Comprehensive Dataset Research Articles

Precision Agriculture: UAV-Based Soil Mapping and Remote Sensing Applications

Using Unmanned Aerial Vehicle (UAV) equipped with remote sensing cameras that measure the electromagnetic spectrum of light reflected back from the ground below, this technology does soil mapping and field analysis. The way that different elements reflect different light wavelengths can be used to identify them. Reams of data on those wavelengths are gathered by multispectral image sensors, which power sophisticated AI software that can identify even the smallest variations in the elemental composition of soil. There are numerous positives to soil mapping, including helping farmers choose the best crops, applying the right fertilizers, and scheduling irrigation. Soil mapping also makes it easier to create plans for soil conservation and restoration that work. UAVs facilitate rapid deployment, enabling data collection near end-users, while delivering a more comprehensive and precise dataset compared to traditional camera-based methods. This study will provide effective technical support and decision-making assistance for future agricultural land planning.

Simultaneously quantifying a novel five-component anti- migraine formulation containing ergotamine, propyphenazone, caffeine, camylofin, and mecloxamine using UV spectrophotometry and chemometric models

This study presents a new method for simultaneously quantifying a complex anti-migraine formulation containing five components (ergotamine, propyphenazone, caffeine, camylofin, and mecloxamine) using UV spectrophotometry and chemometric models. The formulation presents analytical challenges due to the wide variation in component concentrations (ERG: PRO: CAF: CAM: MEC ratio of 0.075:20:8:5:4) and highly overlapping UV spectra. To create a comprehensive validation dataset, the Kennard-Stone Clustering Algorithm was used to address the limitations of arbitrary data partitioning in chemometric methods. Three different chemometric models were evaluated: Classical Least Squares (CLS), Partial Least Squares (PLS), and Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS). Among these, MCR-ALS demonstrated excellent performance, achieving recovery values of 98–102% for all components, accompanied by minimal root mean square errors of calibration (0.072–0.378) and prediction (0.077–0.404). Moreover, the model exhibited high accuracy, with relative errors ranging from 1.936 to 3.121%, bias-corrected mean square errors between 0.074 and 0.389, and a good sensitivity (0.2097–1.2898 μg mL−1) for all components. The Elliptical Joint Confidence Region analysis further confirmed the predictive performance of the models, with MCR-ALS consistently showing the smallest ellipses closest to the ideal point (slope = 1, intercept = 0) for most analytes, indicating superior accuracy and precision. The approach's sustainability was rigorously assessed using six advanced metrics, validating its environmental friendliness, economic viability, and practical application. This approach effectively resolves complex pharmaceutical formulations, contributing to sustainable development objectives in quality control processes.

Machine Learning Models for the Classification of Histopathological Images of Colorectal Cancer

The aim of this study was to explore the application of computational models for the analysis of histopathological images in the context of colon cancer. A comprehensive dataset of colon cancer images annotated into eight distinct categories based on their representation of cancerous cell portions was used. The primary objective was to employ various image classification algorithms to assess their efficacy in the context of cancer classification. Additionally, this study investigated the use of feature extraction techniques to derive meaningful data from the images, contributing to a more nuanced understanding of cancerous tissues, comparing the performance of different image classification algorithms in the context of colon cancer image analysis. The findings of this research suggested that XGboost provides the highest accuracy (89.79%) and could contribute to the growing body of knowledge in computational pathology. Other algorithms, such as the random forest, SVM, and CNN, also provided satisfactory results, offering insights into the effectiveness of image classification algorithms in distinguishing between different categories of cancerous cells. This work holds implications for the development of more accurate and efficient tools, underscoring the potential of computational models in enhancing the analysis of histopathological images and improving diagnostic capabilities in cancer research.

