StackDPP: Stacking-Based Explainable Classifier for Depression Prediction and Finding the Risk Factors among Clinicians.

Mental health is a major concern for all classes of people, but especially physicians in the present world. A challenging task is to identify the significant risk factors that are responsible for depression among physicians. To address this issue, the study aimed to build a machine learning-based predictive model that will be capable of predicting depression levels and finding associated risk factors. A raw dataset was collected to conduct this study and preprocessed as necessary. Then, the dataset was divided into 10 sub-datasets to determine the best possible set of attributes to predict depression. Seven different classification algorithms, KNN, DT, LGBM, GB, RF, ETC, and StackDPP, were applied to all the sub-datasets. StackDPP is a stacking-based ensemble classifier, which is proposed in this study. It was found that StackDPP outperformed on all the datasets. The findings indicate that the StackDPP with the sub-dataset with all the attributes gained the highest accuracy (0.962581), and the top 20 attributes were enough to gain 0.96129 accuracy by StackDPP, which was close to the performance of the dataset with all the attributes. In addition, risk factors were analyzed in this study to reveal the most significant risk factors that are responsible for depression among physicians. The findings of the study indicate that the proposed model is highly capable of predicting the level of depression, along with finding the most significant risk factors. The study will enable mental health professionals and psychiatrists to decide on treatment and therapy for physicians by analyzing the depression level and finding the most significant risk factors.

Design of a Highly Sensitive Photonic Crystal Fiber Sensor for Sulfuric Acid Detection.

In this research, a photonic crystal fiber (PCF)-based sulfuric acid detector is proposed and investigated to identify the exact concentration of sulfuric acid in a mixture with water. In order to calculate the sensing and propagation characteristics, a finite element method (FEM) based on COMSOL Multiphysics software is employed. The extensive simulation results verified that the proposed optical detector could achieve an ultra-high sensitivity of around 97.8% at optimum structural and operating conditions. Furthermore, the proposed sensor exhibited negligible loss with suitable numerical aperture and single-mode propagation at fixed operating conditions. In addition, the circular air holes in the core and cladding reduce fabrication complexity and can be easily produced using the current technology. Therefore, we strongly believe that the proposed detector will soon find its use in numerous industrial applications.

Discovering Common Pathophysiological Processes between COVID-19 and Cystic Fibrosis by Differential Gene Expression Pattern Analysis.

Coronaviruses are a family of viruses that infect mammals and birds. Coronaviruses cause infections of the respiratory system in humans, which can be minor or fatal. A comparative transcriptomic analysis has been performed to establish essential profiles of the gene expression of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) linked to cystic fibrosis (CF). Transcriptomic studies have been carried out in relation to SARS-CoV-2 since a number of people have been diagnosed with CF. The recognition of differentially expressed genes demonstrated 8 concordant genes shared between the SARS-CoV-2 and CF. Extensive gene ontology analysis and the discovery of pathway enrichment demonstrated SARS-CoV-2 response to CF. The gene ontological terms and pathway enrichment mechanisms derived from this research may affect the production of successful drugs, especially for the people with the following disorder. Identification of TF-miRNA association network reveals the interconnection between TF genes and miRNAs, which may be effective to reveal the other influenced disease that occurs for SARS-CoV-2 to CF. The enrichment of pathways reveals SARS-CoV-2-associated CF mostly engaged with the type of innate immune system, Toll-like receptor signaling pathway, pantothenate and CoA biosynthesis, allograft rejection, graft-versus-host disease, intestinal immune network for IgA production, mineral absorption, autoimmune thyroid disease, legionellosis, viral myocarditis, inflammatory bowel disease (IBD), etc. The drug compound identification demonstrates that the drug targets of IMIQUIMOD and raloxifene are the most significant with the significant hub DEGs.

Applying supervised contrastive learning for the detection of diabetic retinopathy and its severity levels from fundus images.

