Search

This work presents an optimization framework for the simultaneous design of low-carbon hydrogen and carbon dioxide capture, use, and storage supply chains. The proposed model is a large-scale multiperiod mixed-integer linear programming that maximizes the net present value. Our approach addresses technological hydrogen production and carbon capture pathways, multimodal transportation of materials, electricity generation and transmission, and carbon dioxide utilization via enhanced oil recovery. The novelty of this work is twofold: the design of a low-carbon hydrogen supply chain and its integration into an oil field development of infrastructure designed to produce incremental crude oil via carbon dioxide injection; thus, it can be captured, utilized, and stored. We consider crude oil production profiles, characteristics of oil reservoirs, and well drilling constraints to represent the CO 2 -EOR operation. We conduct a nationwide long-term planning case study considering the potential for enhanced oil recovery operations and the availability of biomass, solar, and wind resources in Colombia to produce low-carbon hydrogen and monetize carbon dioxide through incremental crude oil production. Moreover, a graphical user interface (GUI) was designed to guide the end user in setting up case studies and analyzing the results through a series of tabular structures. • Modelling framework for the simultaneous design of low-carbon hydrogen supply chains and CO 2 -EOR operations. • Prediction of CO 2 injection considering petrophysical properties of the oil field. • Low-carbon hydrogen production from biomass and renewable energy. • Nationwide study applied to the Colombian context. [ABSTRACT FROM AUTHOR]

We present TDS Simulator, a new software tool aimed at modelling thermal desorption spectroscopy (TDS) experiments. TDS is a widely used technique for quantifying key characteristics of hydrogen-material interactions, such as diffusivity and trapping. However, interpreting the output of TDS experiments is non-trivial and requires appropriate post-processing tools. This work introduces the first software tool capable of simulating TDS curves for arbitrary choices of material parameters and hydrogen trap characteristics, using the primary hydrogen diffusion and trapping models (Oriani, McNabb–Foster). Moreover, TDS Simulator contains a specific functionality for loading experimental TDS data and conducting the inverse calibration of a selected transport model, providing automatic estimates of the density and binding energy of each hydrogen trap type in the material. In its first version, TDS Simulator is provided as a MATLAB App, which is made freely available to the community and provides a simple graphical user interface (GUI) to make use of TDS Simulator straightforward. As reported in the present manuscript, the outputs of TDS Simulator have been extensively validated against literature data. Demonstrations of automatic determination of trap characteristics from experimental data through the optimization tool are also provided. The present work enables an efficient and straightforward characterization of hydrogen-material characteristics relevant to multiple applications, from nuclear fusion to the development of hydrogen-compatible materials for the hydrogen economy. TDS Simulator can be downloaded from https://mechmat.web.ox.ac.uk/codes. • TDS Simulator is a free, user–friendly App for simulating H desorption experiments. • Multiple H transport theories (Oriani, McNabb-Foster) and trap sites are considered. • The provided inference algorithm enables identifying trap quantities from test data. • Each of the elements implemented in TDS Simulator is extensively validated. • Lab data is analysed, proving strong inference abilities and bringing new insight. [ABSTRACT FROM AUTHOR]

This software assistant aims at calculating the dose-response relations of tumors and normal tissues, or clinically assessing already determined values by other researchers. It can also indicate the optimal dose prescription by optimizing the expected treatment outcome. The software is developed solely in python programming language, and it employs PSFL license for its Graphical User Interface (GUI), NUMPY, MATPLOTLIB, and SCIPY libraries. It comprises of two components. The first is the Dose-response relations derivation component, which takes as input the dose volume histograms (DVHs) of patients and their recorded responses regarding a given clinical endpoint to determine the parameters of different tumor control probability (TCP) or normal tissue complication probability (NTCP) models. The second is the Treatment Plan Assessment component, which uses the DVHs of a plan and the dose-response parameters values of the involved tumors and organs at risk (OARs) to calculate their expected responses. Additionally, the overall probabilities of benefit (P B), injury (P I) and complication-free tumor control (P +) are calculated. The software calculates rapidly the corresponding generalized equivalent uniform doses (gEUD) and biologically effective uniform doses ( D ‾ ‾) for the Lyman-Kutcher-Burman (LKB), parallel volume (PV) and relative seriality (RS) models respectively, determining the model parameters. In the Dose-Response Relations Derivation component, the software plots the dose-response curves of the irradiated organ with the relevant confidence internals along with the data of the patients with and without toxicity. It also calculates the odds ratio (OR) and the area under the curve (AUC) of different dose metrics or model parameter values against the individual patient outcomes to determine their discrimination capacity. It also performs a goodness-of-fit evaluation of any model parameter set. The user has the option of viewing plots like Scatter, 3D surfaces, and Bootstrap plots. In the Treatment Plan Assessment part, the software calculates the TCP and NTCP values of the involved tumors and OARs, respectively. Furthermore, it plots the dose-response curves of the TCPs, NTCPs, P B , P I , and P + for a range of prescription doses for different treatment plans. The presented software is ideal for efficiently conducting studies of radiobiological modeling. Furthermore, it is ideal for performing treatment plan assessment, comparison, and optimization studies. [ABSTRACT FROM AUTHOR]

The list of standard abbreviations for JDS is available at adsa.org/jds-abbreviations-24. Nonstandard abbreviations are available in the Notes. Dairy farmers face increasing pressure to reduce GHG emissions (i.e., carbon dioxide, CO 2 ; methane, CH 4 ; and nitrous oxide, N 2 O), but measuring on-farm GHG emissions directly is costly or impractical. Therefore, the dairy industry has relied upon mathematical models to estimate these emissions. However, current models tend to be not user-friendly, difficult to access, or sometimes very research-focused, limiting their practical use. To address this, we introduce the DairyPrint model, a user-friendly tool designed to estimate GHG emissions from dairy farming. The model integrates herd dynamics, manure management, crop, and feed costs considerations, simplifying the estimation process while providing comprehensive insights. The herd module simulates monthly herd dynamics based on inputs as total cows, calving interval, and culling rate, outputting average annual demographics and estimating various animal-related variables (i.e., DMI, milk yield, manure excretion, and enteric CH 4 emissions). These outputs feed into other modules, such as the manure module, which calculates emissions based on manure, weather data, and facility type. The manure module processes manure according to farm practices, and the crop module accounts for GHG emissions from manure, fertilizers, and limestone application, also estimating nutrient balances. The DairyPrint model was developed using the Shiny framework and the Golem package for robust production-grade Shiny applications in the R programming language. We evaluated the model across 32 simulation scenarios by combining various factors and considering a standard freestall system with 1,000 dairy cows averaging 40 kg/d of milk production. These factors included 2 NDF-ADF levels in the diet (28%–22.8% and 24%–19.5%), the presence or absence of 3-nitrooxypropanol (3-NOP) dietary addition (yes or no) at an average dose of 70 mg/kg DM per cow daily, the type of bedding used (sawdust or sand), the frequency of manure pond emptying (once yearly, only in fall; or twice a year, in fall and spring), and the use or nonuse of a biodigester plus solid-liquid separator (Biod + SL). In our results across the 32 scenarios simulated, the average GHG emission was 0.811 kg CO 2 eq/kg of milk, corrected for fat and protein contents (4% and 3.3%, respectively), ranging from 0.644 to 1.082. Notably, the scenario yielding the lowest GHG emission (i.e., 0.644 kg CO 2 eq/kg) involved a combination of factors, including a lower NDF-ADF level in the diet in addition to incorporation of 3-NOP, use of sand as bedding, application of Biod + SL, and strategic manure pond emptying in both fall and spring. Conversely, the scenario that resulted in the highest GHG emission (i.e., 1.082 kg CO 2 eq/kg) involved a combination of a higher NDF-ADF level in the diet and excluded incorporation of 3-NOP, use of sawdust as bedding, no application of Biod + SL, and manure pond emptying only in fall. All these scenarios can easily be simulated in the DairyPrint model, with results obtained immediately for user evaluation. Therefore, the DairyPrint model can help farmers move toward improved sustainability, providing a user-friendly and intuitive graphical user interface allowing the user to ask what-if questions. [ABSTRACT FROM AUTHOR]

[Display omitted] • Six machine learning models were used to predict MR and CT. • GPR model demonstrated the best predictive performance. • SHAP and PDP were introduced for model interpretation. • An easy-to-use GUI was developed for the prediction of MR and MC. Preparation of activated carbons is an important way to utilize municipal sludge (MS) resources, while drying is a pretreatment method for making activated carbons from MS. In this study, machine learning techniques were used to develop moisture ratio (MR) and composting temperature (CT) prediction models for the thermally assisted biodrying process of MS. First, six machine learning (ML) models were used to construct the MR and CT prediction models, respectively. Then the hyperparameters of the ML models were optimized using the Bayesian optimization algorithm, and the prediction performances of these models after optimization were compared. Finally, the effect of each input feature on the model was also evaluated using SHapley Additive exPlanations (SHAP) analysis and Partial Dependence Plots (PDPs) analysis. The results showed that Gaussian process regression (GPR) was the best model for predicting MR and CT, with R2 of 0.9967 and 0.9958, respectively, and root mean square errors (RMSE) of 0.0059 and 0.354 ℃. In addition, graphical user interface software was developed to facilitate the use of the GPR model for predicting MR and CT by researchers and engineers. This study contributes to the rapid prediction, improvement, and optimization of MR and CT during thermally assisted biodrying of MS, and also provides valuable guidance for the dynamic regulation of the drying process. [ABSTRACT FROM AUTHOR]