Random generation of three-dimensional realistic ballast particles using generative adversarial networks

Natural ballast particles exhibit a wide array of shape characteristics and follow complex probability distributions, which are difficult to quantify using explicit formulas due to their implicit nature. Traditional methods for generating ballast particles are limited to shapes that adhere to specific shape indices. In this study, generative adversarial networks (GANs) were utilized to randomly generate realistic three-dimensional (3D) ballast particles. Representative ballast particle shapes were captured using 3D structured light scanning technology. The scanned shapes were then voxelized and augmented to create a comprehensive dataset representing various ballast particle shapes. This dataset served to train the generator and discriminator components of the GANs. The trained generator successfully produced 2,000 detailed 3D ballast particles. Further refinement of these particles was achieved using Gaussian blur, followed by the Laplace smoothing algorithms and marching cubes algorithm for surface reconstruction. The authenticity of the generated ballast particles was validated by comparing their shape indices with those of natural ballast particles, thus demonstrating the effectiveness of the trained GANs model. The generated ballast particles are suitable for use as templates in discrete element method (DEM) simulations, specifically for clumps and polyhedrons.

Optimizing Fire Scene Analysis: Hybrid Convolutional Neural Network Model Leveraging Multiscale Feature and Attention Mechanisms

The rapid and accurate detection of fire scenes in various environments is crucial for effective disaster management and mitigation. Fire scene classification is a critical aspect of modern fire detection systems that directly affects public safety and property preservation. This research introduced a novel hybrid deep learning model designed to enhance the accuracy and efficiency of fire scene classification across diverse environments. The proposed model integrates advanced convolutional neural networks with multiscale feature extraction, attention mechanisms, and ensemble learning to achieve superior performance in real-time fire detection. By leveraging the strengths of pre-trained networks such as ResNet50, VGG16, and EfficientNet-B3, the model captures detailed features at multiple scales, ensuring robust detection capabilities. Including spatial and channel attention mechanisms further refines the focus on critical areas within the input images, reducing false positives and improving detection precision. Extensive experiments on a comprehensive dataset encompassing wildfires, building fires, vehicle fires, and non-fire scenes demonstrate that the proposed framework outperforms existing cutting-edge techniques. The model also exhibited reduced computational complexity and enhanced inference speed, making it suitable for deployment in real-time applications on various hardware platforms. This study sets a new benchmark for fire detection and offers a powerful tool for early warning systems and emergency response initiatives.

MULTICLASS CLASSIFICATION FOR STUNTING PREDICTION USING DEEP NEURAL NETWORKS

Stunting is a chronic nutritional issue that hinders child growth and leads to serious long-term health and developmental impacts, particularly in developing countries. Therefore, early and accurate prediction of stunting is crucial for implementing effective interventions. This research aims to develop a multiclass classification model based on Deep Neural Networks (DNNs) to predict stunting status. The model is trained using a comprehensive dataset that encompasses various health variables related to stunting. The research process includes data collection, data preprocessing, dataset splitting, and training and evaluation of the DNNs model. The model can classify stunting status into four categories: stunted, severely stunted, normal, and tall. Further analysis is conducted to evaluate the influence of various parameters on the model's performance, including dataset splitting ratios (80:20 and 70:30) and learning rates (0.001, 0.0001, and 0.00001). The results show that a learning rate of 0.0001 yields the highest prediction accuracy, at 93.64% and 93.83% for the two data-splitting schemes. This indicates that this learning rate has achieved an optimal balance between convergence speed and the model's generalization capability. Additionally, the developed DNNs model can identify complex patterns hidden within the data without being affected by noise. These findings confirm that appropriate parameter selection, particularly the dataset splitting ratio and learning rate, can significantly enhance the DNNs model's ability to identify complex data patterns.

Does financial inclusion enhance farmers' resilience to climate change? Evidence from rural Ethiopia

AbstractFinancial inclusion is recognized as a vital driver of sustainable development and serves as a fundamental pillar of climate action. It is crucial to enhance the climate resilience of smallholder farmers in the face of severe and unpredictable climate shocks, which disproportionately affect them. However, the level of financial inclusion in Ethiopia remains low, and its impact on the climate resilience of smallholder farmers has not been thoroughly examined using rigorous model and comprehensive dataset. This study investigates the impact of financial inclusion on the climate resilience of rural households, using a large data set from the Ethiopian Socio‐Economic Survey. The principal component analysis was applied to construct a climate resilience index. The financial inclusion was measured using an index that encompasses three dimensions: penetration, availability, and usage. In order to address the endogenous nature of financial inclusion, an instrumental variable approach was employed, using the distance to the nearest financial institution and religion as instrumental variables. The results demonstrated a positive and significant impact of financial inclusion on the climate resilience of rural households. Therefore, the government should strengthen the provision of essential financial and related infrastructures in rural Ethiopia to improve access to financial products and services. Furthermore, it is essential for policymakers to initiate and implement financial sector reforms that ensure the availability of affordable and tailored financial services. These reforms should also prioritize the development of climate‐resilient agricultural finance, thereby contributing to the achievement of climate action goal of sustainable development.

