Nothing Special   »   [go: up one dir, main page]

You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Article Types

Countries / Regions

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Search Results (9,925)

Search Parameters:
Keywords = Intelligent algorithms

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
18 pages, 919 KiB  
Article
Deep Learning for Ultrasonographic Assessment of Temporomandibular Joint Morphology
by Julia Lasek, Karolina Nurzynska, Adam Piórkowski, Michał Strzelecki and Rafał Obuchowicz
Tomography 2025, 11(3), 27; https://doi.org/10.3390/tomography11030027 (registering DOI) - 27 Feb 2025
Abstract
Background: Temporomandibular joint (TMJ) disorders are a significant cause of orofacial pain. Artificial intelligence (AI) has been successfully applied to other imaging modalities but remains underexplored in ultrasonographic evaluations of TMJ. Objective: This study aimed to develop and validate an AI-driven method for [...] Read more.
Background: Temporomandibular joint (TMJ) disorders are a significant cause of orofacial pain. Artificial intelligence (AI) has been successfully applied to other imaging modalities but remains underexplored in ultrasonographic evaluations of TMJ. Objective: This study aimed to develop and validate an AI-driven method for the automatic and reproducible measurement of TMJ space width from ultrasonographic images. Methods: A total of 142 TMJ ultrasonographic images were segmented into three anatomical components: the mandibular condyle, joint space, and glenoid fossa. State-of-the-art architectures were tested, and the best-performing 2D Residual U-Net was trained and validated against expert annotations. The algorithm for joint space width measurement based on TMJ segmentation was proposed, calculating the vertical distance between the superior-most point of the mandibular condyle and its corresponding point on the glenoid fossa. Results: The segmentation model achieved high performance for the mandibular condyle (Dice: 0.91 ± 0.08) and joint space (Dice: 0.86 ± 0.09), with notably lower performance for the glenoid fossa (Dice: 0.60 ± 0.24), highlighting variability due to its complex geometry. The TMJ space width measurement algorithm demonstrated minimal bias, with a mean difference of 0.08 mm and a mean absolute error of 0.18 mm compared to reference measurements. Conclusions: The model exhibited potential as a reliable tool for clinical use, demonstrating accuracy in TMJ ultrasonographic analysis. This study underscores the ability of AI-driven segmentation and measurement algorithms to bridge existing gaps in ultrasonographic imaging and lays the foundation for broader clinical applications. Full article
(This article belongs to the Topic AI in Medical Imaging and Image Processing)
29 pages, 3479 KiB  
Article
Estimating Daily Reference Crop Evapotranspiration in Northeast China Using Optimized Empirical Models Based on Heuristic Intelligence Algorithms
by Zongyang Li, Zhengxin Zhao, Liwen Xing, Lu Zhao, Ningbo Cui and Huanjie Cai
Agronomy 2025, 15(3), 599; https://doi.org/10.3390/agronomy15030599 (registering DOI) - 27 Feb 2025
Abstract
Accurately estimating reference crop evapotranspiration (ETo) improves agricultural water use efficiency. However, the accuracy of ETo estimation needs to be further improved in the Northeast region of China, the country’s main grain production area. In this research, meteorological data from 30 sites in [...] Read more.
Accurately estimating reference crop evapotranspiration (ETo) improves agricultural water use efficiency. However, the accuracy of ETo estimation needs to be further improved in the Northeast region of China, the country’s main grain production area. In this research, meteorological data from 30 sites in Northeast China over the past 59 years (1961–2019) were selected to evaluate the simulation accuracy of 11 ETo estimation models. By using the least square method (LSM) and three population heuristic intelligent algorithms—a genetic algorithm (GA), a particle swarm optimization algorithm (PSO), and a differential evolution algorithm (DE)—the parameters of eleven kinds of models were optimized, respectively, and the ETo estimation model suitable for northeast China was selected. The results showed that the radiation-based Jensen and Haise (JH) model had the best simulation accuracy for ETo in Northeast China among the 11 empirical models, with R2 of 0.92. The Hamon model had an acceptable estimation accuracy, while the combination model had low simulation accuracy in Northeast China, with R2 ranges of 0.74–0.88. After LSM optimization, the simulation accuracy of all models had been significantly improved by 0.58–12.1%. The results of heuristic intelligent algorithms showed that Hamon and Door models optimized by GA and DE algorithms had higher simulation accuracy, with R2 of 0.92. Although the JH model requires more meteorological factors than the Hamon and Door model, it shows better stability. Regardless of the original empirical formula or the optimization of various algorithms, JH has higher simulation accuracy, and R2 is greater than 0.91. Therefore, when only temperature or radiation factors were available, it was recommended to use the Hamon or Door model optimized by GA to estimate ETo, respectively; both models underestimated ETo with an absolute error range of 0.01–0.02 mm d–1 compared to the reference Penman–Monteith (P–M) equation. When more meteorological factors were available, the JH model optimized by LSM or GA could be used to estimate ETo in Northeast China, with an absolute error of less than 0.01 mm d–1. This study provided a more accurate ETo estimation method within the regional scope with incomplete meteorological data. Full article
(This article belongs to the Section Precision and Digital Agriculture)
21 pages, 1963 KiB  
Article
A Particle Swarm Optimization-Based Ensemble Broad Learning System for Intelligent Fault Diagnosis in Safety-Critical Energy Systems with High-Dimensional Small Samples
by Jiasheng Yan, Yang Sui and Tao Dai
Mathematics 2025, 13(5), 797; https://doi.org/10.3390/math13050797 (registering DOI) - 27 Feb 2025
Abstract
Intelligent fault diagnosis (IFD) plays a crucial role in reducing maintenance costs and enhancing the reliability of safety-critical energy systems (SCESs). In recent years, deep learning-based IFD methods have achieved high fault diagnosis accuracy extracting implicit higher-order correlations between features. However, the excessive [...] Read more.
Intelligent fault diagnosis (IFD) plays a crucial role in reducing maintenance costs and enhancing the reliability of safety-critical energy systems (SCESs). In recent years, deep learning-based IFD methods have achieved high fault diagnosis accuracy extracting implicit higher-order correlations between features. However, the excessive long training time of deep learning models conflicts with the requirements of real-time analysis for IFD, hindering their further application in practical industrial environments. To address the aforementioned challenge, this paper proposes an innovative IFD method for SCES that combines the particle swarm optimization (PSO) algorithm and the ensemble broad learning system (EBLS). Specifically, the broad learning system (BLS), known for its low time complexity and high classification accuracy, is adopted as an alternative to deep learning for fault diagnosis in SCES. Furthermore, EBLS is designed to enhance model stability and classification accuracy with high-dimensional small samples by incorporating the random forest (RF) algorithm and an ensemble strategy into the traditional BLS framework. In order to reduce the computational cost of the EBLS, which is constrained by the selection of its hyperparameters, the PSO algorithm is employed to optimize the hyperparameters of the EBLS. Finally, the model is validated through simulated data from a complex nuclear power plant (NPP). Numerical experiments reveal that the proposed method significantly improved the diagnostic efficiency while maintaining high accuracy. In summary, the proposed approach shows great promise for boosting the capabilities of the IFD models for SCES. Full article
