In relativity, the Newtonian concepts of velocity and acceleration are observer-dependent quantities that vary with the chosen frame of reference. It is well established that in the comoving frame, cosmic expansion is currently accelerating; however, in the rest frame, this expansion is actually decelerating. In this paper, we explore the implications of this distinction. The traditional measure of cosmic acceleration, denoted by q, is derived from the comoving frame and describes the acceleration of the scale factor a for a 3D space-like homogeneous sphere. We introduce a new parameter q_E representing the acceleration experienced between observers within the light cone. By comparing q_E to the traditional q using observational data from Type Ia supernovae (SN) and the radial clustering of galaxies and quasars (BAO)—including the latest results from DESI2024—our analysis demonstrates that q_E aligns more closely with these data. The core argument of the paper is that Λ—regardless of its origin—creates an event horizon that divides the manifold into two causally disconnected regions analogous to conditions inside a black hole’s interior, thereby allowing for a rest-frame perspective q_E in which cosmic expansion appears to be decelerating and the horizon acts like a friction term. Such a horizon suggests that the universe cannot maintain homogeneity outside. The observed cosmological constant Λ can then be interpreted not as a driver of new dark energy or a modification of gravity but as a boundary term exerting an attractive force, akin to a rubber band, resisting further expansion and preventing event horizon crossings. This interpretation calls for a reconsideration of current cosmological models and the assumptions underlying them.

Color image binarization plays a pivotal role in image preprocessing work, significantly impacting subsequent tasks, particularly in text recognition. This paper concentrates on Document Image Binarization (DIB), aiming to separate a image into foreground (text) and background (non-text content). Through a thorough analysis of conventional and deep learning-based approaches, we conclude that prevailing DIB methods leverage deep learning technology. Furthermore, we explore the receptive fields pre- and post-network training to underscore the Transformer model's advantages. Subsequently, we introduce a lightweight model based on the U-Net structure, enhanced with the Mobile ViT module to better capture global information features in document images. Given its adeptness at learning both local and global features, our proposed model exhibits superior performance on two standard datasets (DIBCO2012 and DIBCO2017) compared to state-of-the-art methods. Notably, our proposed DIB method presents a straightforward end-to-end model devoid of additional image preprocessing or post-processing. Moreover, its parameter count is less than a quarter of the HIP'23 model, which achives best results on three datasets(DIBCO2012, DIBCO2017 and DIBCO2018). Finally, two sets of ablation experiments were conducted to verify the effectiveness of the proposed binarization model.

The exceptional performance advantages of the fiber optic gyroscope (FOG) position it as a dominant player in middle and high-end inertial navigation systems. To prevent the loss of sensor precision caused by algorithm design and simplify the complex modeling strategy in traditional methods. We gradually demonstrate the significant role of Convolutional Neural Network (CNN) in the navigation system based on FOG, and utilize the particle swarm optimization algorithm (PSO) to expedite the convergence of the network. The experimental results demonstrate that the initial alignment method based on deep learning is more accurate than the traditional method. The attitude angle error is reduced by 81.25%, 92.54% and 36.53% respectively. The research provides support for the future application of deep learning in optical navigation systems.

Numerous image processing techniques (IPTs) have been employed to detect crack defects, offering an alternative to human-conducted onsite inspections. These IPTs manipulate images to extract defect features, particularly cracks in surfaces produced through Additive Manufacturing (AM). This article presents a vision-based approach that utilizes deep convolutional neural networks (CNNs) for crack detection in AM surfaces. Traditional image processing techniques face challenges with diverse real-world scenarios and varying crack types. To overcome these challenges, our proposed method leverages CNNs, eliminating the need for extensive feature extraction. Annotation for CNN training is facilitated by LabelImg without the requirement for additional IPTs. The trained CNN, enhanced by OpenCV preprocessing techniques, achieves an outstanding 99.54% accuracy on a dataset of 14,982 annotated images with resolutions of 1536 × 1103 pixels. Evaluation metrics exceeding 96% precision, 98% recall, and a 97% F1-score highlight the precision and effectiveness of the entire process.