Leveraging diverse genomic data to guide equitable carrier screening: Insights from gnomAD v.4.1.0.

Analysis of exome data from the latest release of the Genome Aggregation Database (gnomAD v.4.1.0) revealed a significant carrier burden of pathogenic/likely pathogenic (P/LP) variants in genes associated with autosomal-recessive conditions across diverse ancestral populations. Carrier screening panels are routinely offered to reproductive partners to inform their risk of having an affected child. Current guidelines from the American College of Medical Genetics and Genomics (ACMG) recommend screening for genes with a carrier frequency of at least 1/200 and associated with moderate/severe conditions. Here, we systematically analyzed >700,000 gnomAD v.4.1.0 exomes spanning eight ancestries to estimate the carrier frequency of P/LP variants in 2,987 genes associated with autosomal-recessive conditions. After expert curation for clinical severity, we identified 286 genes meeting the criteria for carrier screening. The number of genes exceeding the 1/200 threshold varied across populations, with 40 in the South Asian ancestry and up to 119 in the Ashkenazi Jewish population. Simulations showed that pan-ethnic screening panels offer advantages for individuals of diverse or admixed ancestry, while ancestry-specific panels may be preferable for genetically homogeneous populations. This study leveraged the most comprehensive genomic dataset to date to provide an updated candidate gene list for equitable carrier screening across diverse populations. Our findings highlight the need for continued expansion of genomic resources to better understand rare disease risk and inform screening efforts in underrepresented groups.

Interpretable machine learning model for digital lung cancer prescreening in Chinese populations with missing data

We developed an interpretable model, BOUND (Bayesian netwOrk for large-scale lUng caNcer Digital prescreening), using a comprehensive EHR dataset from the China to improve lung cancer detection rates. BOUND employs Bayesian network uncertainty inference, allowing it to predict lung cancer risk even with missing data and identify high-risk factors. Developed using data from 905,194 individuals, BOUND achieved an AUC of 0.866 in internal validation, with time- and geography-based external validations yielding AUCs of 0.848 and 0.841, respectively. In datasets with 10%–70% missing data, AUC ranged from 0.827 – 0.746. The model demonstrates strong calibration, clinical utility, and robust performance in both balanced and imbalanced datasets. A risk scorecard was also created, improving detection rates up to 6.8 times, available free online (https://drzhang1.aiself.net/). BOUND enables non-radiative, cost-effective lung cancer prescreening, excels with missing data, and addresses treatment inequities in resource-limited primary healthcare settings.

Advancing dermoscopy through a synthetic hair benchmark dataset and deep learning-based hair removal.

Early detection of melanoma is crucial for improving patient outcomes, and dermoscopy is a critical tool for this purpose. However, hair presence in dermoscopic images can obscure important features, complicating the diagnostic process. Enhancing image clarity by removing hair without compromising lesion integrity can significantly aid dermatologists in accurate melanoma detection. We aim to develop a novel synthetic hair dermoscopic image dataset and a deep learning model specifically designed for hair removal in melanoma dermoscopy images. To address the challenge of hair in dermoscopic images, we created a comprehensive synthetic hair dataset that simulates various hair types and dimensions over melanoma lesions. We then designed a convolutional neural network (CNN)-based model that focuses on effective hair removal while preserving the integrity of the melanoma lesions. The CNN-based model demonstrated significant improvements in the clarity and diagnostic utility of dermoscopic images. The enhanced images provided by our model offer a valuable tool for the dermatological community, aiding in more accurate and efficient melanoma detection. The introduction of our synthetic hair dermoscopic image dataset and CNN-based model represents a significant advancement in medical image analysis for melanoma detection. By effectively removing hair from dermoscopic images while preserving lesion details, our approach enhances diagnostic accuracy and supports early melanoma detection efforts.