License plate recognition technology is a key to realizing intelligent traffic management. Nowadays, most of the license plate recognition technologies are only designed for the preferred lighting conditions. But in the natural scenes, the change of the weather and the light intensity throughout the day carry many difficulties to the location and recognition of the license plate system. To solve this problem, we proposed a license plate recognition system based on OpenCV and SVM (Support Vector Machines) in this paper. First, the Sobel Edge Detection algorithm combined with Mathematical Morphology was used to determine the approximate outline of the license plate. And Orthographic Projection was then used to segment the characters in the license plate. Finally, the segmented characters were identified by using SVM. The recognition accuracy of the proposed system is 96.52%. In order to improve the operability of the system, we have designed a Graphical User Interface (GUI) based on PyQt5, which makes the interface more beautiful and simple. [ABSTRACT FROM AUTHOR]

This paper uses Stable Diffusion ComfyUI platform to construct a set of artificial intelligence working system in interior design field. Stable Diffusion is a deep learning image generation model that is capable of transforming text descriptions into high-quality images. Its open source nature promotes the wide application and innovative development of the technology. ComfyUI, the extension interface of Stable Diffusion, uses a node-based graphical user interface and modularization to simplify image generation. These efficient artificial intelligence image processing tools are practical in interior design, optimizing the process and improving efficiency for designers. [ABSTRACT FROM AUTHOR]

In this paper, we explore how to use a Nvidia Jetson Nano and Python to create a system that detects weariness in a person's face using computer vision and machine learning techniques. The system captures the person's face in real time, preprocesses the picture to cut noise, then detects the face using Haar cascades. Next, using computer vision algorithms, characteristics associated with weariness, such as the eyes, are retrieved. These characteristics are then used to build a machine learning model that can predict weariness in the live stream feed. Lastly, using a graphical user interface, the findings are shown, and the system may be fine-tuned to increase accuracy. This device might be used in applications such as driver monitoring and traffic safety. [ABSTRACT FROM AUTHOR]

• FSP was performed on WE 43 Mg alloy to assess the corrosion behavior using ML. • Synthetic data was generated using PSO for virtual sample generation. • Five different machine-learning models were compared using the original dataset. • A sensitivity analysis was performed using Shapley Additive Explanations. • A graphical user interface (GUI) has been created to predict the corrosion rate. The corrosion rate is a crucial factor that impacts the longevity of materials in different applications. After undergoing friction stir processing (FSP), the refined grain structure leads to a notable decrease in corrosion rate. However, a better understanding of the correlation between the FSP process parameters and the corrosion rate is still lacking. The current study used machine learning to establish the relationship between the corrosion rate and FSP process parameters (rotational speed, traverse speed, and shoulder diameter) for WE43 alloy. The Taguchi L 27 design of experiments was used for the experimental analysis. In addition, synthetic data was generated using particle swarm optimization for virtual sample generation (VSG). The application of VSG has led to an increase in the prediction accuracy of machine learning models. A sensitivity analysis was performed using Shapley Additive Explanations to determine the key factors affecting the corrosion rate. The shoulder diameter had a significant impact in comparison to the traverse speed. A graphical user interface (GUI) has been created to predict the corrosion rate using the identified factors. This study focuses on the WE43 alloy, but its findings can also be used to predict the corrosion rate of other magnesium alloys. [ABSTRACT FROM AUTHOR]

The popularity of photography has grown as more individuals have access to camera equipment. Photos seem significant in our daily lives because they are rich with information. However, to learn more, images must often be enhanced. While many technologies are available to enhance the photographs, they are often used to produce phony shots, which leads to the spread of false information. The serious problem of image forgeries has to be addressed. To detect picture fraud, several well-established frameworks have been developed in the past. Recently, there has been an increase in the use of convolutional neural networks (CNNs), which has affected the localization of photo forgeries. One of the most difficult methods of picture falsification is image splicing, which involves cutting a section of one image into another. The picture forgery localization methods based on literature have a lot of drawbacks. As such, developing a technique for accurately and reliably identifying forgeries in photos that have been manipulated is crucial. We provide a potent deep learning-based approach that combines CNN with error level analysis (ELA) to detect picture fraud. To ascertain whether or not a picture has been manipulated, the ELA from an image is extracted and sent to CNN. Transfer learning with UNET (encoder-decoder model), and ELA is used to locate the tampered area. The recommended methods for predicting the changed location. The experiment results on the CASIA 2.0 database are quite encouraging, and these methodologies have been duly evaluated. The results of the recommended model outperform other well-known Transfer learning methodologies compared to the findings. The study suggests that the proposed method might get 0.917 and 0.935 accuracy and precision, respectively. Moreover, the graphical user interface—which enables us to determine the veracity of an image—is designed using these models. [ABSTRACT FROM AUTHOR]

Real-Time rendering is the technique which allows us to have graphical applications in our everyday life, whether it is a 3D game or a tool with graphical user interface. Nowadays graphics rendering is handled by the GPU (Graphics Processing Unit) in our device. There are many layers of abstraction above the programming of GPUs through libraries and graphics engines, though the most low-level way of accessing a GPU in user-mode applications is using a Graphics API. Due to the need of high performance and low-level capabilities usually these APIs are used from C or C++, but we realized the need to utilize these APIs in higher level languages as well. In our approach we're using the.NET C# language for developing multi-platform real-time graphical applications instead of the C or C++ languages. Using the modern.NET environment, we're able to use Graphics APIs for rendering onto common.NET UI Frameworks while consuming all our previously implemented C# libraries and.NET technologies in the same application. To maintain compatibility with multiple platforms we're developing a library system allowing the use different Graphics APIs from the same C# source-code. The library system contains a Graphics API abstraction layer with multiple Graphics API implementations of this layer in C# and a C# to shader language compiler for cross-API shader development in C#. In this paper, we're proposing our considerations for implementing a library to be able to use the Vulkan and OpenGL APIs through a single C# codebase. We provide solutions for multi-platform rendering and dealing with the low-level challenges of using the two deeply different APIs, while maintaining performance capable to do real-time rendering. [ABSTRACT FROM AUTHOR]

This study presents an integrated framework of machine learning models (Artificial Neural Network, Ensembled Learning Tree, Support Vector Machine, and Gaussian Process Regression) and particle swarm optimization (PSO) to predict and optimize hydrogen production from SCWG using sewage sludge characteristics and process parameters. According to the results, ELT-PSO is preferred for forecasting hydrogen yield (Coefficient of determination (R2) = 0.997, Root Mean Square Error (RMSE) = 0.093), and it is highly suggested for handling complex variable-target correlation. However, Support Vector Machine (SVM) performed poorly, with R2 = 0.83 and RMSE = 2.28. According to SHAP feature importance Temperature, Carbon, Hydrogen, and Pressure are among the parameters that have strong impact. In addition, by adjusting the ML hyperparameters, optimization method (PSO) was used to maximize H2 yield. The optimized ELT-PSO model was utilized to establish a Graphical User Interface that made it simple to calculate H2 yield. [Display omitted] • ML with PSO used for prediction and analysis of H2 yield from SCWG. • PCC show temperature, oxygen, hydrogen, and volatile matter impact significantly. • ELT-PSO performed better than other machine learning optimization integrated model. • GUI is developed for H2 yield prediction for ease in the process. [ABSTRACT FROM AUTHOR]

Smart manufacturing requires easily reconfigurable robotic systems to increase the flexibility in presence of market uncertainties by reducing the set-up times for new tasks. One enabler of fast reconfigurability is given by intuitive robot programming methods. On the one hand, offline skill-based programming (OSP) allows the definition of new tasks by sequencing pre-defined, parameterizable building blocks termed as skills in a graphical user interface. On the other hand, programming by demonstration (PbD) is a well known technique that uses kinesthetic teaching for intuitive robot programming, where this work presents an approach to automatically recognize skills from the human demonstration and parameterize them using the recorded data. The approach further unifies both programming modes of OSP and PbD with the help of an ontological knowledge base and empowers the end user to choose the preferred mode for each phase of the task. In the experiments, we evaluate two scenarios with different sequences of programming modes being selected by the user to define a task. In each scenario, skills are recognized by a data-driven classifier and automatically parameterized from the recorded data. The fully defined tasks consist of both manually added and automatically recognized skills and are executed in the context of a realistic industrial assembly environment. [ABSTRACT FROM AUTHOR]

Degradation in reinforced concrete (RC) members associated with rebar corrosion in old buildings is one major concern that can significantly affect the strength and serviceability of RC structures. Strengthening existing structures using FRP can comprise complex assessment and design processes. This study addresses the critical issue of modeling the behavior associated with steel reinforcement corrosion in RC slab-column joints (SCJs). Three modeling methods, namely artificial neural network (ANN), Eureqa's symbolic regression algorithm, and analytical modeling, are proposed to predict the punching shear (PS) capacity of corroded RC SCJs by accounting for the changes in material properties due to corrosion and the strengthening provided by different types of FRP. The accuracy of the proposed models is validated through comparisons with experimental results and a verified finite element (FE) model that considers material deterioration and bond degradation. The results demonstrate that the proposed models offer more accurate PS capacity predictions than existing design codes. Among the considered prediction models, the ANN model exhibits superior performance and is recommended for PS capacity prediction. A practical design tool in the form of a graphical user interface (GUI) was created to facilitate the assessment and prediction of the PS behavior of corroded RC SCJs. Engineers can conveniently use this GUI to evaluate and predict the performance of such joints. [ABSTRACT FROM AUTHOR]