Ocean exploration is crucial for utilizing its extensive resources. Images captured by underwater robots suffer from issues such as color distortion and reduced contrast. To address the issue, we propose an innovative enhancement algorithm that integrates Transformer and Convolutional Neural Network (CNN) in a parallel fusion manner. Firstly, a novel transformer model is intro-duced to capture local features, employing peak-signal-to-noise ratio (PSNR) attention and linear operations. Subsequently, to extract global features, both temporal and frequency domain features are incorporated to construct convolutional neural network. Finally, the Fourier’s high and low-frequency information of the original image are utilized to fuse different features. To demon-strate the algorithm's effectiveness, underwater images with various levels of color distortion are selected for both qualitative and quantitative analyses. The experimental results demonstrate that our approach surpasses other mainstream methods, achieving superior PSNR and structural sim-ilarity index measure (SSIM) metrics and leading to a detection performance improvement of over ten percent.

We propose novel spectrometer designs that aim to enhance the measured spectral range 1 of ions on a finite-sized detector. In contrast to the traditional devices that use uniform magnetic 2 field, in which the deflection of particles increases inversely proportional to their momentum, in a 3 gradient magnetic field, the deflection of particles will be decreased due to the reduction of the 4 magnetic field along their propagation. Low energy ions are particularly impacted, as they are 5 deflected less and are more likely to reach the detector compared to being driven out of it while 6 using a uniform magnetic field. By the means of a gradient magnetic field, the nonlinear dispersion 7 of ions in a homogeneous magnetic field can approach a nearly linear dispersion. Nonetheless, 8 the dispersion of low energy ions in a homogeneous magnetic field remains unnecessarily high. 9 In this article, we explore linear, quadratic, and higher-order field profiles in comparison to the 10 homogeneous field case. We discuss the employed methodology and present simulation results of 11 the spectrometer with an extended ion spectral range, focusing on the minimum detectable energy 12 (energy dynamic range) and energy resolution

The rapid advancements in satellite technology have led to a significant increase in high-resolution remote sensing (RS) images, necessitating advanced processing methods. Additionally, a patent analysis revealed a significant increase in deep learning and machine learning applications in remote sensing, highlighting the growing importance of these technologies. Therefore, this paper introduces the Kolmogorov-Arnold Network (KAN) model to remote sensing, aiming to enhance efficiency and performance in RS applications. We conducted several experiments to validate KAN's applicability, starting with the EuroSAT dataset, where we combined the KAN layer with multiple pretrained CNN models. The optimal performance was achieved with ConvNeXt, leading to the development of the KonvNeXt model. KonvNeXt was evaluated on the Optimal-31, AID, and Merced datasets for validation, and it achieved accuracies of 90.59%, 94.1%, and 98.1%, respectively. The model also showed fast processing speed, with the Optimal-31 and Merced datasets completed in 107.63 seconds each, while the bigger and more complicated AID dataset took 545.91 seconds. This result is meaningful since it achieved faster speeds and comparable accuracy compared to the existing study which utilized VIT and proved KonvNeXt's applicability for remote sensing classification tasks. Furthermore, we investigated the model's interpretability by utilizing Occlusion Sensitivity and by displaying the influential regions, it validated its potential use in a variety of domains including medical imaging and weather forecasting. This paper is meaningful in that it is the first use of KAN in remote sensing classification, proving its adaptability and efficiency.

Health management in industrial systems is more critical in maintenance management and it plays an important role in productivity, fault diagnosis, safety, efficiency, and economy in manufacturing industries. Early detection of faults in machinery may increase the effectiveness of maintenance actions and will avoid unwanted consequences in process operations and maintenance. Existing fault diagnosis methods have limitations such as insufficient accuracy, slow detection rate, and handling large and complex data sets. In this digital age, Industry 4.0 techniques have been applied across all fields, including the condition monitoring of machines. This research addresses the gaps in traditional fault diagnosis by using deep learning, a modern AI technique effective for diagnosing faults in various machines. Deep learning algorithms Multilayer Perceptron (MLP), Coevolutionary Neural Network (CNN), and Recurrent Neural Network (RNN) with Long Short-Term Memory (LSTM) are tested for fault diagnosis using vibration datasets collected from Spectra Quest Machinery Fault Simulator (SQMFS). In this research work, NI-DAQ (National Instruments- Data Acquisition) system, accelerometer, and LabVIEW software are used to collect vibration signals. Preprocessing of the signals has been done using a sampling strategy, shuffling, standardization, and reshaping data augmentation. The result shows that MLP accuracy in the prediction of fault is 0.9, CNN reached 0.95, and RNN and LSTM with 0.57 and 0.45 respectively. The high performance of CNN is due to its ability to effectively capture spatial patterns in vibration data which is crucial for fault diagnosis in rotating machinery followed by MLP due to its faster convergence during training. However, when scaling the data, MLP outperformed CNN, demonstrating superior adaptability to increased data complexity and volume. Due to the need for larger datasets and temporal patterns in the vibration data, which RNN and LSTM are designed to handle, they resulted in a lower accuracy. This study shows that CNN has given better results than other deep learning algorithms MLP, RNN, and LSTM in fault diagnosis of rotating machinery. Future research could explore the application of these techniques to different types of machinery and fault conditions.