A global effort to benchmark predictive models and reveal mechanistic diversity in long-term stroke outcomes.

Stroke remains a leading cause of mortality and long-term disability worldwide, with variable recovery trajectories posing substantial challenges in anticipating post-event care and rehabilitation planning. In response, we established the NeuralCup consortium to address these challenges by benchmarking predictive models of stroke outcome through a collaborative, data-driven approach. This study presents the findings of 15 participating teams worldwide who used a comprehensive dataset including clinical and imaging data, to identify and compare predictors of motor, cognitive, and emotional outcomes one-year post-stroke. Analyses integrated traditional and novel approaches, including machine learning algorithms to discover 'optimal recipes' for predicting each domain. The differences in these 'optimal recipes' reflect distinct brain mechanisms in response to different tasks. Key predictors across all domains included lesion characteristics, T1-weighted MRI sequences, and demographic factors. Additionally, the integration of FLAIR imaging and white matter tract analysis significantly improved the prediction of cognitive and motor outcomes, respectively. These findings support a multifaceted approach to stroke outcome prediction, underscoring the potential of collaborative data science to develop personalized care strategies that enhance recovery and quality of life for stroke survivors.

AgriFusion: A Low‐Carbon Sustainable Computing Approach for Precision Agriculture Through Probabilistic Ensemble Crop Recommendation

ABSTRACTOptimizing crop production is essential for sustainable agriculture and food security. This study presents the AgriFusion Model, an advanced ensemble‐based machine learning framework designed to enhance precision agriculture by offering highly accurate and low‐carbon crop recommendations. By integrating Random Forest, Gradient Boosting, and LightGBM, the model combines their strengths to boost predictive accuracy, robustness, and energy efficiency. Trained on a comprehensive dataset of 2200 instances covering key parameters like nitrogen, phosphorus, potassium, temperature, humidity, pH, rainfall, and crop type, the model underwent rigorous preprocessing for data integrity. The RandomizedSearchCV method was employed to do hyperparameter tuning, namely improving the number of trees in the Random Forest algorithm and the learning rates in the Gradient Boosting algorithm. This ensemble approach achieves a remarkable accuracy rate of 99.48%, optimizes computer resources, lowers carbon footprint, and responds efficiently to a variety of agricultural situations. The model's performance is confirmed using metrics including cross‐validation, accuracy, precision, recall, and F1 score. This demonstrates how the model might improve agricultural decision‐making, make the most use of available resources, and promote ecologically responsible farming practices.

Boosting the efficacy of green accounting for better firm performance: artificial intelligence and accounting quality as moderators

Purpose This study aims to deepen our understanding of how conventional technologies and robust accounting education standards can impact the effectiveness of green accounting practices in enhancing firm performance. To achieve this, the paper explores the moderating effects of artificial intelligence (AI) and accounting education quality on the relationship between green accounting and firm performance. Design/methodology/approach Using generalized method of moments estimation, this research uses a comprehensive dataset comprising 32,680 firm-year observations of listed companies from ten prominent countries – Canada, the UK, the USA, China, France, Germany, India, Japan, South Korea and Italy – over the period from 2012 to 2022. These countries, selected based on their high gross domestic product rankings as reported by the International Monetary Fund, ensure a diverse representation of economic strengths and capture a wide range of green accounting practices. Findings The study shows that green accounting practices positively impact current firm performance. Country-level AI positively moderates this relationship, suggesting that advanced AI infrastructure enhances the benefits of green accounting through improved data accuracy and decision-making. However, country-level accountancy education quality negatively moderates the relationship, indicating that stringent implementation of green accounting standards in these regions may introduce complexities and costs that reduce firm performance. Practical implications Integrating AI enhances data processing, predictive analytics and decision-making, improving green accounting effectiveness. High-quality accounting education ensures accurate reporting and greater transparency. These insights, when applied, can empower businesses to optimize sustainability strategies, assist policymakers in developing targeted regulations and guide educators in preparing accountants for the evolving demands of green accounting. Originality/value To the best of the authors’ knowledge, this study is the first to explore the combined moderating effects of AI and accounting education quality on the relationship between green accounting and firm performance. By highlighting the synergistic role of digital innovation and robust educational standards, this research offers novel insights into how these factors can enhance the effectiveness of green accounting practices and improve financial outcomes.