[Display omitted] Cells cultured in 3D fibrous biopolymer matrices exert traction forces on their environment that induce deformations and remodeling of the fiber network. By measuring these deformations, the traction forces can be reconstructed if the mechanical properties of the matrix and the force-free matrix configuration are known. These requirements limit the applicability of traction force reconstruction in practice. In this study, we test whether force-induced matrix remodeling can instead be used as a proxy for cellular traction forces. We measure the traction forces of hepatic stellate cells and different glioblastoma cell lines and quantify matrix remodeling by measuring the fiber orientation and fiber density around these cells. In agreement with simulated fiber networks, we demonstrate that changes in local fiber orientation and density are directly related to cell forces. By resolving Rho-kinase (ROCK) inhibitor-induced changes of traction forces, fiber alignment, and fiber density in hepatic stellate cells, we show that the method is suitable for drug screening assays. We conclude that differences in local fiber orientation and density, which are easily measurable, can be used as a qualitative proxy for changes in traction forces. The method is available as an open-source Python package with a graphical user interface. [ABSTRACT FROM AUTHOR]

Despite advancements in interaction with immersive Virtual Reality (VR) systems, using hand gestures for all interactions still imposes some challenges, especially in interactions with graphical user interfaces that are usually performed with point-and-click interfaces. Therefore, exploring the use of alternative hands-free methods for selection is essential to overcome usability problems and provide natural interaction for users. The results and insights gained from this exploration can lead to enhanced user experiences in VR applications. This study aims to contribute to the literature with the evaluation of the usability of the most commonly used hands-free methods for selection and system control tasks in immersive VR and their impact on standard and validated experience and usability metrics, namely the sense of presence, cybersickness, system usability, workload, and user satisfaction. A Fitts' selection task was performed using a within-subjects design by nine participants experienced in VR. The methods evaluated were the handheld controllers, the head gaze, eye gaze, and voice commands for pointing at the targets, and dwell time and voice commands to confirm the selections. Results show that the methods provide similar levels of sense of presence and low cybersickness while showing low workload values and high user satisfaction, matching the experience of traditional handheld controllers for non-multimodal approaches. The assisted eye gaze with dwell was the preferred hands-free method and the one with the highest values of usability. Still, developers should minimize the number of gaze movements to reduce fatigue. The evaluation also showed that using a multimodal approach for selections, especially using the voice, decreases user satisfaction and increases users' frustration. [Display omitted] • VR experience was not significantly different between the methods. • The workload was low and user satisfaction high through all hands-free methods. • Assisted eye gaze with dwell was the preferred hands-free method. • Voice confirmations decrease satisfaction and increase frustration. • Hands-free methods with dwell matched the usability of controllers. [ABSTRACT FROM AUTHOR]

• A systematic evaluation of four machine learning algorithms were explored to rationalise the complex relationships in Mg-alloy data. • The random forest regression model and neural network were robust models for predicting Mg alloys properties. • A graphical user interface (GUI) webtool was developed to facilitate the use of the machine learning models in predicting the mechanical properties of Mg alloys. • Machine learning (ML) models provide significant implications for the design and development of high-performance Mg alloys. Machine learning (ML) models provide great opportunities to accelerate novel material development, offering a virtual alternative to laborious and resource-intensive empirical methods. In this work, the second of a two-part study, an ML approach is presented that offers accelerated digital design of Mg alloys. A systematic evaluation of four ML regression algorithms was explored to rationalise the complex relationships in Mg-alloy data and to capture the composition-processing-property patterns. Cross-validation and hold-out set validation techniques were utilised for unbiased estimation of model performance. Using atomic and thermodynamic properties of the alloys, feature augmentation was examined to define the most descriptive representation spaces for the alloy data. Additionally, a graphical user interface (GUI) webtool was developed to facilitate the use of the proposed models in predicting the mechanical properties of new Mg alloys. The results demonstrate that random forest regression model and neural network are robust models for predicting the ultimate tensile strength and ductility of Mg alloys, with accuracies of ∼80% and 70% respectively. The developed models in this work are a step towards high-throughput screening of novel candidates for target mechanical properties and provide ML-guided alloy design. [ABSTRACT FROM AUTHOR]

Characterization of the optic nerve through measurement of optic nerve diameter (OND) and optic nerve sheath diameter (ONSD) using transorbital sonography (TOS) has proven to be a useful tool for the evaluation of intracranial pressure (ICP) and multiple neurological conditions. We describe a deep learning–based system for automatic characterization of the optic nerve from B-mode TOS images by automatic measurement of the OND and ONSD. In addition, we determine how the signal-to-noise ratio in two different areas of the image influences system performance. A UNet was trained as the segmentation model. The training was performed on a multidevice, multicenter data set of 464 TOS images from 110 subjects. Fivefold cross-validation was performed, and the training process was repeated eight times. The final prediction was made as an ensemble of the predictions of the eight single models. Automatic OND and ONSD measurements were compared with the manual measurements taken by an expert with a graphical user interface that mimics a clinical setting. A Dice score of 0.719 ± 0.139 was obtained on the whole data set merging the test folds. Pearson's correlation was 0.69 for both OND and ONSD parameters. The signal-to-noise ratio was found to influence segmentation performance, but no clear correlation with diameter measurement performance was determined. The developed system has a good correlation with manual measurements, proving that it is feasible to create a model capable of automatically analyzing TOS images from multiple devices. The promising results encourage further definition of a standard protocol for the automatization of the OND and ONSD measurement process using deep learning–based methods. The image data and the manual measurements used in this work will be available at 10.17632/kw8gvp8m8x.1. [ABSTRACT FROM AUTHOR]

[Display omitted] • Complete method for SPVWPS deployment; site selection & sizing. • Comprehensive user-friendly tool for SPVWPS deployment. • Case study for the SPVWPS deployment in a remote and complex terrain. • Tool's FAHP-based ranking functionality for any MCDM problem. • Optimized energy and water resource utilization with the software tool. Water resources are essential for human survival, environmental stability, sustainable development, and economic growth. However, the lack of access to clean water poses a significant challenge to people's well-being. Solar photovoltaic-powered water pumping systems offer a sustainable solution to this problem. Despite their implementation in various locations, there is currently no established methodology for optimal site selection and sizing. To address this gap, this study thoroughly investigates and analyzes the design and deployment steps of a solar PV water pumping system, including site selection and sizing of the components. A graphical user interface is developed based on fundamental equations to assist users in appropriately sizing the system for the most suitable site, aiming to minimize costs. The tool is rigorously tested and validated through a case study focusing on economically disadvantaged communities in remote areas with complex terrains. The results align with ground data, providing valuable insights into system design and deployment. This study serves as a comprehensive guide for designing solar PV water pumping systems, applicable worldwide with relevant data inputs. It contributes to the planning and implementation of sustainable solutions, addressing water access challenges through a scientific approach. [ABSTRACT FROM AUTHOR]

Time-domain NMR has been extensively utilised to study various characteristics of fluid-saturated porous,materials for instance their mobility, dynamics, stiffness, viscosity and rigidity features, particularly for solid hydrocarbons, rubbers and other polymers. As a unique time-domain technique available for over 30 years, NMR cryoporometry (NMRC) may be used to obtain pore-size distributions of the measured samples. To accurately control the sample temperature, a Peltier thermo-electrically cooled variable temperature probe has been developed and integrated with a highly compact precision NMR time-domain relaxation spectrometer, therefore providing the community with a high-performance instrument for NMR Cryoporometry. To extend the application of aforementioned high-performance NMRC instrument into more senarios, we designed a series of light-weight, compact and integral models with optional NMR frequencies from 12 MHz up to 23 MHz. The measured sample temperature can be precisely controlled from about −60 °C to +80 °C, with an excellent temperature resolution of 10 mK or better near the probe liquid bulk melting point. Therefore, it offers a fairly wide NMRC pore-size distribution ranging from about 1 nm to 2 μ m by using water as the probe liquid in the pores, significantly wider than is possible when applying generic NMR Spectrometers for NMRC. A preliminary example of NMR Cryoporometric measurements on two special cement samples is shown in the paper in which the measured pore scales as well as their repeatability are demonstrated. Furthermore, various nano-materials, such as MOF, zeolite and shale kerogen would be potential materials to study by using these new available NMRC instrument models. We aim to offer this technique as a quantitative and easy-to-apply unitary benck-top tool for an even wider range of porous material. [Display omitted] • A unitary table-top NMR Cryoporometer with thermo-electric cooling down to -60C and heating up to +80C. • Measures pore-size distributions from about 1 nm to over 1 µm, with 10mK precision near the probe liquid bulk melting point. • A built-in palm-top precision compact time-domain NMR relaxation spectrometer offering good sensitivity (able to study low pore-volume samples) and fast recovery (able to study brittle ice and and hydrocarbons in pores). • Able to study low pore-volume samples and brittle ice and and hydrocarbons in pores. • Measures over a wider pore range and to better precision than is usually available in generic NMR Spectrometers. • With intuitive Graphical User Interfaces for control. • This instrument should be of significant interest to researchers interested in studying pore-size distributions in porous materials. [ABSTRACT FROM AUTHOR]