16 pages, 654 KiB  
Article
Intelligent Scheduling of a Pulsating Assembly Flow Shop Considering a Multifunctional Automated Guided Vehicle
by Hailong Song, Shengluo Yang, Shuoxin Yin, Junyi Wang and Zhigang Xu
Appl. Sci. 2025, 15(5), 2593; https://doi.org/10.3390/app15052593 (registering DOI) - 27 Feb 2025
Abstract
The pulsating assembly line is widely used in modern manufacturing, particularly in high-precision industries such as aerospace, where it greatly enhances production efficiency. To achieve overall optimization, both product scheduling and Automated Guided Vehicle (AGV) scheduling must be simultaneously optimized. However, existing research [...] Read more.
The pulsating assembly line is widely used in modern manufacturing, particularly in high-precision industries such as aerospace, where it greatly enhances production efficiency. To achieve overall optimization, both product scheduling and Automated Guided Vehicle (AGV) scheduling must be simultaneously optimized. However, existing research predominantly focuses on product scheduling, with limited attention given to AGV scheduling. This paper proposes an optimized solution for the pulsating assembly line scheduling problem, incorporating multifunctional AGV scheduling. A mathematical model is developed and three AGV selection strategies and three AGV standby strategies are designed to optimize AGV scheduling and control. To improve scheduling efficiency, nine heuristic strategies are introduced, along with the Variable Neighborhood Descent (VND) algorithm as a metaheuristic method for product scheduling. The VND algorithm refines the solution through multiple neighborhood searches, enhancing both the precision and efficiency of product scheduling. Our experimental results demonstrate that the proposed strategies significantly improve the production efficiency of pulsating assembly workshops, reduce AGV scheduling costs, and optimize overall production workflows. This study offers novel methods for intelligent scheduling in pulsating assembly workshops, contributing to the advancement of manufacturing toward “multiple varieties, small batches, and customization.” Full article
20 pages, 1226 KiB  
Article
Research on Precise Segmentation and Center Localization of Weeds in Tea Gardens Based on an Improved U-Net Model and Skeleton Refinement Algorithm
by Zhiyong Cao, Shuai Zhang, Chen Li, Wei Feng, Baijuan Wang, Hao Wang, Ling Luo and Hongbo Zhao
Agriculture 2025, 15(5), 521; https://doi.org/10.3390/agriculture15050521 (registering DOI) - 27 Feb 2025
Abstract
The primary objective of this research was to develop an efficient method for accurately identifying and localizing weeds in ecological tea garden environments, aiming to enhance the quality and yield of tea production. Weed competition poses a significant challenge to tea production, particularly [...] Read more.
The primary objective of this research was to develop an efficient method for accurately identifying and localizing weeds in ecological tea garden environments, aiming to enhance the quality and yield of tea production. Weed competition poses a significant challenge to tea production, particularly due to the small size of weed plants, their color similarity to tea trees, and the complexity of their growth environment. A dataset comprising 5366 high-definition images of weeds in tea gardens has been compiled to address this challenge. An enhanced U-Net model, incorporating a Double Attention Mechanism and an Atrous Spatial Pyramid Pooling module, is proposed for weed recognition. The results of the ablation experiments show that the model significantly improves the recognition accuracy and the Mean Intersection over Union (MIoU), which are enhanced by 4.08% and 5.22%, respectively. In addition, to meet the demand for precise weed management, a method for determining the center of weed plants by integrating the center of mass and skeleton structure has been developed. The skeleton was extracted through a preprocessing step and a refinement algorithm, and the relative positional relationship between the intersection point of the skeleton and the center of mass was cleverly utilized to achieve up to 82% localization accuracy. These results provide technical support for the research and development of intelligent weeding equipment for tea gardens, which helps to maintain the ecology of tea gardens and improve production efficiency and also provides a reference for weed management in other natural ecological environments. Full article
(This article belongs to the Special Issue Applications of Remote Sensing in Agricultural Soil and Crop Mapping)
17 pages, 13551 KiB  
Article
Lithology Identification of Buried Hill Reservoir Based on XGBoost with Optimized Interpretation
by Bin Zhao and Wenlong Liao
Processes 2025, 13(3), 682; https://doi.org/10.3390/pr13030682 (registering DOI) - 27 Feb 2025
Abstract
Buried hill reservoirs are characterized by complex formation conditions and highly heterogeneous rock structures, which result in the poor performance of traditional crossplot methods in stratigraphic lithology classification. Logging curves, as comprehensive reflections of various petrophysical properties, are influenced by complex geological factors, [...] Read more.
Buried hill reservoirs are characterized by complex formation conditions and highly heterogeneous rock structures, which result in the poor performance of traditional crossplot methods in stratigraphic lithology classification. Logging curves, as comprehensive reflections of various petrophysical properties, are influenced by complex geological factors, leading to overlapping response values even among different lithologies with similar physical properties. This overlap negatively impacts the accuracy of intelligent lithology identification methods. To address this challenge, this study leverages logging response data, experimental data, and mud logging data to propose an optimized inversion method for mineral content, introducing mineral curves to resolve the curve overlap issue. By analyzing six wells in the study area, models were constructed using the calculated mineral content curves and conventional logging features to mitigate the feature overlap. The XGBoost algorithm was employed to identify lithologies by addressing the nonlinear relationships inherent in complex reservoirs. The experimental results indicate that the optimized mineral curves significantly enhance the model’s discriminative capability, effectively addressing the decline in identification accuracy due to feature overlap. Compared to models such as Random Forest (RF) and Support Vector Machine (SVM), the XGBoost model demonstrated superior accuracy and stability, providing a reliable basis for precise reservoir identification in the study area. Full article
(This article belongs to the Section Energy Systems)
Show Figures