This paper proposed emotion recognition methods for consecutive facial images, and implements the inference of neural network model on a field-programmable gate array (FPGA). The proposed emotion recognition methods are based on a neural network model architecture that combines convolution neural networks (CNNs), long short-term memory (LSTM), and fully connected neural networks (FCNNs), called CLDNN or ConvLSTM-FCN. This type of neural network mod-el can analyze the local feature sequences obtained through convolution of data, making it suita-ble for processing time-series data such as consecutive facial images. In this paper, sequences of facial images are sampled from videos corresponding to various emotional state of subjects. The sampled images are then pre-processed with the processes includes facial detection, greyscale conversion, resize, and data augmentation if necessary. The 2-D CNN in ConvLSTM-FCN is used for feature extraction for these pre-processed facial images. These sequences of facial im-age features are time sequences with dependent properties between the elements within them. The LSTM is then used to model these time sequences followed by fully connected neural net-works (FCNNs) for classification. The proposed consecutive facial emotion recognition method achieves an average recognition rate of 99.51% on RAVDESS dataset, 87.80% on BAUM-1s dataset and 96.82% on eNTERFACE’05 data set, using 10-fold cross-validation on the PC. Some com-parisons of recognition accuracies between the proposed method and the other existing related works are conducted in this paper. According to the comparisons, the proposed emotion recog-nition methods outperform the existing related researches. The proposed emotion recognition methods are then implemented on an FPGA chip using the neural network model inference algo-rithms in this paper, and the accuracies of the experiments conducted on the FPGA chip are iden-tical to those obtained on the PC. This verifies that the proposed neural network model imple-mented on the FPGA chip performs well.

Seaports are critical hubs in global trade, facing increasing pressure to accelerate green transi-tions while navigating evolving economic and environmental landscapes. This research analyses the intricate relationship between port organizational ecosystem resilience and the successful implementation of green transitions, highlighting the challenges and opportunities in moving towards a more sustainable future. The research employed a combined focus group interview and participatory observation methodology to gather insights from experts. Analysis revealed that while technological advancements are a primary driver for greening port operations, a lack of managerial capacity is a significant impediment to the successful integration of these innova-tions. It was identified several challenges, including outdated infrastructure, cybersecurity threats, and the need for ongoing investments in new technologies, along with the need to foster a sustainable behaviour culture. Experts emphasized the importance of developing managerial competencies and continuous learning. The research also highlights the role of strong stake-holder partnerships, including collaboration with government agencies, industry associations, and environmental organizations. This research provides valuable insights for policymakers, port managers, and other stakeholders seeking to navigate the challenges for ensuring a sus-tainable future for the maritime sport organizational ecosystems and maritime industry.

This study explores novel techniques to improve the detection accuracy of skarn iron deposits using advanced image processing methodologies. Leveraging the capabilities of Aster image, Band Ratio (BR) images, and Principal Component Analysis (PCA), alongside the power of 3D Convolutional Neural Networks (3D-CNN), the research aims to enhance the precision and efficiency of ore detection in complex geological environments. The proposed method employs a specific 3D-CNN architecture, accepting input as a 7×7×C image patch, where C represents the combined number of selected Aster image bands, Principal Component (PC) bands, and computed BR images. To evaluate the accuracy of the proposed method, five distinct image bands combinations, including the proposed bands combination, were tested and evaluated based on the overall accuracy (OA), average accuracy (AA), and kappa coefficient. The results demonstrated that while the incorporation of BR images alongside Aster bands initially seemed promising, it introduced significant confusion in certain classifications, leading to unexpected mis-classification rates. Surprisingly, utilizing solely Aster bands as input parameters yielded higher accuracy rates (OA= 93.13%, AA = 91.96%, Kappa = 90.91%) compared to scenarios involving the integration with band ratios (OA = 87.02%, AA = 79.15, and Kappa = 82.60%) or the integration of BR images to PC bands (OA= 87.78%, AA = 82.39%, Kappa = 83.81%). However, the amalgamation of Aster bands with selected PC bands showed slight improvements in accuracy (OA= 94.65%, AA = 92.93%, Kappa = 93.45%), although challenges in accurately classifying certain features persisted. Ultimately, the proposed combination of Aster bands, PC bands, and BR images (proposed bands combination) presented the most visually appealing and statistically accurate results (OA = 96.95%, AA= 94.87%, and Kappa = 95.93%), effectively addressing misclassifications observed in the other combinations. These findings underscore the synergistic contributions of each of the Aster bands, PC bands, and BR images, with Aster bands proving pivotal for optimal skarn classification, PC bands enhancing intrusions classification accuracy, and BR images strengthening wall rock classification accuracy. In conclusion, the proposed combination of input image bands emerges as a robust and comprehensive methodology, demonstrating unparalleled accuracy in the remote sensing detection of skarn iron minerals.