• An Artificial neural network model was developed for predicting contact area of the tyre. • Considered contact area of 4 different bias-ply tractor tyres for model development. • A tkinter Graphical User Interface was built for easy interaction with the user. • The developed model was deployed in GUI. • The developed model coupled with GUI shows good predictability. An Artificial Neural Network (ANN) model was developed for predicting the contact area and ground pressure of bias-ply tractor tyres. A Graphical User Interface (GUI) using Tkinter was created to deploy the developed ANN model. A total of 538 datasets for the 12.4–28, 13.6–28, 14.9–28 and 16.9–28 tyres by varying load on tyres (600 to 1950 kg) and inflation pressure (69 to 234 kPa) were collected from the experimental trials and divided in a ratio of 70:30 for the training and testing of the ANN model, respectively. Another 50 new datasets of all tyres were used for validation purpose. The nominal width and overall diameter of tyre, normal load, and inflation pressure were considered as input parameters for the developed ANN model and the contact area of the tyre was taken as the output parameter. All the input parameters significantly affected the contact area of the tyre, as evident from the sensitivity analysis. The well-trained ANN model coupled with GUI predicted contact area and ground pressure beneath the tyre on firm surface at any inflation pressure and load with a maximum deviation of 2 % from the actual values measured as compared to 10 % and −9% in case of Komandi model. [ABSTRACT FROM AUTHOR]

To design, construct, and evaluate a system for assisted placement of brachytherapy applicators using electromagnetic tracking (EMT) technology that has been registered to CT or MRI images. The system provides real-time localization of needles during the insertion process, a three-dimensional display of planned needle sites, visibility of the anatomy and needle position during placement, and reference tracking to account for generator or target anatomy shifts. Such a system might be used to reduce brachytherapy procedure times, improve correspondence between intended and actual needle positions, or decrease the trainee learning curve. The system is notable for the following features: 1) Real-time visual and quantitative feedback of needle placement with respect to the underlying anatomy, as visualized by MRI or CT image, without continuous or repeated imaging 2) Pre-planning capability with a graphical overlay of target needle trajectories 3) Reference tracking to account for electromagnetic field generator or target anatomy shifts 4) DICOM-coordinate digital reconstruction of applicator locations for treatment planning and/or pre-treatment quality assurance 5) Compatibility with standard brachytherapy workflows including fixed table CT and MRI systems, procedures within or outside of a brachy suite, and insertion of needles in dorsal lithotomy position 6) Consists of commercially available EMT technology components The system was constructed using an Aurora (Northern Digital Instruments, Waterloo, ON) planar 20 × 20 cm2 field generator (EFG), System Control Unit, Sensor Interface Unit, and 6DOF and 5DOF Flextube sensor tools. The graphical user interface was written as a Matlab application with native functions and toolboxes. EMT-to-DICOM registration was based upon intracavitary applicators (e.g., tandem and ovoids), placed prior to imaging and digitized automatically using thresholding methods. The EFG was positioned above the pelvis (Figure 1A), EMT sensors were translated through the tandem and ovoid channels, and the EMT system was registered to the DICOM image set using an iterative closest-point algorithm. Next, a 5DOF EMT sensor was loaded into the distal inner lumen of a brachytherapy needle for placement. The system display provides axial, coronal, sagittal, and 3D-volumetric CT/MRI views. Proof-of concept and system accuracy were evaluated in phantom and human cadaver by comparing EM-tracked needle positions with ground-truth, post-implant CTs. Proof of concept was demonstrated for EMT-assisted placement of brachytherapy needles in a realistic clinical environment and on a brachy suite CT table. Figure 1B provides an example of how a pre-planned needle location (blue) can be visualized alongside real-time needle placement (red) to provide feedback to the user. The left-hand panel in Figure 1B shows an initial attempt where a guided needle is not in the intended location. The right-hand panel shows placement after readjustment. Accuracy in phantom (mean ± standard deviation) was 0.76 ± 0.13mm for needle tips placed up to 75mm from the tandem/ovoids and 0.52 ± 0.27mm for needle shafts at distances up to 100mm from the tandem/ovoids. Figure 1C illustrates the human cadaver EM-tracked needle positions and manually digitized positions from a post-implant CT scan. Tip and shaft accuracies were 0.77 ± 0.14mm and 0.40 ± 0.21mm, respectively. An EMT-based guidance system provided sub-millimeter accuracy for the placement of brachytherapy needles without repeated or continuous imaging. The technology can be used to reduce brachytherapy procedure times, improve the correspondence between intended and actual needle positions, or decrease the trainee learning curve. [ABSTRACT FROM AUTHOR]

Palliative radiotherapy (RT) for oncologic emergencies poses distinct patient safety challenges due to limited access to standard clinical resources beyond normal treatment hours. This clinic has historically utilized a vendor-provided unplanned treatment mode (UTM) available on the Linac console to deliver emergent RT after-hours. However, this workflow necessitates significant user intervention and is susceptible to a heightened risk of mistreatment given the condensed timeline, complexity, and infrequency of care. As such, an automated simulation-free platform was developed to better assist on-call teams in planning and delivering expedited procedures. The practical development of this tool and initial observations are described herein. The in-house platform was designed using a C#-based application programming interface (API), enabling access to scripting tools integrated directly with the planning system. Generated scripts were compiled as a stand-alone executable, and a graphical user interface was customized via an integrated development environment. The software is designed as a framework for accelerated cone-beam CT (CBCT)-based RT (no standard simulation image acquired), and streamlined workflows within the tool adhere to all processes within standard planning procedures as the application systematically guides users through each step of the RT process. When achievable, full automation is utilized to limit user intervention (e.g., for plan generation, documentation, physics second check) supplemented by detailed graphical instructions for tasks requiring manual execution. Upon completing each step, an automatic audit is conducted to ensure all needed tasks have been performed prior to proceeding. Over several months, numerous phantom-based dry runs (for palliative spine RT) were conducted to refine this tool and quantify relevant timing metrics. Dry runs indicate the entire RT workflow can be completed in under an hour (~50 min on average) with use of the in-house tool, compared to 1.5-3.5 hours for similar treatments with UTM. The software generates satisfactory multi-field CBCT-based 3D treatment plans (for spine) in ~20 seconds following definition of the desired treatment area, whereas manual plan generation on 2D radiographs with UTM (based on hand calculations) requires ~10 min. Overall, therapist/physician feedback indicates the platform is far less complex than UTM and requires fewer steps to accomplish the identical task of emergent RT delivery. A novel in-house platform for emergent RT was successfully developed utilizing an API and is undergoing clinical implementation pending departmental approval. By introducing full automation when feasible, this application can improve the quality and safety of after-hours RT while greatly reducing workflow duration. Additional work has commenced to extend use of this platform to more complex treatment techniques and sites beyond emergent care. [ABSTRACT FROM AUTHOR]

Monte Carlo (MC) based treatment planning solutions for FLASH have typically been confined within proprietary treatment planning systems (TPS), lacking a true open-source implementation. This study aims to establish an open-source treatment planning workflow by integrating the Tool for Particle Simulation (TOPAS) MC code with a Python graphical user interface (GUI) to accurately calculate 3D dose across various electron cut-out apertures and geometries. The treatment head of the FLASH-enabled Linac was modeled using CAD files. Depth dose and beam profiles were acquired using radiographic film with different circular collimators. A beam model was created through inverse Bayesian optimization, minimizing the difference between measured and simulated profiles. The total particle count for each radiation pulse passing through the beam current transformer (BCT) on the mobile electron linear accelerator was estimated from measurements and then the BCT was integrated into the MC simulation, establishing a calibration relationship between MC output, absolute dose and number of histories. The whole workflow was then used for end-to-end testing incorporating an anatomically correct 3D printed mouse phantom. Inverse optimization yielded a beam energy of 8.4 MeV with a spread of 1 MeV and a source spot size of 2 mm with an angular spread of 4 degrees. The simulated and measured depth dose for the 10 cm collimator agreed to within <1% of each other. Simulated beam profiles for various insert sizes exhibited good agreement (<3%) with the measured profiles, although variations were noted in the low dose region for all profiles. A Python GUI was developed enabling the importation of CT scans and beam apertures to calculate 3D dose distributions by invoking the TOPAS MC code for simulation. A recent multi-institutional dosimetric audit uncovered discrepancies of up to 10%, particularly in areas with tissue inhomogeneities. Given the tissue-protective benefits of FLASH radiation are on the order of 10-20%, it is imperative to employ precise planning and delivery tools. Our open-source MC beam model and Python GUI will be invaluable resources for ensuring accuracy in FLASH research, with the potential for seamless adaptation into clinical trials. [ABSTRACT FROM AUTHOR]

• Well-known and advanced feature selection methods within a unified framework have been implemented. • Filter, wrapper, and embedded approaches are considered in the tool. • Essential auxiliary tools for performance evaluation, result visualization, and statistical analysis have been provided. • Researchers can use the tool through its graphical user interface or as a library in their Java codes. • UniFeat has been completely implemented in Java and can be run on a wide range of platforms. The Universal Feature Selection Tool (UniFeat) is an open-source tool developed entirely in Java for performing feature selection processes in various research areas. It provides a set of well-known and advanced feature selection methods within its significant auxiliary tools. This allows users to compare the performance of feature selection methods. Moreover, due to the open-source nature of UniFeat, researchers can use and modify it in their research, which facilitates the rapid development of new feature selection algorithms. [ABSTRACT FROM AUTHOR]

The research aimed to test the influence on text entry efficiency of supporting features in gaze text entry systems (aka gaze keyboards), leading to dynamic changes in key layout and size. We examined three gaze-controlled on-screen keyboards, that differ in the degree of support, which translates into the amount of change that occurs in the arrangement of the keys and their size - three degrees of dynamism of keyboards' graphical user interface. We also attempted to examine the importance of user attention while using dynamic interface and its impact on text input performance. [ABSTRACT FROM AUTHOR]