Figure 1

Figure 1
<p>XGBoost model flowchart.</p>
Full article ">Figure 2
<p>Relationship between lithology and mineral content in the study area.</p>
Full article ">Figure 3
<p>Relationship between quartz, potassium feldspar, and plagioclase content in the buried hill.</p>
Full article ">Figure 4
<p>Relationship between quartz, feldspar, and clay content in granite buried hill.</p>
Full article ">Figure 5
<p>Mineral Content calculation results for Well H1.</p>
Full article ">Figure 6
<p>Overview of the study area.</p>
Full article ">Figure 7
<p>Typical logging response curve characteristics of different lithologies in the study area.</p>
Full article ">Figure 8
<p>Gamma and density crossplot for different lithologies.</p>
Full article ">Figure 9
<p>Gamma and neutron crossplot for different lithologies.</p>
Full article ">Figure 10
<p>Density–sonic crossplot of different lithologies in the study area.</p>
Full article ">Figure 11
<p>Optimal parameter combination for the model obtained through optimization and cross-validation.</p>
Full article ">Figure 12
<p>t-SNE visualization of the distribution of lithology samples: (<b>a</b>) dataset distribution without mineral curve data, (<b>b</b>) dataset distribution with mineral curve data.</p>
Full article ">Figure 13
<p>Comparison of confusion matrices for model prediction results with and without mineral content curves: (<b>a</b>) model prediction results without using mineral content curves for training, (<b>b</b>) model prediction results using mineral content curves for training.</p>
Full article ">Figure 14
<p>Lithology identification results for Well H2.</p>
Full article ">
22 pages, 3368 KiB  
Article
Diagnosis of Lung Cancer Using Endobronchial Ultrasonography Image Based on Multi-Scale Image and Multi-Feature Fusion Framework
by Huitao Wang, Takahiro Nakajima, Kohei Shikano, Yukihiro Nomura and Toshiya Nakaguchi
Tomography 2025, 11(3), 24; https://doi.org/10.3390/tomography11030024 - 27 Feb 2025
Viewed by 46
Abstract
Lung cancer is the leading cause of cancer-related deaths globally and ranks among the most common cancer types. Given its low overall five-year survival rate, early diagnosis and timely treatment are essential to improving patient outcomes. In recent years, advances in computer technology [...] Read more.
Lung cancer is the leading cause of cancer-related deaths globally and ranks among the most common cancer types. Given its low overall five-year survival rate, early diagnosis and timely treatment are essential to improving patient outcomes. In recent years, advances in computer technology have enabled artificial intelligence to make groundbreaking progress in imaging-based lung cancer diagnosis. The primary aim of this study is to develop a computer-aided diagnosis (CAD) system for lung cancer using endobronchial ultrasonography (EBUS) images and deep learning algorithms to facilitate early detection and improve patient survival rates. We propose M3-Net, which is a multi-branch framework that integrates multiple features through an attention-based mechanism, enhancing diagnostic performance by providing more comprehensive information for lung cancer assessment. The framework was validated on a dataset of 95 patient cases, including 13 benign and 82 malignant cases. The dataset comprises 1140 EBUS images, with 540 images used for training, and 300 images each for the validation and test sets. The evaluation yielded the following results: accuracy of 0.76, F1-score of 0.75, AUC of 0.83, PPV of 0.80, NPV of 0.75, sensitivity of 0.72, and specificity of 0.80. These findings indicate that the proposed attention-based multi-feature fusion framework holds significant potential in assisting with lung cancer diagnosis. Full article
Show Figures