Diabetic Retinopathy (DR) is a major complication in human eyes among the diabetic patients. Early detection of the DR can save many patients from permanent blindness. Various artificial intelligent based systems have been proposed and they outperform human analysis in accurate detection of the DR. In most of the traditional deep learning models, the cross-entropy is used as a common loss function in a single stage end-to-end training method. However, it has been recently identified that this loss function has some limitations such as poor margin leading to false results, sensitive to noisy data and hyperparameter variations. To overcome these issues, supervised contrastive learning (SCL) has been introduced. In this study, SCL method, a two-stage training method with supervised contrastive loss function was proposed for the first time to the best of authors' knowledge to identify the DR and its severity stages from fundus images (FIs) using "APTOS 2019 Blindness Detection" dataset. "Messidor-2" dataset was also used to conduct experiments for further validating the model's performance. Contrast Limited Adaptive Histogram Equalization (CLAHE) was applied for enhancing the image quality and the pre-trained Xception CNN model was deployed as the encoder with transfer learning. To interpret the SCL of the model, t-SNE method was used to visualize the embedding space (unit hyper sphere) composed of 128 D space into a 2 D space. The proposed model achieved a test accuracy of 98.36%, and AUC score of 98.50% to identify the DR (Binary classification) and a test accuracy of 84.364%, and AUC score of 93.819% for five stages grading with the APTOS 2019 dataset. Other evaluation metrics (precision, recall, F1-score) were also determined with APTOS 2019 as well as with Messidor-2 for analyzing the performance of the proposed model. It was also concluded that the proposed method achieved better performance in detecting the DR compared to the conventional CNN without SCL and other state-of-the-art methods.

Identification of Potential Key Genes and Molecular Mechanisms of Medulloblastoma Based on Integrated Bioinformatics Approach.

BACKGROUND: Medulloblastoma (MB) is the most occurring brain cancer that mostly happens in childhood age. This cancer starts in the cerebellum part of the brain. This study is designed to screen novel and significant biomarkers, which may perform as potential prognostic biomarkers and therapeutic targets in MB. METHODS: A total of 103 MB-related samples from three gene expression profiles of GSE22139, GSE37418, and GSE86574 were downloaded from the Gene Expression Omnibus (GEO). Applying the limma package, all three datasets were analyzed, and 1065 mutual DEGs were identified including 408 overexpressed and 657 underexpressed with the minimum cut-off criteria of ∣log fold change | >1 and P < 0.05. The Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and WikiPathways enrichment analyses were executed to discover the internal functions of the mutual DEGs. The outcomes of enrichment analysis showed that the common DEGs were significantly connected with MB progression and development. The Search Tool for Retrieval of Interacting Genes (STRING) database was used to construct the interaction network, and the network was displayed using the Cytoscape tool and applying connectivity and stress value methods of cytoHubba plugin 35 hub genes were identified from the whole network. RESULTS: Four key clusters were identified using the PEWCC 1.0 method. Additionally, the survival analysis of hub genes was brought out based on clinical information of 612 MB patients. This bioinformatics analysis may help to define the pathogenesis and originate new treatments for MB.

Graphene-MoS2 with TiO2-SiO2 layers based surface plasmon resonance biosensor: Numerical development for formalin detection.

In this article, numerically a surface plasmon resonance (SPR) biosensor is developed based on Graphene-M O S 2 with TiO 2 -SiO 2 hybrid structure for the detection of formalin. Based on attenuated total reflection (ATR) method, we used angular interrogation technique to sense the presence the formalin by observing the change of "minimum reflectance with respect to SPR angle" and "maximum transmittance with respect to surface plasmon resonance frequency (SPRF)". Here, we used Chitosan as probe analyte to perform chemical reaction with formalin (formaldehyde) which is consider as target analyte. Simulation results show a negligible variation of SPRF and SPR angle for improper sensing of formalin that confirms absence of formalin whereas for proper sensing is considerably countable that confirms the presence of formalin. Thereafter, a comparison of sensitivity for different sensor structure is made. It is observed that the sensitivity without TiO2, SiO2, MoS2 and Graphene (conventional structure) is very poor and 73.67% whereas the sensitivity with graphene but without TiO2, SiO2 and MoS2 layers is 74.67% consistently better than the conventional structure. This is due to the electron loss of graphene, which is accompanying with the imaginary dielectric constant. Furthermore, the sensitivity without TiO2, SiO2 and graphene but with MoS2 layer is 79.167%. After more if both graphene and MoS2 are used and TiO2 and SiO2 layers are not used then sensitivity improves to 80.5%. This greater than before performance is due to the absorption ability and optical characteristics of graphene biomolecules and high fluorescence quenching ability of MoS2. Further again, if TiO2-SiO2 composite layer is used with the Graphene-MoS2 then the sensitivity enhances from 80.5% to 82.5%. Finally, the sensitivity for the proposed structure has been carried out, and result is 82.83%, the highest value among all the previous structures to integrate the advantages of graphene, MoS2, TiO2 and SiO2.