• RELAX is a fully automated EEG cleaning pipeline, freely available on GitHub, and easy to use through a graphical user interface. • RELAX provided amongst the best performance out of a range of common EEG pre-processing pipelines at cleaning all artifact types. • RELAX provided high values for the variance explained by the difference between outcomes in common experimental manipulations. Electroencephalographic (EEG) data are often contaminated with non-neural artifacts which can confound experimental results. Current artifact cleaning approaches often require costly manual input. Our aim was to provide a fully automated EEG cleaning pipeline that addresses all artifact types and improves measurement of EEG outcomes We developed RELAX (the Reduction of Electroencephalographic Artifacts). RELAX cleans continuous data using Multi-channel Wiener filtering [MWF] and/or wavelet enhanced independent component analysis [wICA] applied to artifacts identified by ICLabel [wICA_ICLabel]). Several versions of RELAX were compared using three datasets (N = 213, 60 and 23 respectively) against six commonly used pipelines across a range of artifact cleaning metrics, including measures of remaining blink and muscle activity, and the variance explained by experimental manipulations after cleaning. RELAX with MWF and wICA_ICLabel showed amongst the best performance at cleaning blink and muscle artifacts while preserving neural signal. RELAX with wICA_ICLabel only may perform better at differentiating alpha oscillations between working memory conditions. RELAX provides automated, objective and high-performing EEG cleaning, is easy to use, and freely available on GitHub. We recommend RELAX for data cleaning across EEG studies to reduce artifact confounds, improve outcome measurement and improve inter-study consistency. [ABSTRACT FROM AUTHOR]

A Bayesian web-based Decision Support System (DSS) was designed and developed to assist various stakeholders in improving biophysical and socio-economic conditions in a stressed and fragile watershed by systematically evaluating current watershed characteristics, introducing feasible solutions as well as choosing the best ones. The need to deliver a flexible DSS has been the motivation to look for a decision toolbox accommodating the concerns and priorities of a wide range of stakeholders as the potential users. This assists the users to make informed decisions based on quantified potential outcomes. The DSS main components include a database management system, a model base management system, multi-attribute decision-making models, and a graphical user interface. This prototype DSS provides a brief introduction of a pilot watershed characterized with semi-arid conditions in the Mid-north of Iran and then presents the problem structure using the DPSIR approach. A set of vegetation-based solutions and their spatial distribution considering some scenario rules are provided to the DSS users. The various impacts of the proposed solutions are predicted using several models in the Bayesian network modeling context integrating qualitative and quantitative knowledge and data. The DSS developed in this study facilitates knowledge transfer and involvement of stakeholders including practitioners and local communities in the decision making process. This DSS has been designed into a web-based and open-source framework for easier and inclusive access. The DSS can be tailored to fit a broader class of environmental problems. • The Bayesian network is used to integrate different disciplinary models and multiple sources of data and information. • Through using the DPSIR approach, the DSS assists in understanding and analyzing the structure of problems. • The DSS allows the testing of alternative management scenarios and predicting their impacts. • The web-based DSS assists users to engage in the management of environmental problems at a watershed scale. • The DSS provides an easy-to-use graphical user interface for non-technical users facilitating informed decision making. [ABSTRACT FROM AUTHOR]

• Introduces the largest waste dataset, comprising 35,264 images. • Parallel lightweight depth-wise separable CNN (DP-CNN) feature extractor. • Proposed a novel three-stage waste classification using Ensemble ELM. • Model interpretability is highlighted using various Explainable AI. • Prototype hardware system with GUI for real-time waste categorization. The growing negative effects caused by inadequate waste processing have led to the widespread implementation of waste classification systems. One effective approach is to develop an automated classification system that uses advanced waste recognition technology. This method can decrease the amount of manual labor required for waste separation and recycling activities. In the present study, a novel three-stage waste classification system was proposed. It incorporates the parallel lightweight depth-wise separable convolutional neural network (DP-CNN) in conjunction with the ensemble extreme learning machine (En-ELM) classifier. Waste items are first classified into two main categories: biodegradable and non-biodegradable. The dataset is then split into nine distinct categories in the second stage based on the overall waste characteristics. The final stage of the classification process involves a more detailed granularity, as all images are assigned to one of thirty-six specific classes. With an average accuracy, precision, recall, f1, and ROC-AUC values of 96 %, 95.0 ± 0.02 %, 95.0 ± 0.02 %, 95.0 ± 0.02 %, and 98.77 %, respectively, the proposed model demonstrates promising results in the first stage of two-class classification. Advancing to the second stage, the framework showed excellent results in nine-class classification, with performance rates of 91.00 %, 90.0 ± 0.04 %, 89.44 ± 0.06 %, 89.66 ± 0.05 %, and 98.57 %, respectively. Similar to the previous stages, the model continued to perform effectively in the third stage, achieving 85.25 % accuracy, 85.02 % precision, 85.25 % recall, 84.54 % f1-score, and 98.68 % AUC in the thirty-six-class classification. The En-ELM classifier, a fusion of pseudoinverse ELM (PI-ELM) and L1 regularized ELM (L1-RELM), achieved impressive results with an average testing time of 0.00001 s. A novel comprehensive dataset titled TriCascade WasteImage, which combines four smaller preexisting datasets, was used to measure the performance of the DP-CNN-En-ELM model. With only nine layers and 1.09 million parameters, the proposed approach precisely extracts pertinent information from images to classify diverse waste materials. The effectiveness of the model is confirmed by comparing it to advanced transfer learning methods. Various explainable AI (XAI) methods are used to explore the interpretability and decision-making capability of the proposed model. Additionally, this study presents a comprehensive prototype hardware architecture for rapid waste categorization in an augmented environment, enabling autonomous waste sorting in industrial applications. [ABSTRACT FROM AUTHOR]

Fabric Reinforced Cementitious Matrix (FRCM) is an effective intervention method for improving the shear strength of existing reinforced concrete (RC) beams, yet predictive analyses are scarce. This study introduces and compares nine machine learning (ML) models to estimate the shear capacity of FRCM-strengthened RC beams, utilizing a dataset characterized by 14 distinct variables. Among these, the Extreme Gradient Boosting (XGBoost) model distinguishes itself with the highest accuracy, with R² , MAE, RMSE, and CoV values of 98.98 %, 5.36 kN, 10.07 kN, and 9.75 %, surpassing traditional empirical and mechanical formulations. Additionally, robustness analysis across random dataset splits further confirms the stability of the XGBoost model's predictions. The superiority of XGBoost and other ML models over conventional approaches is further validated through Taylor diagrams, which demonstrate closer alignment with experimental outcomes. Shapley Additive Explanations (SHAP) analysis identifies key influential factors, including beam depth, concrete compressive strength, stirrup reinforcement ratio, and thickness of mortar, while Partial Dependence Plots (PDP) reveal their marginal effects on shear strength. These insights are crucial for enhancing design practices and guidelines for FRCM-strengthened beams. A practical graphical user interface (GUI) tool was also developed, facilitating direct application of findings in the study. • 9 ML models are evaluated to predict the shear capacity of FRCM strengthened beams. • ML approaches show superior precision versus existing models in comparative analysis. • XGBoost model significantly reduces prediction errors. • SHAP analysis highlights key factors in XGBoost's predictions. [ABSTRACT FROM AUTHOR]

This article presents a new, versatile, user-friendly, and seamlessly integrated modeling platform for guided wave propagation in elastic waveguides of arbitrary cross-sections. The software exploits the semi-analytical finite element (SAFE) method for the calculation of wave-propagation characteristics. The graphical user interface (GUI) allows for the modeling of guided waves in plate-like, cylindrical, and arbitrary cross-sectional waveguides considering piezoelectric effects. The frequency-dependent mode shapes for two-dimensional waveguide cross sections are displayed in the form of vivid harmonic motion animations. A structural cross-sectional scanning system is seamlessly integrated with the modeling platform as an initiative for tunneling the digital model with the physical domain. This may open a new paradigm for future physical-virtual interfacing supercomputing within the nondestructive evaluation (NDE) and structural health monitoring (SHM) communities. Finally, experimental verifications are conducted on a rail structure via the Scanning Laser Doppler Vibrometry (SLDV). The measured wavefields corroborate the perfect superposition between the computational and practical wave dispersive features in a free rail. The paper finishes with a summary, concluding remarks, and suggestions for future work. • Physical-virtual interfacing is realized with a structural cross-sectional scanning system. • A GUI Platform for modeling wave propagation in waveguides of arbitrary cross-sections. • Piezoelectric effects are implemented in the semi-analytical finite element formulation. • Numerical simulations and experimental measurements compared well with one another. [ABSTRACT FROM AUTHOR]

[Display omitted] • Performance evaluation and validation of the COMPASS software. • Focus on fasted intake of immediate-release oral drug products. • Rapid and straightforward assessment of the bioequivalence probability. • Extensive investigation of the variable gastric emptying kinetics influence. • Comparable predictions as complex virtual bioequivalence trials. COMPASS is a novel tool for predicting the in vivo similarity of immediate-release oral dosage forms. It bridges the gap between straightforward but purely mathematical f2 factor-based assessment of dissolution profiles and complex physiologically-based pharmacokinetic simulations. This Python-based program uses semi-mechanistic modeling to evaluate the relative differences between key pharmacokinetic endpoint parameters − C max and AUC − at variable gastric emptying rate (kGE) and the housekeeping wave time (GET) under fasting conditions. Through a graphical interface the user uploads dissolution results and fits a modified Noyes-Whitney model. Then, after providing the pharmacokinetic absorption and disposition parameters (effective permeability, clearances and distribution volumes) of a given drug, COMPASS generates an interactive report with predicted differences for C max and AUC as a function of kGE and GET (3D surface plot), as well as the full plasma concentration–time profiles. Using a dataset comprising 15 formulations (reference and test pairs) containing 11 different drug substances (BCS class I − IV), we determined the measures for discriminating between bioequivalent vs. nonbioequivalent test and reference pairs. We also found that COMPASS-based bioequivalence predictions were similar as those from physiologically-based pharmacokinetic models built in commercially available software. Therefore, COMPASS can support the development of immediate-release dosage forms by fast screening of formulations for bioequivalence/non-bioequivalence, particularly in the early stages, thereby boosting the success rate of clinical trials. [ABSTRACT FROM AUTHOR]