Figure 1

Figure 1
<p>This figure illustrates the detailed process for generating multi-scale EBUS images: First, three EBUS images are selected from the four available types. Each selected EBUS image is then converted from RGB to gray-scale and resized to 224 × 224 pixels. Finally, the three resized gray-scale images are merged to form a single multi-scale RGB image. This approach utilizes varying image resolutions to effectively capture multi-scale features within one composite image.</p>
Full article ">Figure 2
<p>This figure illustrates the architecture of our proposed <math display="inline"><semantics> <mrow> <msup> <mi mathvariant="normal">M</mi> <mn>3</mn> </msup> <mrow> <mtext>-</mtext> <mi>Net</mi> </mrow> </mrow> </semantics></math>, which is composed of three key components: a feature extraction module, a feature fusion module, and a classifier. Each branch in the <math display="inline"><semantics> <mrow> <msup> <mi mathvariant="normal">M</mi> <mn>3</mn> </msup> <mrow> <mtext>-</mtext> <mi>Net</mi> </mrow> </mrow> </semantics></math> architecture corresponds to a specific type of EBUS image, with each image type processed by an individual CNN encoder for feature extraction. The extracted features are then sent to the feature fusion module, where they are combined. Finally, the fused features are passed into the classifier to generate the final class prediction. The feature extraction module extracts distinct features, <span class="html-italic">F</span><sub>1</sub>, <span class="html-italic">F</span><sub>2</sub>, and <span class="html-italic">F</span><sub>3</sub>, from different input images. These extracted features are then processed through an attention-based feature fusion module, where they are integrated. Finally, the fused features are passed into the classifier to generate probability predictions for the target categories.</p>
Full article ">Figure 3
<p>This figure illustrates the framework of feature fusion module version 1 (FFM-v1).</p>
Full article ">Figure 4
<p>This figure illustrates the framework of feature fusion module version 2 (FFM-v2).</p>
Full article ">Figure 5
<p>This figure illustrates the framework of feature fusion module version 3 (FFM-v3).</p>
Full article ">Figure 6
<p>This figure represents the encoder framework based on the fine-tuned DenseNet-121 model.</p>
Full article ">Figure 7
<p>This figure illustrates 12 CNN models that were trained on the polar type 3 EBUS dataset using both unweighted cross-entropy and weighted cross-entropy as loss functions to evaluate their AUC.</p>
Full article ">Figure 8
<p>This figure shows the best AUC achieved when training 12 CNN models with weighted cross-entropy loss on each of the 14 EBUS image datasets.</p>
Full article ">
12 pages, 201 KiB  
Review
Advances in Autism Spectrum Disorder (ASD) Diagnostics: From Theoretical Frameworks to AI-Driven Innovations
by Christine K. Syriopoulou-Delli
Electronics 2025, 14(5), 951; https://doi.org/10.3390/electronics14050951 (registering DOI) - 27 Feb 2025
Viewed by 22
Abstract
This study provides a comprehensive analysis of the evolution of Autism Spectrum Disorder (ASD) diagnostics, tracing its progression from psychoanalytic origins to the integration of advanced artificial intelligence (AI) technologies. The study explores, through scientific data bases like Pub Med, Scopus, and Google [...] Read more.
This study provides a comprehensive analysis of the evolution of Autism Spectrum Disorder (ASD) diagnostics, tracing its progression from psychoanalytic origins to the integration of advanced artificial intelligence (AI) technologies. The study explores, through scientific data bases like Pub Med, Scopus, and Google Scholar, how theoretical frameworks, including psychoanalysis, behavioral psychology, cognitive development, and neurobiological paradigms, have shaped diagnostic methodologies over time. Each paradigm’s associated assessment tools, such as the Autism Diagnostic Observation Schedule (ADOS) and the Vineland Adaptive Behavior Scales, are discussed in relation to their scientific advancements and limitations. Emerging technologies, particularly AI, are highlighted for their transformative impact on ASD diagnostics. The application of AI in areas such as video analysis, Natural Language Processing (NLP), and biodata integration demonstrates significant progress in precision, accessibility, and inclusivity. Ethical considerations, including algorithmic transparency, data security, and inclusivity for underrepresented populations, are critically examined alongside the challenges of scalability and equitable implementation. Additionally, neurodiversity- informed approaches are emphasized for their role in reframing autism as a natural variation of human cognition and behavior, advocating for strength-based, inclusive diagnostic frameworks. This synthesis underscores the interplay between evolving theoretical models, technological advancements, and the growing focus on compassionate, equitable diagnostic practices. It concludes by advocating for continued innovation, interdisciplinary collaboration, and ethical oversight to further refine ASD diagnostics and improve outcomes for individuals across the autism spectrum. Full article
(This article belongs to the Special Issue Robotics: From Technologies to Applications)
18 pages, 7601 KiB  
Article
Intelligent Detection Algorithm for Concrete Bridge Defects Based on SATH–YOLO Model
by Lanlin Zou and Ao Liu
Sensors 2025, 25(5), 1449; https://doi.org/10.3390/s25051449 - 27 Feb 2025
Viewed by 44
Abstract
Amid the era of intelligent construction and inspection, traditional object detection models like YOLOv8 struggle in bridge defect detection due to high computational complexity and limited speed. To address this, the lightweight SATH–YOLO model was proposed in this paper. First, the Star Block [...] Read more.
Amid the era of intelligent construction and inspection, traditional object detection models like YOLOv8 struggle in bridge defect detection due to high computational complexity and limited speed. To address this, the lightweight SATH–YOLO model was proposed in this paper. First, the Star Block from StarNet was used to build the STNC2f module, enriching semantic information and improving multi-scale feature fusion while reducing parameters and computation. Second, the SPPF module was replaced with an AIFI module to capture finer-grained local features, improving feature-fusion precision and adaptability in complex scenarios. Lastly, a lightweight TDMDH detection head with shared convolution and dynamic feature selection further reduced computational costs. With the SATH–YOLO model, parameter count, computation, and model size were reduced significantly by 39.9%, 8.6%, and 36.2%, respectively. Meanwhile, the average detection precision was not impacted but improved by 1%, which meets the demands of edge devices and resource-constrained environments. Full article
(This article belongs to the Section Intelligent Sensors)
Show Figures