Evolution of institutional long-term care costs based on health factors

As many developed countries face the challenges of an aging population, the need to efficiently plan and finance long-term care (LTC) becomes increasingly important. Understanding the dynamics of care requirements and their associated costs is essential for sustainable healthcare systems. In this study, we employ a multi-state Markov model to analyze the transitions between care states of elderly individuals within institutional LTC in the canton of Geneva, Switzerland. Utilizing a comprehensive dataset of 21494 elderly residents, we grouped care levels into four broader categories reflecting the range from quasi-autonomy to severe dependency. Our model considers fixed covariates at admission, such as demographic details, medical diagnoses, and levels of dependence, to forecast transitions and associated costs. The main results illustrate significant variations in care trajectories and LTC costs across different health profiles, notably influenced by gender and initial care state. Females generally require longer periods with less intensive care, while conditions like severe and nervous diseases show quicker progression to more intensive care and higher initial costs. These transitions and expected length of stay in each state directly impact LTC costs, highlighting the necessity of advanced strategies to manage the financial burden. Our findings offer insights that can be utilized to optimize LTC services in response to the specific needs of institutionalized elderly people. These findings can be applied to enhance healthcare planning, the preparedness of infrastructure, and the design of insurance products.

CAUSALdb2: an updated database for causal variants of complex traits.

Unraveling the causal variants from genome wide association studies (GWASs) is pivotal for understanding genetic underpinnings of complex traits and diseases. Despite continuous efforts, tools to refine and prioritize GWAS signals need enhancement to address the direct causal implications of genetic variations. To overcome challenges related to statistical fine-mapping in identifying causal variants, CAUSALdb has been updated with novel features and comprehensive datasets, morphing into CAUSALdb2. This expanded repository integrates 15057 updated GWAS summary statistics across 10839 unique traits and implements both LD-based and LD-free fine-mapping approaches, including innovative applications of approximate Bayes Factor and SuSiE. Additionally, by incorporating larger LD reference panels such as TOPMED and UK Biobank, and integrating functional annotations via PolyFun, CAUSALdb2 enhances the accuracy and context of fine-mapping results. The database now supports interrogation of additional causal signals and offers sophisticated visualizations to aid researchers in deciphering complex genetic architectures. By facilitating a deeper and more precise characterisation of causal variants, CAUSALdb2 serves as a crucial tool for advancing the genetic analysis of complex diseases. Available freely, CAUSALdb2 continues to set benchmarks in the post-GWAS era, fostering the development of targeted diagnostics and therapeutics derived from responsible genetic research. Explore these advancements at http://mulinlab.org/causaldb.

Prediction of strata settlement in undersea metal mining based on deep forest

Undersea mining encounters challenges due to the presence of seawater. An influx of seawater into stop in undersea can result in enormous disaster. Predicting strata settlement is a crucial measure to ensure the safety of undersea mining. This study proposed an intelligent model based on deep forest (DF) to evaluate the strata settlement during undersea mining. Initially, the strata displacement was monitored in the Xishan mining area of Sanshandao gold mine, China. Comprehensive datasets encompassing roof displacement and twelve influencing factors were compiled from 120 observations. Then, these datasets were statistically analyzed and used to train the DF model. The developed DF model achieved a training R2 of 0.971 and a testing R2 of 0.936. Compared with other machine learning models, the DF model has superior performance in the prediction of strata settlement. Moreover, a graphical user interface was designed to facilitate the application of the DF model. Finally, to validate model feasibility, displacement monitoring was conducted in the Xinli mining area of Sanshandao gold mine. Additional datasets were collected to validate the capability of the DF model. The results suggested that the DF model can be used to predict strata subsidence in undersea mining effectively.