In response to the burgeoning global demand for premium agricultural products, particularly within the competitive nut market, this paper introduces an innovative methodology aimed at enhancing the grading process for almonds and their shells. Leveraging state-of-the-art Deep Convolutional Neural Networks (CNNs), specifically the AlmondNet-20 architecture, our study achieves exceptional accuracy exceeding 99%, facilitated by the utilization of a 20-layer CNN model. To bolster robustness in differentiating between almonds and shells, data augmentation techniques are employed, ensuring the reliability and accuracy of our classification system. Our model, meticulously trained over 1000 epochs, demonstrates remarkable performance, boasting an accuracy rate of 99% alongside a minimal loss function of 0.0567. Rigorous evaluation through test datasets further validates the efficacy of our approach, revealing impeccable precision, recall, and F1-score metrics for almond detection. Beyond its technical prowess, this advanced classification system offers tangible benefits to both industry experts and non-specialists alike, ensuring globally reliable almond classification. The application of deep learning algorithms, as showcased in our study, not only enhances grading accuracy but also presents opportunities for product patents, thereby contributing to the economic value of our nation. Through the adoption of cutting-edge technologies such as the AlmondNet-20 model, we pave the way for future advancements in agricultural product classification, ultimately enriching global trade and economic prosperity.

Machine unlearning, the process of removing the influence of specific data points from trained machine learning models, has become increasingly important in light of modern data privacy regulations, such as the GDPR and the "right to be forgotten." This paper explores the challenges and solutions associated with implementing machine unlearning in three widely used neural network architectures: Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), and Graph Neural Networks (GNNs). Each of these networks presents unique challenges due to their distinct architectures and learning processes. Techniques such as selective retraining, influence functions, and knowledge distillation have been proposed to address these challenges. The paper also introduces the concept of full parameter unlearning, which adjusts all trainable parameters using two key techniques: gradient ascent based on first-order information and Fisher information based on second-order information. These methods ensure comprehensive unlearning, but also introduce computational complexity. We discuss examples, potential solutions, and future research directions to make full parameter unlearning more scalable and efficient, thus providing a framework for balancing data privacy with model performance.

In this paper, the mechanism of combined process of ion beam sputtering (IBS) and HF acid etching on the chemical structure defects of fused silica and its laser damage resistance perfor-mance were investigated. During the removal process of surface material, the sputtering effect causes lattice atoms to detach from their original positions, forming a large number of unsaturated structures on silica surface, which improves the chemical activity of Si and O atoms, accelerates the chemical reaction process between surface atoms and water molecules, increases the content of hydroxyl (OH-) groups in the shallow layer, and enhances the photothermal weak absorption intensity. However, the presence of OH- groups reduces the bonding strength of Si-O, destroys the spatial network structure of silica bulk, and reduces the mechanical strength of silica surface, resulting in a decrease in its laser damage resistance performance. The paper reveals for the first time the mechanism of IBS to change the structure characteristics of silica material, accelerate the surface hydroxylation process, and thereby reduces the laser damage resistance performance. This work provides technical guidance for effectively suppressing chemical structure defects on silica surface and improving laser damage resistance performance of optical components under high-flux laser irradiation.

In this paper, we explore the use of superluminal ionization fronts to accelerate and amplify electromagnetic radiation. These fronts are defined as optical boundaries between two regions of a gas, the neutral region and the plasma region, characterized by two different values of the refractive index. For that reason, the front velocity is not necessarily related with the motion of material particles, such as neutral atoms, ions and electrons, which can stay at rest. The fronts can therefore become superluminal without violating causality. In recent years, different experimental configurations, such as the flying-focus, showed that it is possible to create superluminal fronts in the laboratory. These fronts can easily be described theoretically in a special reference frame, called the time-frame, which is used here. In this frame, superluminal fronts reduce to time-refraction, a process that is symmetrical to the well-known optical refraction. We show that propagation through such fronts can lead to considerable frequency shift and energy amplification of probe laser beams. This could eventually be used to develop new sources of turnable radiation.