• Comprehensive optimization method of full-color holographic stereogram printing. • Single-shot depth estimation for proper 3D visualization of real objects. • Full-color hogel via integral imaging and unique non-iterative phase modulation. • Time-controlled exposure of full-color hogel using a single spatial light modulator. • Automated holographic printing with a custom-designed user interface. A comprehensive optimization method for a full-color holographic stereogram (HS) printing system based on single-shot depth estimation for real-world objects and time-controlled exposure is proposed. Both processing steps, including digital content generation and optical printing, are optimized to ensure possible high-quality three-dimensional (3D) holographic image printing, rapid computation, and proper full-color visualization. In the digital content generation, first a high-resolution two-dimensional (2D) image of the real object is captured, and its depth map is then estimated via a pre-trained convolutional neural network (CNN) model, ensuring an identical resolution with a given 2D image. As a post-processing, the unnecessary scenes and background are removed from the captured color image, without losing the main information of a primary object. Then, a hogel array (HA) is obtained by utilizing the estimated depth map and a post-processed color image through a fast inverse-directed propagation (IDP) method. Each hogel undergoes unique non-iterative phase modulation in a quite short time without the degradation of image quality while the chromatic dispersion errors are minimized. Finally, the hogels are sequentially printed onto holographic material using a time-controlled exposure, to provide the color-balanced full-color reconstruction using a single spatial light modulator (SLM). The overall procedure is seamlessly performed automatically via custom-designed graphical user interface. This study experimentally confirmed a simple and effective optimization for HS printing systems in both digital content generation and optical printing unit. [ABSTRACT FROM AUTHOR]

Graph convolutional networks (GCNs) are an effective skeleton-based human action recognition (HAR) technique. GCNs enable the specification of CNNs to a non-Euclidean frame that is more flexible. The previous GCN-based models still have a lot of issues: (I) The graph structure is the same for all model layers and input data. GCN model's hierarchical structure and human action recognition input diversity make this a problematic approach; (II) Bone length and orientation are understudied due to their significance and variance in HAR. For this purpose, we introduce an Extended Multi-stream Temporal-attention Adaptive GCN (EMS-TAGCN). By training the network topology of the proposed model either consistently or independently according to the input data, this data-based technique makes graphs more flexible and faster to adapt to a new dataset. A spatial, temporal, and channel attention module helps the adaptive graph convolutional layer focus on joints, frames, and features. Hence, a multi-stream framework representing bones, joints, and their motion enhances recognition accuracy. Our proposed model outperforms the NTU RGBD for CS and CV by 0.6% and 1.4%, respectively, while Kinetics-skeleton Top-1 and Top-5 are 1.4% improved, UCF-101 has improved 2.34% accuracy and HMDB-51 dataset has significantly improved 1.8% accuracy. According to the results, our model has performed better than the other models. Our model consistently outperformed other models, and the results were statistically significant that demonstrating the superiority of our model for the task of HAR and its ability to provide the most reliable and accurate results. • Skeleton HAR using GCNs has excellent results, but problems in earlier GCN models. • EMS-TAGCN model for HAR can train network topology based on input data. • The proposed model focuses on joints, frames, and features, improving accuracy. • NTU CS = 0.6% and CV = 1.4%, Kinetics- Top-1 and Top-5 are 1.4% improved. [ABSTRACT FROM AUTHOR]

FSUDAQ is a versatile, multi-threaded, lightweight, and open-source data acquisition software with a graphical user interface, designed to fully utilize the capabilities of first-generation CAEN x725, x730, and x740 series digitizers equipped with various Digital Pulse Processing (DPP) firmware, including Pulse-Height Analysis (PHA), Pulse-Shape Discrimination (PSD), and Charge-Digital Conversion (QDC). It emphasizes user-friendliness, stability, scalability, high throughput, and low latency. The software includes features such as an online waveform scope, scalar panel, and real-time single spectrum display for each input channel, along with an online event builder and analyzer capable of generating 1D and 2D histograms and applying graphical cuts. Users can also create and integrate custom online analyzers to meet specific experimental requirements. FSUDAQ has been successfully tested at the John D. Fox laboratory at Florida State University (FSU) with a diverse set of experiments using the Encore and ANASEN active target detectors, the Super-Enge Split-Pole Spectrograph, and the CATRiNA neutron detectors. In terms of performance, FSUDAQ can handle up to approximately 500k triggers per second per channel without waveform recording, or data rates of around 65 MB/s per optical fiber, with or without waveform recording. [ABSTRACT FROM AUTHOR]

Veterinary surveillance frequently requires study design for freedom-from-disease testing, specifying a sample size to balance higher statistical power with larger sample sizes against increased research and ethics costs, with the recognition that tests can generate false positive and negative results: i.e., tests exhibit imperfect sensitivity and specificity. In this paper, we revisit the mathematics behind exact calculations of sample size in terms of the binomial and hypergeometric distributions, and present a new algorithm – implemented and available to use in R as a Shiny application with a graphical user interface – to determine sample size for practical situations. Often, sample size calculations are based upon simulations or approximations, but we show here that exact calculations are feasible. In addition, we relax the liberal assumption – which provides conservative sample-size estimates – that sensitivity and specificity are known exactly, and instead assume both are Beta distributed with known hyperparameters. The application presented here was originally designed as a learning tool for students and is now made available for wider use. [ABSTRACT FROM AUTHOR]

Phosphate (PO 4 (III)) contamination in water bodies poses significant environmental challenges, necessitating efficient and accurate methods to predict and optimize its removal. The current study addresses this issue by predicting the adsorption capacity of PO 4 (III) ions onto biochar-based materials using five probabilistic machine learning models: eXtreme Gradient Boosting LSS (XGBoostLSS), Natural Gradient Boosting, Bayesian Neural Networks (NN), Probabilistic NN, and Monte-Carlo Dropout NN. Utilizing a dataset of 2952 data points with 16 inputs, XGBoostLSS demonstrated the highest R2 (0.95) on new adsorbents. SHapely Additive exPlanations analysis showed that adsorption experimental conditions had the most significant impact (43%), followed by synthesis conditions (29%) and adsorbent characteristics (28%). Optimized conditions included an initial PO 4 (III) concentration of 125 mg/L, carbon content of 11.5%, oxygen content of 23%, a contact time of 1440 min, a heating rate of 5 °C/min, 200 rpm, and a surface area of 410 m2/g, using Ra-LDO adsorbent synthesized from rape cabbage feedstock. This study developed and presented a practical online framework for predicting PO 4 (III) removal onto biochar using a web-based graphical user interface. [Display omitted] • 5014 data-points with 16 features on PO 4 (III) adsorption onto 132 different biochars. • Five ML models were used to predict the PO 4 (III) adsorption capacity onto biochar. • ML findings suggest that the NGBoostLSS model outperformed other models. • SHAP analysis proposed that the C i , C (%) and O (%) were the most controlling inputs. • A web-based GUI was developed for wider application of developed ML models. [ABSTRACT FROM AUTHOR]

This study investigates the free vibration and transient dynamic response of functionally graded material (FGM) sandwich plates with stepped face sheets (FGM-SPSFS) supported by viscoelastic foundation under blast loading. The research focuses on the effects of geometric configurations and material property variations across segments. Each plate comprises three layers: a homogeneous hard core and two FGM face sheets, divided horizontally into two segments with differing face sheet thicknesses, which enhance structural stiffness while maintaining a consistent total thickness. The material properties of the sandwich plates follow a power-law distribution. The formulations are based on higher-order shear deformation plate theory and von Kármán geometric nonlinearity, and are solved using Galerkin's method. Validation is achieved by comparing the results with published literature and finite element analysis (FEA). Artificial neural network (ANN) models are developed to predict natural frequencies without extensive computational runs, employing Bayesian Regularization (BR) and Levenberg–Marquardt (LM) algorithms in MATLAB. A new graphical user interface (GUI) tool facilitates frequency predictions using the proposed ANN model. Key findings indicate that modifications to the stepped face sheets and core layers affect stiffness, natural frequency, and vibration amplitudes. Increasing the core-to-total thickness ratio enhances stiffness, resulting in higher frequencies and reduced displacement amplitudes. The LM algorithm outperforms the BR algorithm, with errors generally below 1%, compared to 2% to 4% for BR with the log-sigmoid function. This study offers valuable insights into the design and analysis of FGM sandwich structures for engineering applications. • Higher-order shear deformation theory and Galerkin method are applied for structural analysis. • The effect of stepped face sheets on structures under blast loading is thoroughly investigated. • Increasing the core-to-thickness ratio raises natural frequencies while reducing displacement amplitudes. • Artificial Neural Network (ANN) models are developed to accurately predict the natural frequencies of plates. • A Graphical User Interface (GUI) tool leveraging ANN models is created for frequency predictions. [ABSTRACT FROM AUTHOR]