Developing WasteSAM: A novel approach for accurate construction waste image segmentation to facilitate efficient recycling

The escalating volume of construction activities and resultant waste generation underscores the imperative for developing sophisticated segmentation models to facilitate efficient sorting and recycling processes. This study introduces WasteSAM, an enhanced iteration of the segment anything model (SAM), specifically tailored to address the intricate complexities inherent in construction waste imagery. Drawing upon a comprehensive dataset comprising over 15,000 masks representing five distinct categories of construction materials, WasteSAM exhibits notably superior segmentation capabilities. Quantitative analysis demonstrates significant performance improvements, with WasteSAM outperforming the original SAM model by an average of 23.9% in dice similarity coefficient and 30.0% in normalized surface distance metrics. The integration of stereo-image techniques in refining the training dataset has facilitated WasteSAM in more accurately discerning the three-dimensional structure of waste materials, thereby augmenting the precision of waste classification. Noteworthy is the model’s adeptness in handling intricate textures and patterns across diverse imaging modalities, including varying lighting conditions and complex object interactions. While showing promising results, this study also highlights the need for high-quality, diverse datasets that reflect real-world construction site complexities, rather than merely larger datasets.

MCNN-AAPT: accurate classification and functional prediction of amino acid and peptide transporters in secondary active transporters using protein language models and multi-window deep learning

Secondary active transporters play a crucial role in cellular physiology by facilitating the movement of molecules across cell membranes. Identifying the functional classes of these transporters, particularly amino acid and peptide transporters, is essential for understanding their involvement in various physiological processes and disease pathways, including cancer. This study aims to develop a robust computational framework that integrates pre-trained protein language models and deep learning techniques to classify amino acid and peptide transporters within the secondary active transporter (SAT) family and predict their functional association with solute carrier (SLC) proteins. The study leverages a comprehensive dataset of 448 secondary active transporters, including 36 solute carrier proteins, obtained from UniProt and the Transporter Classification Database (TCDB). Three state-of-the-art protein language models, ProtTrans, ESM-1b, and ESM-2, are evaluated within a deep learning neural network architecture that employs a multi-window scanning technique to capture local and global sequence patterns. The ProtTrans-based feature set demonstrates exceptional performance, achieving a classification accuracy of 98.21% with 87.32% sensitivity and 99.76% specificity for distinguishing amino acid and peptide transporters from other SATs. Furthermore, the model maintains strong predictive ability for SLC proteins, with an overall accuracy of 88.89% and a Matthews Correlation Coefficient (MCC) of 0.7750. This study showcases the power of integrating pre-trained protein language models and deep learning techniques for the functional classification of secondary active transporters and the prediction of associated solute carrier proteins. The findings have significant implications for drug development, disease research, and the broader understanding of cellular transport mechanisms.

Flower Recognition System using CNN

Flower recognition stands as a pivotal task across diverse fields such as botany, agriculture, and environmental studies. The manual process of identifying flower species is not only laborious but also susceptible to inaccuracies. In response to this challenge, this report introduces a Flower Recognition System employing Convolutional Neural Networks (CNNs) to automate and enhance this process. By harnessing the capabilities of deep learning, our system demonstrates remarkable proficiency in accurately classifying various flower species from input images. The CNN architecture, meticulously designed and trained on a comprehensive dataset of labeled flower images, exhibits robustness and efficiency in flower identification tasks. Through rigorous experimentation and evaluation, this report meticulously examines the development and performance of the Flower Recognition System. Furthermore, it explores the potential applications of this system in facilitating automated flower species identification, thereby contributing to advancements in botanical research, agricultural practices, and environmental conservation efforts.

A Dataset of 3M Single-DOF Planar 4-, 6-, and 8-Bar Linkage Mechanisms With Open and Closed Coupler Curves for Machine Learning-Driven Path Synthesis

Abstract In recent years, there has been a strong interest in applying machine learning techniques to path synthesis of linkage mechanisms. However, progress has been stymied due to a scarcity of high-quality datasets. In this article, we present a comprehensive dataset comprising nearly three million samples of 4-, 6-, and 8-bar linkage mechanisms with open and closed coupler curves. Current machine learning approaches to path synthesis also lack standardized metrics for evaluating outcomes. To address this gap, we propose six key metrics to quantify results, providing a foundational framework for researchers to compare new models with existing ones. We also present a variational autoencoder-based model in conjunction with a k-nearest neighbor search approach to demonstrate the utility of our dataset. In the end, we provide example mechanisms that generate various curves along with a numerical evaluation of the proposed metrics.