In this paper, inspired by quantum field theory, or more specifically QED, we propose a dynamical model for relativity. By adopting the approach provided by this dynamical model, we provide a dynamical explanation for relativistic phenomena such as length contraction, time dilation and clock desynchronization. The result of the dynamical approach for clock desynchronization is especially interesting. Our dynamical approach exhausts the freedom for different clock synchronizations on frames accelerated from an initial reference frame, which is allowed by Einstein’s principle approach. Furthermore, our dynamical approach arrives at the interesting conclusion that Einstein’s equivalence principle and gravitational redshift are necessary to explain special relativistic clock desynchronization. This result shows that special relativity by itself is an incomplete theory. If our dynamical model assumptions are assumed to be valid at a fundamental level, it implies some modifications to quantum field theory and general relativity via the equivalence principle. These modifications have been discussed.

The article presents an innovative method for synthesizing radiation patterns efficiently by 1 combining the Taguchi method and neural networks, while validating the results on a 10-element 2 antenna array. The Taguchi method aims to minimize product and process variability, while neural 3 networks are used to model the relationship between antenna design parameters and radiation 4 pattern characteristics. This approach utilizes Taguchi parameters as inputs for the neural network, 5 which is then trained on a dataset generated by the Taguchi method. After training, the network is 6 validated using a real 10-element antenna array. Analytical results demonstrate that this method 7 enables efficient synthesis of radiation patterns with a significant reduction in computation time 8 compared to traditional approaches. Furthermore, validation on the antenna array confirms the 9 accuracy and robustness of the approach, showing a high correlation between predicted performances 10 by the neural network model and actual measurements on the antenna array.In summary, our article 11 highlights that the combined use of the Taguchi method and neural networks, with validation on a 12 real antenna array, offers a promising approach for efficient synthesis of antenna radiation patterns. 13 This approach combines speed, accuracy, and reliability in antenna system design.

Acceleration rates of vehicles are crucially important statistics needed in two main aspects: first, they are essential to incident reconstructor professionals as having a verified database on where basing the cinematic reconstructions is definitely difficult and moreover, they are useful in order to understand how to create the safest possible environment on the road. This key topic gives es-sential information for road design like intersection schemes, acceleration and deceleration lane length as well as traffic simulation modelling and furthermore, to educate vehicle drivers when entering the traffic flow and to provide limit acceleration values needed to develop autonomous vehicle’s software in order to guarantee a comfort drive. Literature reports tons of linear acceleration stats mainly referred to vehicle performances claimed by car manufacturers, but these data don’t recreate real drivers’ behaviour when entering the road by turning left or right or crossing an intersection. This work aims to integrate the most reliable data from literature with a hybrid and an electric vehicles statistic, adding also two contemporary thermal engine vehicles, extracted conducting probes that recreates real road environment. The entirety of these experiments reconstructs the scenario in which a car is entering the traffic of a two-lanes road, from a stationary position at a stop sign and the acceleration rates are collected using a M.E.M.S. (Microelectromechanical system). Finally, the space needed between the car that is about to enter or crossing the road and the car that is approaching the intersection is calculated, basing the computations on the acceleration values extracted from the probes. These spaces are pictured, as a way to give a concrete idea to readers.

In this article it is shown how Milgrom’s acceleration constant is related with the Ω_Λ, Ω_m and Ω_B values of the matter distribution within observable universe as defined in the Standard Model of Cosmology, established by the Planck Collaboration Group. The result is a two-parameter model for the observable universe. The two parameters are the age of the universe and Milgrom’s acceleration constant. It is shown that these are sufficient to calculate the amounts of matter and dark energy in the universe, as well as the contributions of dark matter and baryonic matter in the matter part. The numerical results are in agreement with those of the Lamda-CDM model.

Structural displacement is an important metric for assessing structural conditions because it has a direct relationship with the structural stiffness. Many bridge displacement measurement techniques have been developed, but most methods require fixed reference points in the vicinity of the target structure which limits field implementations. A promising alternative is to use reference-free measurement techniques that indirectly estimate the displacement by using measurements such as acceleration, and strain. This paper proposes novel reference-free bridge displacement estimation by the fusion of single acceleration with pseudo-static displacement derived from co-located strain measurements. First, we propose a conversion of the strain at the center of a beam into displacement based on the geometric relationship between strain and deflection curves with reference-free calibration. Second, an adaptive Kalman filter is proposed to fuse the displacement generated by strain with acceleration by recursively estimate the noise covariance of displacement from strain measurements which is vulnerable to measurement condition. Both numerical and experimental validations are presented to demonstrate the efficiency and robustness of the proposed approach.

