Spatial distribution of heavy metal abundance at distance gradients of roadside agricultural soil from the busiest highway in Bangladesh: A multi-index integration approach.

Bangladesh is currently experiencing significant infrastructural development in road networking system through the construction or reconstruction of multiple roads and highways. Consequently, there is a rise in traffic intensity on roads and highways, along with a significant contamination of adjacent agricultural soils with heavy metals. The purpose of this study was to evaluate the ecological risk, health risk and the abundance of seven heavy metals (Cu, Mn, Pb, Cd, Cr, As, and Ni) in three distance gradients (0, 300, and 500 m) of agricultural soil along the Dhaka-Chattogram highway. The concentration of heavy metals was measured with an Atomic Absorption Spectrophotometer (AAS) on a total of 36 soil samples that were taken from 12 different sampling sites. Based on the findings, Cd had a high contamination factor for all distance gradients, whereas Cr had a moderate contamination factor in 67% of the study areas. According to the Pollution Load Index (PLI), Cd, Cr, and Pb were the predominant pollutants. Principal component analysis (PCA) result shows these metals mainly came from anthropogenic sources. The considerable positive correlations between Cu-Pb, Cu-Cd, Pb-Cd, and Cr-Ni all pointed to shared anthropogenic origins. As per Potential Ecological Risk Assessment (PERI) analysis, Pb, Cd, Cr, and Ni each contribute significantly and pose a moderate threat. The Target Hazard Quotient (THQ) values for all pathways of exposure to Pb and Cr in soils were more than 1, which would pose a significant risk to human health in the following order: THQadult female > THQadult male > THQchildren. This study will help to evaluate the human health risk and develop a better understanding of the heavy metal abundance scenario in the agricultural fields adjacent to this highway.

Learning Meta-Learning (LML) dataset: Survey data of meta-learning parameters.

The 'Learning Meta-Learning' dataset presented in this paper contains both categorical and continuous data of adult learners for 7 meta-learning parameters: age, gender, degree of illusion of competence, sleep duration, chronotype, experience of the imposter phenomenon, and multiple intelligences. Convenience sampling and Simple Random Sampling methods are used to structure the anonymous online survey data collection voluntarily for LML dataset creation. The responses from the 54 survey questionnaires contain raw data from 1021 current students from 11 universities in Bangladesh. The entire dataset is stored in an excel file and the entire questionnaire is accessible at (10.5281/zenodo.8112213) In this article mean and standard deviation for the participant's baseline attributes are given for scale parameters, and frequency and percentage are calculated for categorical parameters. Academic curriculum, courses as well as professional training materials can be reviewed and redesigned with a focus on the diversity of learners. How the designed courses will be learned by learners along with how they will be taught is a significant point for education in any discipline. As the survey questionnaires are set for adult learners and only current university students have participated in this survey, this dataset is appropriate for study andragogy and heutagogy but not pedagogy.

Application of deep learning in detecting neurological disorders from magnetic resonance images: a survey on the detection of Alzheimer's disease, Parkinson's disease and schizophrenia.

Neuroimaging, in particular magnetic resonance imaging (MRI), has been playing an important role in understanding brain functionalities and its disorders during the last couple of decades. These cutting-edge MRI scans, supported by high-performance computational tools and novel ML techniques, have opened up possibilities to unprecedentedly identify neurological disorders. However, similarities in disease phenotypes make it very difficult to detect such disorders accurately from the acquired neuroimaging data. This article critically examines and compares performances of the existing deep learning (DL)-based methods to detect neurological disorders-focusing on Alzheimer's disease, Parkinson's disease and schizophrenia-from MRI data acquired using different modalities including functional and structural MRI. The comparative performance analysis of various DL architectures across different disorders and imaging modalities suggests that the Convolutional Neural Network outperforms other methods in detecting neurological disorders. Towards the end, a number of current research challenges are indicated and some possible future research directions are provided.