[Display omitted] • A graph-learning-based BANE model improves the prediction performances. • The BANE-XGBoost has the higher performance improvement and best prediction. • Illumination intensity and KH 2 PO 4 (TOP 2) dominate the growth of Porphyridium. • SHAP-based heatmap and group reach the combined roles of cultivation parameters. • The one-way and two-way PDP models find the optimal factors for the two targets. A graph learning [Binarized Attributed Network Embedding (BANE)] model enhances the single-target and multi-target prediction performances of random forest and eXtreme Gradient Boosting (XGBoost) by learning complex interrelationships between cultivation parameters of Porphyridium. The BANE-XGBoost has the best prediction performance (train R2 > 0.96 and test R2 > 0.87). Based on Shapley Additive Explanation (SHAP) model, illumination intensity, culture time, and KH 2 PO 4 are the most critical factors for Porphyridium growth. The combined facilitating roles of cultivation parameters are found using the SHAP value-based heat map and group. To reach high biomass and daily production rate concurrently, one-way and two-way partial dependent plots models find the optimal conditions. The top 2 critical parameters (illumination intensity and KH 2 PO 4) were selected to verify using the graphical user interface website based on the optimized model and lab experiments, respectively. This study shows the graph-learning-based model can improve prediction performance and optimize intricate low-carbon microalgal cultivation. [ABSTRACT FROM AUTHOR]

A variety of biosensors have been recently introduced as wearable devices to collect physiological data, with applications ranging from personalized medicine and point-of-care diagnostics to home and fitness monitoring, among others, garnering substantial interest. This interest has been fueled by the increasing demand for ubiquitous, continuous, and pervasive vital signs monitoring, coupled with advancements in biosensor technology and IoT-enabled capabilities. Existing research studies have only relied on a limited number of health- and physiological-related indicators (thus, do not offer a comprehensive health monitoring and assessment system) due to the technical difficulties to integrate multiple sensors. In fact, the issues of multimodality, heterogeneity, and complexity of data as well as the interoperability among sensors make it challenging to seamlessly integrate multiple sensors into one system. This study overcame these technical challenges by leveraging multi-sensor fusion capabilities to develop an intelligent, IoT-enabled wearable multi-modal biosensing device and cloud-based digital dashboard for real-time, comprehensive health, physiological, emotional, and cognitive monitoring. First, 18 different health- and physiological-related indicators were identified. Second, 14 different sensors were used to acquire the entire data for the 18 different indicators using a hardware sensing system designed using four ESP32 microcontroller boards integrated with Wi-Fi and Bluetooth connectivity by fusing the various data from the 14 different sensors. Third, the designed system was developed as a wearable device that can be installed on the hip as well as the right and left feet using 3D printed parts. Fourth, a web-based digital dashboard was created onan edge computing server that was hosted on a microprocessor to instantly publish the data, and a graphical user interface (GUI) was developed to provide intuitive and real-time visualization of the various health-related indicators using the Django and JavaScript-based React.js web development frameworks. The accuracy of the developed IoT-enabled biosensing system was tested and validated by benchmarking and comparing the obtained results from the proposed system with those aquired from various commercially used sensors. The validation outcomes reflected that the proposed system achieved an accuracy of more than 90 % for most of the 18 considered indicators and an accuracy greater than 85 % for all indicators. This study adds to the body of knowledge by being the first research capable of reporting the following 18 indicators into a single biosensing system in real-time: Electrocardiogram (ECG or EKG), Electroencephalogram (EEG), Electrooculogram (EOG), Electromyography (EMG), Photoplethysmography (PPG), heart rate (HR), heart rate variability (HRV), respiratory rate (RR), skin temperature (ST), skin humidity (SH), blood glucose level (BGL), blood pressure (BP), oxygen saturation (SpO2), body weight pressure (BWP), body motion (BM), electrodermal activity (EDA), galvanic skin response (GSR), and skin conductance responses (SCR). The proposed system provides rich information on various vital signs and could be used for a wide window of applications, including monitoring and assessing health status; emotional and arousal status; mental and cognitive status; behavioral, physical, and attention status; and physiological status. The developed system is not specific to a particular industry but rather could be used for any sector of interest. This paper lays the ground to significant advancements in wearable sensor technology, data visualization techniques, and health monitoring practices. [Display omitted] • An IoT-enabled, wearable health and physiological monitoring system is proposed. • The work measures 18 different health-related indicators by integrating 14 sensors. • The system is based on multimodal hardware devices and a digital dashboard. • The work supports real-time multi-sensor data fusion methods for health decisions. [ABSTRACT FROM AUTHOR]

This study presents a nonmodel-based machine learning framework for estimating engineering demand parameters (EDPs) of eccentrically braced frames with soil-structure interaction effects. The objective is to estimate residual and peak story drift ratio, peak floor acceleration, and develop fragility curves using traditional regression equations and advanced machine-learning techniques. Correction coefficients are developed to improve prediction accuracy by accounting for soil-structure interaction. A comprehensive database, including incremental dynamic analysis results of 4- and 8-story frames, is developed, consisting of 109,841 data points. The database includes fixed-base models and models with various soil-structure interaction values, subjected to 44 far-field ground motions. Four scenarios are introduced considering various input variables to compare the impact of soil-structure interaction. Findings reveal the effects of soil-structure interaction features on the performance of machine learning algorithms, increasing by up to 17.61% of the coefficient of determination. Utilizing the predicted story drift ratio, two types of fragility curves indicate more precise predictions, emphasizing the impact of soil-structure interaction effects at lower damage levels. A graphical user interface has been developed to predict fragility curves based on various inputs to promote the practical use of machine learning in engineering. Two new 4-story frames are used as case studies, subjected to unseen ground motions to assess the application of trained machine learning algorithms. Prediction errors in input-output scenarios considering soil-structure interaction range from 3% to 18% for new frames. The proposed approach for predicting EDPs is further acknowledged by evaluating a real instrumented five-story steel frame office building. • Integrates system identification and wavelet-based rDSF using only roof acceleration through an output-only scenario. • Utilizes machine learning techniques and regression equations to evaluate the effect of SSI features for EBF buildings. • Proposes correction coefficients for fixed-base equations to achieve more realistic EDPs considering SSI. • Mapps the acceleration- and rDSF -based fragility curves for EBFs using the best ML algorithm. • Develops a GUI for predicting fragility curves based on Sa(ave) and rDSF variables. [ABSTRACT FROM AUTHOR]

The use of prefabricated vertical drain (PVD) in conjunction with vacuum and heat preloading is an effective improvement approach for dredged marine slurry. Despite it being a typical large-strain thermal-hydro-mechanical (THM) problem, there is currently a lack of reliable large-strain analysis methods for this problem. This study therefore develops a large-strain radial consolidation model, considering a thermal elastoplastic constitutive model of soft clays, self-weight of dredged slurry, radial and vertical flows, nonlinear hydraulic conductivity, nonlinear compressibility during the consolidation process, heat conduction, and heat convection process. The modified alternative direction implicit (ADI) difference approach is used to solve the proposed model. The numerical algorithm is then verified by degraded verification with a well-established radial-large strain consolidation model. A laboratory experiment has been employed to show the effectiveness of the proposed model in predicting the consolidation behavior. A graphical user interface (GUI) has been programmed for potential users based on the proposed model and numerical algorithm. The results indicate that heat preloading improves the permeability of soil, resulting in a larger settlement of soil. Heat preloading serves as an effective method to enhance the efficiency of PVD. [ABSTRACT FROM AUTHOR]

Weather monitoring and forecasting plays a vital role in a great variety of human activities such as agriculture, transportation, and extreme weather phenomena. This study presents the first outcomes of the development of a fully automated system regarding the real-time recording of basic meteorological parameters and their short-range forecasting (nowcasting). The system itself is divided into five core components: a hardware system for monitoring atmospheric conditions (Commercial Off-The-Shelf structures), a system for storing and managing data, a module for distributing data to support applications, a machine learning algorithm for nowcasting, and a user-friendly interface, all made by modern tools and methods, described analytically. Finally, the nowcasting procedure along with the relative accuracy results, is presented. The nowcasting procedure is based on a Long Short-Term Memory (LSTM) model scheme which is parametrized in such a way that reliable forecasts, up to 2 h ahead of time, can be provided. • The proposed system refers to automated weather monitoring and nowcasting. • The nowcasting procedure provides reliable estimations up to 2 h ahead of time. • The autonomous system includes modern design of the backend and frontend environment. • The users can have easy access to the graphical user interface and download daily data. [ABSTRACT FROM AUTHOR]

The phase-field method has become a useful tool for the simulation of classical metallurgical phase transformations as well as other phenomena related to materials science. The thermodynamic consistency that forms the basis of these formulations lends to its strong predictive capabilities and utility. However, a strong impediment to the usage of the method for typical applied problems of industrial and academic relevance is the significant overhead with regard to the code development and know-how required for quantitative model formulations. In this paper, we report the development of an open-source phase-field software stack that contains generic formulations for the simulation of multiphase and multi-component phase transformations. The solvers incorporate thermodynamic coupling that allows the realization of simulations with real alloys in scenarios directly relevant to the materials industry. Further, the solvers utilize parallelization strategies using either multiple CPUs or GPUs to provide cross-platform portability and usability on available supercomputing machines. Finally, the solver stack also contains a graphical user interface to gradually introduce the usage of the software. The user interface also provides a collection of post-processing tools that allow the estimation of useful metrics related to microstructural evolution. [Display omitted] [ABSTRACT FROM AUTHOR]