Traditional population-based metaheuristic algorithms are effective in solving complex real-world problems but require careful strategy selection and parameter tuning. Parameter-less population-based optimization algorithms have gained importance due to their simplicity and efficiency. However, research on their applicability for solving large systems of nonlinear equations is still incipient. In this paper, a systematic review and detailed description of metaphor-less optimization algorithms is made, including the Jaya algorithm, the main Jaya variants for continuous optimization problems, with a focus on the Enhanced Jaya (EJAYA) algorithm, the three Rao algorithms, the best-worst-play (BWP) algorithm, and the very recent max-min-greedy-interaction (MaGI) algorithm. This paper also discusses a few previous parallelizations of the Jaya optimization algorithm, as well as recent GPU-based massively parallel implementations of the Jaya, Enhanced Jaya (EJAYA), Rao, and BWP algorithms developed by the authors. In addition, a novel GPU-accelerated version of the MaGI algorithm is proposed. The GPU-accelerated versions of the metaphor-less algorithms developed by the authors were implemented using the Julia programming language and tested primarily on GeForce RTX 3090 and NVIDIA A100 GPUs, as well as on Tesla V100S and Tesla T4 GPUs, using a set of difficult, large-scale nonlinear equation system problems. The computational experiments carried out produced quite significant speedups, which highlights the efficiency of the GPU-based versions of the metaphor-less algorithms developed for solving large-scale systems of nonlinear equations.

Identification and segmentation of rock particles is very important in geology, mining, engineering construction, and environmental monitoring. Traditional methods for recognizing the rock particle characteristics are time-consuming, labor-intensive, and limited by the geographical scope of the sampling sites, which makes it more difficult to represent the spatial distribution of the rock particles accurately. Recent advancements in deep learning techniques offer promising solutions to automate this process. This study aims at utilizing state-of-the-art (SOTA) Mask R-CNN deep learning algorithm from the Detectron2 framework, to capture global features of rocks that will enable the recognition of rock particles more efficiently, just from having the rock images. These images are subject to image processing techniques such as a Gaussian filter, to denoise and smoothen the image, and an Illumination Adaptive Transformer (IAT) framework, to adjust the lighting exposure; all of which add to improve the quality of the image dataset. Afterward, the Mask R-CNN and Detectron2 models are utilized in training the images using transfer learning. Experimental results from evaluating the performance of the proposed algorithms showcase the effectiveness of the approach, thereby highlighting its potential to revolutionize rock particle segmentation across various domains.

Hand gestures are an essential part of human-to-human communication and interaction, and therefore for technical applications, the aim is increasingly to achieve interaction between humans and computers that is as natural as possible, for example by means of natural language or hand gestures. In the context of human-machine interaction research, these methods are consequently being explored more and more. However, the realization of natural communication between humans and computers is a major challenge. In the field of hand gesture recognition, research approaches are being pursued that use additional hardware such as special gloves to classify gestures with high accuracy. Recently, deep learning techniques using artificial neural networks have been increasingly proposed for the problem of gesture recognition without using such tools. In this context, in our approach convolutional neural network (CNN) will be explored in detail for the task of hand gesture recognition. CNN is a deep neural network that can be used in the field of visual object processing and classification. The goal of this work is to recognize ten types of static hand gestures in front of complex backgrounds and different hand sizes based on raw images without the use of extra hardware. We achieved good results with a CNN network architecture consisting of seven layers. Through data augmentation and skin segmentation, a significant increase of the model accuracy was achieved. On public benchmarks the ten gestures have been classified almost perfectly with a testing accuracy of 96.5%

Convolutional neural network (CNN), a class of artificial neural network (ANN) is attracting interests of researchers in all research domain. CNN was invented for computer vision. They have also shown to be useful for semantic parsing, sentence modeling and other natural language processing related tasks. Here in this paper we discuss the basics of CNN models and their scope to provide a reference/baseline to the researchers interested in using CNN models in their research.

