Browse Articles

Enhanced melanoma and non-melanoma skin cancer classification using a hybrid LSTM-CNN model

Abstract Melanoma is the most dangerous type of skin cancer. Although it accounts for only about 1% of all skin cancer cases, it is responsible for the majority of skin cancer-related deaths. Early detection and accurate diagnosis are crucial for improving the prognosis and survival rates of patients with melanoma. This paper presents a novel approach for the automatic identification of cutaneous lesions by integrating convolutional neural networks (CNNs) with long short-term memory (LSTM) networks. In the proposed approach, the image of each skin lesion is divided into a sequence of tags of a particular size, which is then treated by the LSTM network to capture temporal dependence and relevant relationships between different spatial regions. This patching sequence allows the modeling system to analyze the local pattern in the image. Time CNN layers are later used to extract spatial functions, such as texture, edges, and color variation, on each patch. A Softmax layer is then used for classification, providing a probability distribution over the possible classes. We use the HAM10000 dataset, which contains 10,015 skin lesion images. Experimental results demonstrate that the proposed method outperforms recent models in several metrics, including accuracy, recall, precision, F1 score, and ROC curve performance.

Putrescine-enhanced essential oil and phytochemicals in Melissa officinalis L. : a sustainable approach

Exploring the anti-inflammatory mechanism of geniposide in rheumatoid arthritis via network pharmacology and experimental validation

A Multi-Objective optimization framework for the sustainable machining of Monel 400

Identifying biomarkers for papilledema and pseudopapilledema

Abstract Papilledema describes optic nerve head (ONH) swelling due to raised intracranial pressure which can lead to complications including visual loss. Differentiating it from pseudopapilledema is often challenging and may require invasive investigations. Optical Coherence Tomography (OCT) is a non-invasive imaging modality, allowing visualisation of the ONH. We identified OCT-derived biomarkers for papilledema and pseudopapilledema in the existing literature. 213 patients with confirmed papilledema, optic disc drusen (ODD), tilted optic discs (TOD) or crowded optic discs (COD) were included. OCT scans were analysed for qualitative and quantitative biomarkers, with sensitivity, specificity, and AUC calculated for individual and combined biomarkers. Logistic regression modelling including qualitative biomarkers to differentiate papilledema from ODD and COD demonstrated a sensitivity and specificity of 89% and AUC of 0.96. Inclusion of TOD reduced the sensitivity of the model to 66%. In a model differentiating papilledema from TOD, the best-performing biomarker achieved a sensitivity of 87%, specificity of 61% and AUC of 0.83. Using qualitative biomarkers, we identified a model with high sensitivity and specificity to differentiate papilledema from ODD and COD. Quantitative biomarkers displayed high AUCs for differentiating TOD from papilledema. Our findings show that OCT demonstrates promising utility in differentiating papilledema from pseudopapilledema.

Structural and functional characterization of a porcine intestinal microbial ecosystem developed in vitro

Abstract The mammalian digestive tract harbors a vast microbial community that has the potential to modulate numerous health-related processes. Multicompartment dynamic gut models have been developed to study microbial communities in a controlled environment. To verify the assumption that the experimental results produced in vitro in a mechanical device would be highly similar to those obtained from an in vivo study, in this study fecal samples from four pigs were inoculated in a simulator of the porcine intestinal microbial ecosystem (SPIME) and cultured until reaching steady state. The composition and structure of the resultant microbial communities, and the metabolites produced were compared with those harvested from the intestine of the same pigs. Taxonomic abundance identification based on shallow shotgun metagenomic sequencing revealed only 12.1% of species or 15% of metagenome-assembled genomes (MAGs) being shared across the colon compartments of the source pigs and the SPIME. Despite these overwhelming compositional shifts, higher functional conservation was indicated as measured by functional richness, MAG-level traits, CAZymes, and untargeted metabolomics. Environmental selection and bacterial functional redundancy were considered the two key elements in microbial compositional shifts and functional preservation.

Human umbilical vein endothelial cell miRNA secretome highlights endothelial origin of serum miRNAs

Abstract MicroRNAs are key contributors to blood-based biomarker research, however their potential is hindered by the “noise” of their abundance even in healthy blood. Using HUVEC cultures and their conditioned media as a model for endothelium and blood, we were able to detect 574 different microRNAs. 166 of these were exclusively secreted, 155 only intracellular and 253 were found in both states suggesting a highly ordered role in endocrine and paracrine communication. The identified microRNA signatures exhibited higher degrees of variability based on culture conditions rather than genetic background of donors. We found that the endothelial secreted microRNA signature correlates greatly with those found in blood serum (ρ = 0.749 ± 0.044), more so, than leukocyte secretory microRNAs (ρ = 0.531 ± 0.044). These results demonstrate that the endothelium actively secretes microRNAs dominating the microRNA composition of the blood making the endothelial secretory microRNA signatures ideal representation of background “noise” of healthy serum samples. These microRNA signatures are readily adapted to environmental cues; making the standardisation of culture conditions a key concern. Our results also demonstrate that the microarray technology has great use in precision microRNA biomarker discovery using simple models, which should be utilised for further mapping of cell-type specific healthy signatures to further refine blood-based diagnostics.