This paper proposes a novel procedure for prediction of both load-displacement curve and load-carrying capacity of concrete-filled double-skin steel tube (CFDST) columns under uniaxial compression by using convolutional neural network (CNN)-based regression and Nelder-Mead methods. Firstly, hybrid databases collected from experiments in literature and generated from finite element analyses are employed to build the proposed CNN-based model. The accuracy of the proposed model is described through a comparison between predictive results of the proposed model and unseen data. Two machine learning models, including eXtreme Gradient Boosting and Multilayer Perceptron, are adopted for comparison. It can be observed that the CNN-based model provides the most accurate predictions for both the load-displacement curve and axial compression capacity of CFDST columns in both experimental and numerical databases. An efficient procedure is developed to calibrate the preliminary load-displacement curve estimated by the CNN-based model, and to notably enhance its smoothness and performance. Adjusted formulae (based on well-known equations) are obtained for predicting the load-displacement curve of CFDST columns. The hyperparameters of these formulae are optimized using the Nelder-Mead method. It is indicated that the adjusted load-displacement curves obtained from the proposed procedure outperform the preliminary curves estimated by the CNN-based model. A sensitivity analysis was conducted to investigate the model's performance in predicting the load-displacement curves of CFDST columns with variations of input variables within stochastic environments. Finally, a cloud-based graphical user interface is developed to provide a convenient tool for users to predict axial load-displacement responses of CFDST columns without prior programming knowledge. • A novel procedure for predicting load-displacement curves of CFDST columns is proposed for the first time. • CNN-based model presents superior strength prediction performance compared to XGB and MLP models. • CNN-based model is applied to accurately predict load-displacement curves of CFDST columns. • Adjusted formulae are suggested for predicting load-displacement curves of CFDST columns. • A cloud-based GUI program is made online accessible for practical design purposes. [ABSTRACT FROM AUTHOR]

Employing machine learning to predict the Pb2+ adsorption capacity of biochars is an innovative pursuit in hazardous materials. This study compared artificial neural network (ANN), support vector regression (SVR) and random forest (RF) for Pb2+ adsorption capacity by biochar from a fluidized bed system. Besides developing correlations for comparison, the RF model (R2 = 0.984, RMSE = 0.054) outperformed both ANN (R2 = 0.908, RMSE = 0.316) and SVR (R2 = 0.667) in predicting higher adsorption capacity. Based on the superior performance, the Shapley Additive Explanations (SHAP) were employed on RF. SHAP global explanations indicated that adsorption conditions contributed 69.03% and biochar characteristics contributed 30.21%to adsorption capacity, highlighting Dosage (D) and Carbon (C) as the crucial factors. Regarding biochar characteristics, element compositions contributed 76.59%. The single samples demonstrated that the final predictions align with the experimental results. The synergistic effect of dependence plot explains the Pb2+ adsorption under varying parameter conditions, such as D < 1 g/L, C<45%, Pb in >100 mg/L, H < 2.5, t > 12h, T > 25 °C, pH > 9, H/C > 0.4, the SHAP value is positive, contributing to an increase in adsorption capacity. Furthermore, a graphical user interface (GUI) leveraging SHAP model parameters predicts adsorbent performance, providing novel insights into optimizing biochars production. The obtained findings narrow the search for optimal biochars adsorbents and might guide laboratory experiments and engineering application of Pb2+ removal using biochars. [ABSTRACT FROM AUTHOR]

The high energy consumption and the large amount of emissions associated with roads life cycle makes it necessary for the administrations involved to invest in the implementation of renewable energy technologies on adjacent lands so that the net balance between CO 2 emitted and CO 2 saved is zero, thus helping to decarbonize the transport networks. To facilitate the first step in the long process leading up to the construction of wind and solar farms, a user-friendly tool is here presented that allows an early-stage evaluation of renewable energy projects. The ENROAD web-based GIS tool, through a functional, modern, intuitive, and ergonomic graphical user interface aimed at non-specialized users, optimizes the use of photovoltaic and wind technologies based on shading and wake losses, respectively, taking into account the solar radiation and wind conditions of the location and the geometry of the area selected by the user. In addition, the tool provides a thorough financial analysis, incorporating the effects of user-defined long-term interest rates, inflation rates, loan costs and efficiency losses on the Levelized Cost of Energy as well as traditional financial metrics, enabling users to conduct insightful if-then simulations. While this tool cannot have the precision of the in-depth study needed for the design of complex renewable energy facilities, it can facilitate the decision making by the administrations. Finally, a case study and a validation section have been included to demonstrate the technical and financial evaluation provided by ENROAD and ensure that the results provided are reasonable and accurate enough. • Integration of renewable energies in the vicinity of road infrastructures. • User-friendly GIS-based web tool for the evaluation of photovoltaic/wind farms. • Technical and financial assessment of renewable energy projects. [ABSTRACT FROM AUTHOR]

As intelligent interaction systems evolve, ergonomic requirements (ERs) for graphical user interfaces (GUIs) have become increasingly complex and subjective. Inadequate evaluation during the GUI prototype design phase can lead to significant challenges in later development and testing. To address this issue, we propose a novel Decision-making Trial and Evaluation Laboratory (DEMATEL) approach that integrates rough set (RS) theory, linguistic Pythagorean fuzzy set (LPFS), and models relationships among decision makers (DMs). Specifically, we introduce linguistic Pythagorean fuzzy rough numbers (LPFRNs) to capture DM groups' uncertain and interrelated opinions on ERs, accounting for linguistic uncertainty and inherent roughness. We detail the construction of LPFRNs, including operation rules and aggregation operators based on the Dombi t-norm (DTN) and t-conorm(DTCN), and establish ranking rules and distance measures. By incorporating a DM objective weight determination method based on similarity measures, we enhance the precision of the DEMATEL model. Applying this method, we construct a hierarchical system of 31 ERs based on six GUI design elements (DEs) and evaluate their importance. Our findings reveal that layout-related ERs are critically significant, providing valuable insights for GUI designers to prioritize ERs effectively. The novelty of this work lies in integrating LPFRN and its aggregation operator into the DEMATEL method, which can better meet the complex and subjective correlation evaluation. Potential applications include automated evaluation of GUI design quality and optimization of the design process. We validated the robustness and feasibility of our method through sensitivity and comparative analysis. [Display omitted] • Propose the construction process of linguistic Pythagorean fuzzy rough numbers (LPFRNs). • Present Dombi t-conorm and t-norm operations under the LPFRN environment. • Develop Dombi weighted averaging and geometric aggregation operators of LPFRNs. • A novel DEMATEL model is presented to solve the evaluation of ergonomic requirements with LPFRN data. • The decision maker's subjective and objective weights are combined in proposed DEMATEL model. [ABSTRACT FROM AUTHOR]

Evidence of microplastics (MPs) and nanoplastics (NPs) in foods and daily-use products, along with their frequent detection in the human body, has raised concerns regarding their potential impact on human health through dietary ingestion. However, there is a lack of quantitative tools to simulate their bioaccumulation and tissue distribution following environmental exposure. To address this gap, we developed the first physiologically based toxicokinetic (PBTK) model for predicting the biodistribution of MPs and NPs in mice following oral exposure under various exposure scenarios. This novel model incorporated key kinetic mass transport processes, such as membrane permeability, albumin binding, and cellular uptake. We identified that the absorption rate in the gastrointestinal tract and fecal excretion rate constant had significant impacts on organ dosimetry. Our regression analysis indicated that the size-dependent dissociation constant and urine clearance rate constant sharply increased by a factor of 3 as NPs particle size increased to 1 µm. Finally, we developed a graphical user interface to enable interactive visualization and analysis for future applications, supporting human dietary exposure and risk assessment using available food consumption data and MPs/NPs residue data. The simulation results offer a mechanistic perspective, enhancing understanding of the internal organ dosimetry burden and health impacts from dietary exposure to MPs and NPs. [Display omitted] • This is the first whole-body PBTK model for micro/nanoplastics in mice. • GI tract absorption and fecal excretion rates are sensitive to organ dosimetry. • Size-dependent PBTK parameters change dramatically at particle sizes under 1 µm. • The model is converted to an interactive web dashboard to facilitate applications. • The model can support environmental exposure and human health risk assessment. [ABSTRACT FROM AUTHOR]

In response to escalating global wastewater issues, particularly from dye contaminants, many studies have begun using hydrochar to adsorb dye from wastewater. However, the relationship between the preparation conditions of hydrochar, the properties of hydrochar, experimental conditions, types of dyes, and equilibrium adsorption capacity (Q) has not yet been fully explored. This study conducted a comprehensive assessment using twelve distinct ML models. The Gradient Boosting Regressor (GBR) model exhibited superior performance with R² (0.9629) and RMSE (0.1166) in the test dataset, marking it as the most effective among the evaluated models. Moreover, this study also proved the feasibility of the GBR model through stability testing and residual analysis. A feature importance analysis prioritized the variables as follows: experimental conditions (41.5 %), properties of hydrochar (26.0 %), preparation conditions (18.1 %), and type of dye (14.4 %). Meanwhile, experimental conditions (C 0 > 30 mmol/g, pH > 8, and higher solvent temperatures) and hydrochar properties (the BET surface area > 2000 m²/g, an (O+N)/C molar ratio < 0.6, and an H/C molar ratio of approximately 0.06) show higher Q for dyes. Experimental validation of the GBR model confirmed its practical utility with a suitable predictive accuracy (R² = 0.8704). Moreover, the study developed a Python-based GUI that has integrated the best GBR models to facilitate researchers' ongoing application and improvement of this predictive model. This study not only underscores the efficacy of ML in enhancing the understanding of dye adsorption by hydrochar but also sets a precedent for future research on sustainable contaminants removal through bio-based adsorbents. [Display omitted] • Twelve ML models were tested, with the Gradient Boosting Regressor standing out for its exceptional performance. • The experimental verification showed the promising potential of the developed model. • SHAP and PDP were introduced for model interpretation. • An easy-to-use graphical user interface was developed for hydrochar absorb dyes. [ABSTRACT FROM AUTHOR]