Figure 1

Figure 1
<p>Structure diagram of the YOLOv8 algorithm.</p>
Full article ">Figure 2
<p>StarNet Architecture Principles.</p>
Full article ">Figure 3
<p>STNC2f module structure.</p>
Full article ">Figure 4
<p>AIFI module structure.</p>
Full article ">Figure 5
<p>The feature fusion process of the AIFI module.</p>
Full article ">Figure 6
<p>Conv_GN module structure.</p>
Full article ">Figure 7
<p>TDMDH structure diagram.</p>
Full article ">Figure 8
<p>DYFS structure diagram.</p>
Full article ">Figure 9
<p>SATH–YOLO structure diagram.</p>
Full article ">Figure 10
<p>Comparison of concrete bridge crack detection results. (<b>a</b>–<b>d</b>) are four types of random cracks.</p>
Full article ">Figure 11
<p>Visual comparison of focused feature areas. (<b>a</b>–<b>d</b>) are four types of random cracks.</p>
Full article ">Figure 12
<p>Comparison of mAP@50 curves of YOLO series on bridge crack dataset.</p>
Full article ">
22 pages, 4732 KiB  
Article
Rapid Impedance Measurement of Lithium-Ion Batteries Under Pulse Ex-Citation and Analysis of Impedance Characteristics of the Regularization Distributed Relaxation Time
by Haisen Chen, Jinghan Bai, Zhengpu Wu, Ziang Song, Bin Zuo, Chunxia Fu, Yunbin Zhang and Lujun Wang
Batteries 2025, 11(3), 91; https://doi.org/10.3390/batteries11030091 (registering DOI) - 27 Feb 2025
Viewed by 124
Abstract
To address the limitations of conventional electrochemical impedance spectroscopy (EIS) testing, we propose an efficient rapid EIS testing system. This system utilizes an AC pulse excitation signal combined with an “intelligent fast fourier transform (IFFT) optimization algorithm” to achieve rapid “one-to-many” impedance data [...] Read more.
To address the limitations of conventional electrochemical impedance spectroscopy (EIS) testing, we propose an efficient rapid EIS testing system. This system utilizes an AC pulse excitation signal combined with an “intelligent fast fourier transform (IFFT) optimization algorithm” to achieve rapid “one-to-many” impedance data measurements. This significantly enhances the speed, flexibility, and practicality of EIS testing. Furthermore, the conventional model-fitting approach for EIS data often struggles to resolve the issue of overlapping impedance arcs within a limited frequency range. To address this, the present study employs the Regularization Distributed Relaxation Time (RDRT) method to process EIS data obtained under AC pulse conditions. This approach avoids the workload and analytical uncertainties associated with assuming equivalent circuit models. Finally, the practical utility of the proposed testing system and the RDRT impedance analysis method is demonstrated through the estimation of battery state of health (SOH). In summary, the method proposed in this study not only addresses the issues associated with conventional EIS data acquisition and analysis but also broadens the methodologies and application scope of EIS impedance testing. This opens up new possibilities for its application in fields such as lithium-ion batteries (LIBs) energy storage. Full article
Show Figures