Hierarchical deep Q-network-based optimization of resilient grids under multi-dimensional uncertainties from extreme weather

Expression analysis of plasma extracellular vesicle associated candidate MiRNAs in endometriosis using integrative bioinformatics and experiential data

Microglial exosome TREM2 ameliorates ferroptosis and neuroinflammation in alzheimer’s disease by activating the Wnt/β-catenin signaling

Integrated evaluation of groundwater hydrochemistry using multivariate statistics and irrigation-based water quality indices

Abstract This research evaluates the appropriateness of groundwater for potable and irrigation need in the Mathura District of Uttar Pradesh, India. Fifteen different hydro chemical parameters selected for analysis, they are pH, total dissolved solids (TDS), electrical conductivity (EC), Total Hardness (TH), Mg2+, Cl−, Ca2+, Na+, K+, NO3 −, SO4 2−, PO4 3−, F−, CO3 2−, and HCO3 − In the present research various approaches such as Multivariate statistical techniques (MSTs), Arithmetic water quality index (WQI), sodium absorption ratio (SAR), permeability index (PI), sodium percentage (Na%), US salinity, were examined. The WQI assessment reveals that 65% poor quality, 5% very poor quality and 15% water is not suitable for drinking purpose category. Thus, it indicates that the majority of groundwater surpassed acceptable thresholds for potable water consumption. Additional indices indicate that several metrics surpass their acceptable limits, rendering most samples inappropriate for irrigation. CaMgCl, NaCl and CaNaHCO3 attributed to interfaces between water and rock and ion exchange processes in Sodium–Potassium from water and Calcium-Magnesium from rock. Furthermore, the US Salinity graphic confirms that the majority of groundwater samples demonstrate very high salinity risks and Sodium hazard, especially for elevated salt concentrations. Influence the soil fertility, permeability of soil and crop growth. The findings of this study will be beneficial to policymakers and decision-making authorities in executing sustainable water quality initiatives and efficient management of water resources as per scientific principles of various global and national agencies.

Quality assurance through truncated life tests under the Lomax distribution

A Multi-Scale attention network for building extraction from high-resolution remote sensing images

Molecular dynamics-based explanation of the reinforcement geometry effects on CNT/graphene-reinforced Al0.3CoCrFeNi high-entropy alloys

Study on the spatial and temporal differentiation of intangible cultural heritage and its influencing factors in Shandong province

Health risk assessment of polycyclic aromatic hydrocarbons in common food crops within and outside selected mining areas of Ebonyi State, Nigeria

Body shape and labor market dynamics: analyzing salary information asymmetry and bargaining power

Soil microbes and organic fertilizer efficiency are associated with rice field topography

Anticancer and antibacterial activity of the cellulose nanocrystals modified by cyclodextrin and loaded by propolis

Physiological, nutritional, and productive responses of broilers supplemented with lemon, fenugreek, and sesame oils as feed additives

Abstract This study objective is to evaluate the influence of lemon, sesame, fenugreek essential oils and their mixture as feed additive on the growth performance, carcass criteria, nutrient digestibility, cecum microbial and certain blood parameters of broiler chicks. A total of 240 one day-old unsexed broiler chickens Ross 308 were divided into 5 treatment groups, each treatment had 6 replicates with 8 birds per each. the birds were fed a basal diet without supplementation (control group), basal diet supplemented with lemon essential oil (LEO) at 400 mg/kg (LEO group), basal diet supplemented with sesame essential oil (SEO) at 400 mg/kg (SEO group), basal diet supplemented with fenugreek essential oil (FEO) at 400 mg/kg (FEO group), basal diet supplemented with combination of LEO, SEO and FEO (Mix group), for 5 weeks. The results revealed that FEO supplementation significantly (P < 0.05) improved body weight (BW), body weight gain (BWG), feed intake (FI) and feed conversion ratio (FCR) compared to the control group. At the first week FI was lower in Mix group as well as improving FCR than the control and LEO groups (P < 0.05). Datary containing FEO, LEO, and SEO either alone or in combination significantly (P < 0.01) improved serum levels of total protein, globulin and urea and crude protein digestibility, increased cecal counts of Lactobacillus as well as reducing counts of cecal bacteria, E. coli, and Salmonella. Therefore, it was found that, in comparison to other treatments, broiler chicks provided 400 mg/kg of fenugreek oil demonstrated improved growth performance and general health condition.