The paper demonstrates the analysis of Corona Virus Disease based on a CNN probabilistic model. It involves a technique for classification and prediction by recognizing typical and diagnostically most important CT images features relating to Corona Virus. The main contributions of the research include predicting the probability of recurrences in no recurrence (first time detection) cases at applying our proposed Convolution neural network structure. The Study is validated on 2002 chest X-ray images with 60 confirmed positive covid19 cases and (650 bacterial – 412 viral -880 normal) x-ray images. The proposed CNN compared with traditional classifiers with proposed CHFS feature extraction model. The experimental study has done with real data demonstrates the feasibility and potential of the proposed approach for the said cause. The result of proposed CNN structure has been successfully done to achieve 98.20% accuracy of covid19 potential cases with comparable of traditional classifiers.

In recent years, large language models have demonstrated remarkable performance across various natural language processing (NLP) tasks. However, deploying these models for real-world applications often requires efficient inference solutions to handle the computational demands. In this paper, we explore the utilization of CPUs for accelerating the inference of large language models. Specifically, we introduce a parallelized approach to enhance throughput by 1) Exploiting the parallel processing capabilities of modern CPU architectures, 2) Batching the inference request. Our evaluation shows the accelerated inference engine gives an 18-22x improvement in the generated token per sec. The improvement is more with longer sequence and larger models. In addition to this, we can also run multiple workers in the same machine with NUMA node isolation to further improvement in tokens/s. Table 2, we have received 4x additional improvement with 4 workers. This would also make Gen-AI based products and companies’ environment friendly, our estimates shows that CPU usage for Inference could reduce the power consumption of LLMs by 48.9% (1252 W for A100 with AMD EPYC 7V13 vs 613 W for Intel® Xeon® Gold 6538N) while providing production ready throughput & latency.

We measure the accelerated cosmic expansion rate of the late-time Universe and derive constraints on cosmological parameters from the final baryon acoustic oscillation (BAO) uncorrelated measurements of different surveys after two decades of dedicated spectroscopic observation combined with the Pantheon sample of Type Ia supernovae, the Hubble diagram of gamma-ray bursts & quasars data, the Cosmic Chronometers data, and the recent Hubble constant value measurement from the Hubble Space Telescope and the SH0ES Team as an additional prior. In Λ cold dark matter (ΛCDM) scenario, the model fit yields Ωm=0.313±0.034 and ΩΛ=0.672±0.025. Combining BAO with CC+Pantheon+QSR+GRB data sets we get H0=69.21±1.22 km s−1 Mpc−1, rd=133.46±2.49 Mpc. For the a flat wCDM model, we get w=−1.111±0.030. For a dynamical evolution of dark energy in a flat w0waCDM cosmology, we get wa=−0.292±0.167. We apply the Akaike information criteria probe to compare the three models, and see that all models are favored and cannot be discarded from the current data.

This work focuses on short-term forecasting of PV output power. The multilayer perception (MLP), convolutional neural networks (CNN), and k-nearest neighbour (kNN) neural networks have been used singly or in a hybrid (with other algorithms) to forecast solar PV power or global solar irradiance with good success. Their performances in forecasting PV power have been compared with other algorithms but not with themselves. The study aims to compare performance of a number of neural network algorithms in solar PV energy yield forecasting under different weather conditions. The performance of MLPNN, CNN and kNN are being compared using solar PV (hourly) energy yield data for Grahamstown, Eastern Cape, South Africa. The choice of location is part of the study parameters to provide insight into renewable energy power integration in specific areas in south Africa that may be prone to extreme weather conditions. The kNN algorithm was found to have RMSE value of 4.95% and MAE value of 2.74% at its worst performance, and RMSE value of 1.49% and MAE value of 0.85% at its best performance. It outperformed the others by a good margin and kNN could serve as a fast, easy, and accurate tool to forecast solar PV output power.

In order to improve real-time performance of gesture recognition by micro-Doppler map of mmWave Radar, the point cloud based gesture recognition for mmWave Radar is proposed in this paper.Two steps are carried out for mmWave Radar based gesture recognition. The first step is to estimate the point cloud of the gestures by 3D-FFT and the peak grouping. The second step is to train the TRANS-CNN model by combining the multi-head self-attention and the 1D-convolutional network, so as to extract the features in the point cloud data at a deeper level to categorize the gestures. In the experiments, TI mmWave Radar sensor IWR1642 is used as benchmark to evaluate the feasibility of the proposed approach. The results show that the accuracy of the gesture recognition reaches 98.5%. In order to prove the effectiveness of our approach, a simply 2Tx2Rx Radar sensor is developed in our lab and the accuracy of recognition reaches 97.1%. The results show that our proposed gesture recognition approach achieves the best performance in real-time with limited training data, in comparison with the existing methods.