Figure 1

Figure 1
<p>EIS data principle test graph. (<b>A</b>) Schematic diagram of conventional EIS testing; (<b>B</b>) Comparative analysis of pulse signal and sine wave signals through fourier decomposition.</p>
Full article ">Figure 2
<p>EIS data DRT method analysis graph. (<b>A</b>) Electrochemical model of LIBs and DRT equivalent circuit model; (<b>B</b>,<b>C</b>) Comparative analysis of ideal elements and CEP after DRT processing.</p>
Full article ">Figure 3
<p>Hardware architecture diagram of the EIS rapid testing system. (<b>A</b>) Overall architecture diagram of the EIS rapid testing system; (<b>B</b>) Hardware operational structure diagram of the impedance testing system; (<b>C</b>) Topology diagram of the variable excitation equalization board.</p>
Full article ">Figure 4
<p>Software architecture diagram of the EIS rapid testing system. (<b>A</b>) Schematic diagram of the software architecture of the system; (<b>B</b>) Structural diagram of the data processing module in the upper-level application.</p>
Full article ">Figure 5
<p>Lithium-ion battery aging and impedance testing experimental graph. (<b>A</b>) Flowchart of the LIBs aging experiment; (<b>B</b>) Voltage and current variation chart of the aged LIBs during the aging test; (<b>C</b>) Experimental graph of actual aging test for LIBs; (<b>D</b>) Experimental graph of actual testing on LIBs; (<b>E</b>) Experimental signal graph of EIS testing at 100 Hz.</p>
Full article ">Figure 6
<p>Impedance spectra at different SOC levels and under various cycle numbers. (<b>A</b>,<b>B</b>) Comparison between the EIS testing apparatus and the electrochemical workstation; (<b>C</b>,<b>D</b>) Impedance plots of different cycling numbers; (<b>E</b>,<b>F</b>) Impedance plots of different cycling numbers.</p>
Full article ">Figure 7
<p>Comparison plots of impedance data processed with RDRT for different SOC and cycle numbers. (<b>A</b>,<b>B</b>) DRT impedance plots without inductance at different cycling numbers; (<b>C</b>,<b>D</b>) DRT impedance plots with inductance at different cycling numbers. (<b>E</b>,<b>F</b>) DRT impedance plots without inductance at various SOC levels. (<b>G</b>,<b>H</b>) DRT impedance plots with inductance at various SOC levels.</p>
Full article ">Figure 8
<p>Investigation of the impedance data trends corresponding to S<sub>1</sub> and S<sub>3</sub> at different cycle numbers.</p>
Full article ">
28 pages, 1191 KiB  
Perspective
Shaping the Future of Healthcare: Ethical Clinical Challenges and Pathways to Trustworthy AI
by Polat Goktas and Andrzej Grzybowski
J. Clin. Med. 2025, 14(5), 1605; https://doi.org/10.3390/jcm14051605 - 27 Feb 2025
Viewed by 113
Abstract
Background/Objectives: Artificial intelligence (AI) is transforming healthcare, enabling advances in diagnostics, treatment optimization, and patient care. Yet, its integration raises ethical, regulatory, and societal challenges. Key concerns include data privacy risks, algorithmic bias, and regulatory gaps that struggle to keep pace with [...] Read more.
Background/Objectives: Artificial intelligence (AI) is transforming healthcare, enabling advances in diagnostics, treatment optimization, and patient care. Yet, its integration raises ethical, regulatory, and societal challenges. Key concerns include data privacy risks, algorithmic bias, and regulatory gaps that struggle to keep pace with AI advancements. This study aims to synthesize a multidisciplinary framework for trustworthy AI in healthcare, focusing on transparency, accountability, fairness, sustainability, and global collaboration. It moves beyond high-level ethical discussions to provide actionable strategies for implementing trustworthy AI in clinical contexts. Methods: A structured literature review was conducted using PubMed, Scopus, and Web of Science. Studies were selected based on relevance to AI ethics, governance, and policy in healthcare, prioritizing peer-reviewed articles, policy analyses, case studies, and ethical guidelines from authoritative sources published within the last decade. The conceptual approach integrates perspectives from clinicians, ethicists, policymakers, and technologists, offering a holistic “ecosystem” view of AI. No clinical trials or patient-level interventions were conducted. Results: The analysis identifies key gaps in current AI governance and introduces the Regulatory Genome—an adaptive AI oversight framework aligned with global policy trends and Sustainable Development Goals. It introduces quantifiable trustworthiness metrics, a comparative analysis of AI categories for clinical applications, and bias mitigation strategies. Additionally, it presents interdisciplinary policy recommendations for aligning AI deployment with ethical, regulatory, and environmental sustainability goals. This study emphasizes measurable standards, multi-stakeholder engagement strategies, and global partnerships to ensure that future AI innovations meet ethical and practical healthcare needs. Conclusions: Trustworthy AI in healthcare requires more than technical advancements—it demands robust ethical safeguards, proactive regulation, and continuous collaboration. By adopting the recommended roadmap, stakeholders can foster responsible innovation, improve patient outcomes, and maintain public trust in AI-driven healthcare. Full article
(This article belongs to the Section Clinical Guidelines)
Show Figures

