square errors around 0.3 eV, which represents a prediction accuracy of 98 % given the range of values within the dataset, improving significantly on the state-of-the-art. Models are tested for different defect types as well as for defect charge transition levels. We further show that GNN-based defective structure optimization can take us close to DFT-optimized geometries at a fraction of the cost of full DFT. DFT-GNN models enable prediction and screening across thousands of hypothetical defects based on both unoptimized and partially-optimized defective structures, helping identify electronically active defects in technologically-important semiconductors.
This research focuses on utilizing EEG brainwave data for the crucial task of detecting driver drowsiness a significant concern for road safety. We carefully curated the "Sleepy Driver EEG Brainwave Data" set, excluding less reliable metrics. Employing an ensemble approach, our robust classification model integrates Logistic Regression, K-Nearest Neighbors, Decision Tree, and Random Forest algorithms. The ensemble significantly improved prediction accuracy during real tests. The model demonstrated effectiveness in discerning between awake and asleep states, with rigorous hyper-parameter tuning identifying the optimal Random-Forest classifier. This study highlights the potential of EEG signal analysis and machine learning in establishing a dependable system for driver drowsiness detection. Beyond promising a substantial impact on road safety, our findings advocate for life-saving interventions and encourage safer driving practices, contributing to enhanced public well-being.
Mass quantity of population, emission from conventional power plant, and more demand than supply i.e. load shedding encouraged us to search alternative way to fulfill the energy crisis of humankind and reduce greenhouse gas emission. In this paper, we just study the feasibility of energy from outdoor fan of air conditioning system using small wind turbine (rotor radius 0.25-5m), because nowadays air conditioning is the second largest energy consumption device among commercial as well as home appliances in Bangladesh due to climates fluctuation. Moreover, the extracted energy will store in electrochemical device to use this energy during load shedding. It not only helps to reduce tariff but also remove power cut-off tension and replace high price heavy and solar irradiance dependent on-grid rooftop solar system. We measured the outdoor fan's wind speed using anemometer of various capacities such as 1 to 2 tons' air conditioners. Later the model's performance analyzes numerically by MATLAB software. It shows maximum mechanical power 40W at trip speed ratio 8. In addition, cost analysis has done to know economic feasibility.
With high inflammatory states from both COVID-19 and HIV conditions further result in complications. The ongoing confrontation between these two viral infections can be avoided by adopting suitable management measures.The aim of this study was to figure out the pharmacological mechanism behind apigenin's role in the synergetic effects of COVID-19 to the progression of HIV patients.We employed computer-aided methods to uncover similar biological targets and signaling pathways associated with COVID-19 and HIV, along with bioinformatics and network pharmacology techniques to assess the synergetic effects of apigenin on COVID-19 to the progression of HIV, as well as pharmacokinetics analysis to examine apigenin's safety in the human body.Stress-responsive, membrane receptor, and induction pathways were mostly involved in gene ontology (GO) pathways, whereas apoptosis and inflammatory pathways were significantly associated in the Kyoto encyclopedia of genes and genomes (KEGG). The top 20 hub genes were detected utilizing the shortest path ranked by degree method and protein-protein interaction (PPI), as well as molecular docking and molecular dynamics simulation were performed, revealing apigenin's strong interaction with hub proteins (MAPK3, RELA, MAPK1, EP300, and AKT1). Moreover, the pharmacokinetic features of apigenin revealed that it is an effective therapeutic agent with minimal adverse effects, for instance, hepatoxicity.Synergetic effects of COVID-19 on the progression of HIV may still be a danger to global public health. Consequently, advanced solutions are required to give valid information regarding apigenin as a suitable therapeutic agent for the management of COVID-19 and HIV synergetic effects. However, the findings have yet to be confirmed in patients, suggesting more in vitro and in vivo studies.