Figure 1

Figure 1
<p>Key components of trustworthy AI in healthcare: The intersection of ethics, regulations, and technology, emphasizing privacy, fairness, and accountability for ethical AI integration.</p>
Full article ">Figure 2
<p>Conceptual framework illustrating how helpful internal attributes and challenging external factors intersect to influence AI outcomes. Moving toward the center signifies progress in achieving “<span class="html-italic">Trustworthy AI</span>” by balancing dimensions such as fairness, accountability, transparency, compliance, privacy, and sustainability. Each quadrant represents a key facet—ranging from bias-free fairness to secure privacy measures—that, when integrated and optimized, leads to safer, more effective, and ethically sound AI systems in healthcare.</p>
Full article ">
8 pages, 187 KiB  
Review
Use of Artificial Intelligence in Difficult Airway Assessment: The Current State of Knowledge
by Mateusz Wilk, Wojciech Pikiewicz, Krzysztof Florczak and Dawid Jakóbczak
J. Clin. Med. 2025, 14(5), 1602; https://doi.org/10.3390/jcm14051602 - 27 Feb 2025
Viewed by 65
Abstract
Artificial Intelligence (AI) has become one of the most transformative technologies of the 21st century. It is poised to reshape medicine, as almost every field of hospital treatment has seen an increase in AI’s presence. In this article, we focus on its impact [...] Read more.
Artificial Intelligence (AI) has become one of the most transformative technologies of the 21st century. It is poised to reshape medicine, as almost every field of hospital treatment has seen an increase in AI’s presence. In this article, we focus on its impact in the field of anesthesia. We discuss its possible influence on difficult airway management, as it remains one of the most critical and potentially hazardous aspects of anesthesia, often leading to life-threatening complications. The accurate prediction of difficult airways can significantly improve patient safety. We covered the available literature on AI-based models for difficult airway prediction in comparison to traditional forms of airway assessment, as well as the predictive value of ultrasonography. We also address the narrative that AI-based algorithms show high sensitivity and specificity, with which they significantly outperform classical tests based on complex scales and indices. Full article
(This article belongs to the Special Issue Artificial Intelligence (AI)-Based Diagnosis in Clinical Practice)
17 pages, 905 KiB  
Article
BERT Mutation: Deep Transformer Model for Masked Uniform Mutation in Genetic Programming
by Eliad Shem-Tov, Moshe Sipper and Achiya Elyasaf
Mathematics 2025, 13(5), 779; https://doi.org/10.3390/math13050779 - 26 Feb 2025
Viewed by 254
Abstract
We introduce BERT mutation, a novel, domain-independent mutation operator for Genetic Programming (GP) that leverages advanced Natural Language Processing (NLP) techniques to improve convergence, particularly using the Masked Language Modeling approach. By combining the capabilities of deep reinforcement learning and the BERT transformer [...] Read more.
We introduce BERT mutation, a novel, domain-independent mutation operator for Genetic Programming (GP) that leverages advanced Natural Language Processing (NLP) techniques to improve convergence, particularly using the Masked Language Modeling approach. By combining the capabilities of deep reinforcement learning and the BERT transformer architecture, BERT mutation intelligently suggests node replacements within GP trees to enhance their fitness. Unlike traditional stochastic mutation methods, BERT mutation adapts dynamically by using historical fitness data to optimize mutation decisions, resulting in more effective evolutionary improvements. Through comprehensive evaluations across three benchmark domains, we demonstrate that BERT mutation significantly outperforms conventional and state-of-the-art mutation operators in terms of convergence speed and solution quality. This work represents a pivotal step toward integrating state-of-the-art deep learning into evolutionary algorithms, pushing the boundaries of adaptive optimization in GP. Full article
(This article belongs to the Special Issue Machine Learning and Evolutionary Algorithms: Theory and Applications)
Show Figures

Figure 1

Figure 1
<p>Illustration of Masked Language Modeling using BERT [<a href="#B8-mathematics-13-00779" class="html-bibr">8</a>]: “Super Bowl 50 was an American football game to determine the champion” becomes “Super Bowl 50 was # # # # to determine the champion”, where # represents a mask. The model is then trained to predict the masked tokens, thereby inferring the missing words. This approach enables the model to learn bidirectional contextual representations by incorporating both the left and right contexts of the sentence during training.</p>
Full article ">Figure 2
<p>A transformer encoder block used in BERT (represented by the gray box) [<a href="#B7-mathematics-13-00779" class="html-bibr">7</a>]. The input is first converted into an input embedding, which is combined with positional encoding to retain the order of the sequence. The positional encoding provides information about the position of each token in the input sequence, which is essential, since the transformer architecture lacks inherent sequence order. The gray block is repeated <span class="html-italic">N</span> times.</p>
Full article ">Figure 3
<p>A GP tree and its string representation below, generated by traversing the nodes in infix order.</p>
Full article ">Figure 4
<p>The mask replacement process of the BERT mutation. The trained BERT model takes a masked individual and outputs the probability for each possible replacement. The softmax function samples a possible replacement, and the string with the replaced token is passed again until all masks are replaced. In the first iteration of this example, the masked node is a constant; all non-constant replacements are considered illegal and thus masked. In the second iteration, we replace the mask with an operator that has an arity of two. The red box highlights the masked token that is currently being replaced.</p>
Full article ">Figure 5
<p>The Santa Fe trail problem instance. In the depicted grid, yellow cells are empty, and blue cells contain food.</p>
Full article ">Figure 6
<p>Fitness value of best individual vs. generation, of each mutation operator for the <tt>friedman1</tt> symbolic regression dataset. The fitness value is averaged over 10 runs.</p>
Full article ">Figure 7
<p>Fitness value of best individual vs. generation, of each mutation operator for the <tt>occupancy</tt> symbolic classification dataset. The fitness value is averaged over 10 runs.</p>
Full article ">Figure 8
<p>Fitness value of best individual vs. generation, of each mutation operator for the Artificial Ant <tt>Los Altos trail</tt> instance. The fitness value is averaged over 10 runs.</p>
Full article ">Figure 9
<p>Mean average tree length per generation of each mutation operator for the Artificial Ant <tt>Los Altos trail</tt> instance. The values are averaged over 10 runs.</p>
Full article ">
34 pages, 640 KiB  
Article
Brute Force Computations and Reference Solutions
by Mihail Mihaylov Konstantinov, Petko Hristov Petkov and Ekaterina Borisova Madamlieva
Foundations 2025, 5(1), 7; https://doi.org/10.3390/foundations5010007 - 26 Feb 2025
Viewed by 88
Abstract
In this paper, we consider the application of brute force computational techniques (BFCTs) for solving computational problems in mathematical analysis and matrix algebra in a floating-point computing environment. These techniques include, among others, simple matrix computations and the analysis of graphs of functions. [...] Read more.
In this paper, we consider the application of brute force computational techniques (BFCTs) for solving computational problems in mathematical analysis and matrix algebra in a floating-point computing environment. These techniques include, among others, simple matrix computations and the analysis of graphs of functions. Since BFCTs are based on matrix calculations, the program system MATLAB® is suitable for their computer realization. The computations in this paper are completed in double precision floating-point arithmetic, obeying the 2019 IEEE Standard for binary floating-point calculations. One of the aims of this paper is to analyze cases where popular algorithms and software fail to produce correct answers, failing to alert the user. In real-time control applications, this may have catastrophic consequences with heavy material damage and human casualties. It is known, or suspected, that a number of man-made catastrophes such as the Dharhan accident (1991), Ariane 5 launch failure (1996), Boeing 737 Max tragedies (2018, 2019) and others are due to errors in the computer software and hardware. Another application of BFCTs is finding good initial guesses for known computational algorithms. Sometimes, simple and relatively fast BFCTs are useful tools in solving computational problems correctly and in real time. Among particular problems considered are the genuine addition of machine numbers, numerically stable computations, finding minimums of arrays, the minimization of functions, solving finite equations, integration and differentiation, computing condensed and canonical forms of matrices and clarifying the concepts of the least squares method in the light of the conflict remainders vs. errors. Usually, BFCTs are applied under the user’s supervision, which is not possible in the automatic implementation of computational methods. To implement BFCTs automatically is a challenging problem in the area of artificial intelligence and of mathematical artificial intelligence in particular. BFCTs allow to reveal the underlying arithmetic in the performance of computational algorithms. Last but not least, this paper has tutorial value, as computational algorithms and mathematical software are often taught without considering the properties of computational algorithms and machine arithmetic. Full article
(This article belongs to the Section Mathematical Sciences)
Show Figures

Figure 1

Figure 1
<p>Scaled rounding errors.</p>
Full article ">Figure 2
<p>An oscillating function.</p>
Full article ">Figure 3
<p>Computed first difference of the function <math display="inline"><semantics> <mrow> <mi>f</mi> <mo>(</mo> <mi>x</mi> <mo>)</mo> <mo>=</mo> <mi>x</mi> </mrow> </semantics></math>.</p>
Full article ">
20 pages, 3322 KiB  
Article
Consensus-Based Formation Control for Heterogeneous Multi-Agent Systems in Complex Environments
by Xiaofei Chang, Yiming Yang, Zhuo Zhang, Jiayue Jiao, Haoyu Cheng and Wenxing Fu
Drones 2025, 9(3), 175; https://doi.org/10.3390/drones9030175 - 26 Feb 2025
Viewed by 83
Abstract
The purpose of this paper is to develop formation control strategies for heterogeneous multi-intelligent-agent systems in complex environments, with the goal of enhancing their performance, reliability, and stability. Complex flight conditions, such as navigating narrow gaps in urban high-rise buildings, pose considerable challenges [...] Read more.
The purpose of this paper is to develop formation control strategies for heterogeneous multi-intelligent-agent systems in complex environments, with the goal of enhancing their performance, reliability, and stability. Complex flight conditions, such as navigating narrow gaps in urban high-rise buildings, pose considerable challenges for agent control. To address these challenges, this paper proposes a consensus-based formation strategy that integrates graph theory and multi-consensus algorithms. This approach incorporates time-varying group consistency to strengthen fault tolerance and reduce interference while ensuring obstacle avoidance and formation maintenance in dynamic environments. Through a Lyapunov stability analysis, combined with minimum dwell time constraints and the LaSalle invariance principle, this work proves the convergence of the proposed control scheme under changing network topologies. Simulation results confirm that the proposed strategy significantly improves system performance, mission execution capability, autonomy, synergy, and robustness, thereby enabling agents to successfully maintain formation and avoid obstacles in both homogeneous and heterogeneous clusters in complex environments. Full article
(This article belongs to the Special Issue Swarm Intelligence in Multi-UAVs)
Show Figures

Figure 1

Figure 1
<p>First-order agent consensus state trajectory.</p>
Full article ">Figure 2
<p>Velocity of first-order intelligences over time.</p>
Full article ">Figure 3
<p>Position of second-order intelligences over time.</p>
Full article ">Figure 4
<p>Velocity versus time for second-order intelligences.</p>
Full article ">Figure 5
<p>Acceleration versus time for second-order intelligences.</p>
Full article ">Figure 6
<p>Simulated multi-intelligent-agent formation.</p>
Full article ">Figure 7
<p>Simulation results of formation grouping control for multi-intelligent-agent system.</p>
Full article ">Figure 8
<p>Real-time change curves of one-order and two-order multi-consensus results.</p>
Full article ">Figure 9
<p>Simulation results of formation control for 25 multi-intelligent body systems.</p>
Full article ">
Back to TopTop