A method of direction-of-arrival (DoA) estimation for FMCW radars is presented. In addition to MUSIC and ESPRIT, which are well-known high-resolution DoA estimation algorithms, deep learning has recently emerged as a very promising alternative. It is proposed in this paper to use a 3D convolutional neural network (CNN) for DoA estimation. The 3D-CNN extracts from the radar data cube spectrum features of the region of interest (RoI) centered on the potential positions of the targets, thereby capturing the spectrum phase shift information, which corresponds to DoA, along the antenna axis. Finally, the results of simulations and field experiments are provided to demonstrate the superior performance, as well as the limitations, of the proposed 3D-CNN.

Automatic speech recognition of native languages plays a crucial role in fostering inclusivity and preserving linguistic diversity. The Kashmiri language, an underrepresented Indo-Aryan dialect primarily spoken in the Kashmir Valley, poses substantial challenges for existing AI models due to its phonetic diversity and scant linguistic resources. This study addresses these hurdles by developing a robust spoken Kashmiri recognition system that employs dual feature extraction with spectrogram augmentation and a hybrid Convolutional Neural Network (CNN) and Gated Multi-Layer Perceptron (gMLP) model. Key to this endeavour is the creation of a high-fidelity dataset that captures the phonetic variations across Kashmiri dialects, focusing on twelve specific words. Through the integration of dual feature extraction, spectrogram augmentation, and the innovative hybrid modelling approach, our system attains an impressive 96% accuracy on the test dataset for classifying these twelve spoken words. This research not only enhances the generalization and resilience of spoken Kashmiri recognition systems but also represents a critical step towards advancing technology and safeguarding the Kashmiri language within this underrepresented linguistic domain.

Focusing on the problem of identifying and classifying aero-engine models, this paper measures the infrared spectrum data of aero-engine hot jets using a telemetry Fourier transform infrared spectrometer. Simultaneously, infrared spectral data sets with the six different types of aero-engines are created. For the purpose of classifying and identifying infrared spectral data, a CNN architecture based on the continuous wavelet transform peak seeking attention mechanism (CWT-AM-CNN) is suggested. This method calculates the peak value of middle wave band by continuous wavelet transform, and the peak data is extracted by the statistics of the wave number locations with high frequency. Attention mechanism is used for the peak data, and the attention mechanism is weighted to the feature map of the feature extraction block. The training set, vali-dation set and prediction set are divided in the ratio of 8:1:1 for the infrared spectral data sets. For three different data sets, CWT-AM-CNN proposed in this paper is compared with the classical classifier algorithm based on CO2 feature vector and the popular AE, RNN and LSTM spectral processing networks. The prediction accuracy of the proposed algorithm in the three data sets is as high as 97%, and the lightweight network structure design not only guarantees high precision, but also has a fast running speed, which can realize the rapid and high-precision classification of the infrared spectral data of the aero-engine hot jets.

In a previous series of papers relating to the Combined Gravitational Action (CGA), we have exclusively studied orbital motion without spin. In the present paper, we apply CGA to any self-rotating material body, i.e., an axially spinning massive object, which itself may be locally seen as a gravito-rotational source because it is capable of generating the gravito-rotational acceleration, which seems to be unknown to previously existing theories of gravity. The consequences of such an acceleration are very interesting, particularly for Compact Stellar Objects. Independently of the equation of state, it is found that the minimum radius of a stable neutron star is three times its gravitational radius, R_min = 3GM_NS/c², and its critical and maximum internal magnetic field strength cannot exceed the value of 3×10¹⁸ G.

Accurately predicting the distance an object will travel to its destination is very important in various sports. Acceleration sensors as a means of real-time monitoring are increasingly gaining increasing attention in sports. Due to the low energy output and power density of Triboelectric Nanogenerators (TENGs), recent efforts have focused on developing various acceleration sensors. However, these sensors suffer from significant drawbacks of large size, complexity, high power input requirements, and high cost. Here, we described a portable and cost-effective real-time refreshable strategy design comprising a series of individually addressable and controllable units based on TENGs embedded in a flexible substrate. This results in a highly sensitive, low-cost, and self-powered acceleration sensor. Putting, which accounts for nearly half of the strokes played, is obviously an important component of the golf game. The developed acceleration sensor has an accuracy controlled within 5 %. The initial velocity and acceleration of the forward movement of a rolling golf ball after it is hit by a putter can be displayed, and the stopping distance quickly calculated and predicted in about 7 seconds. This research demonstrates the application of the portable TENG-based acceleration sensor, while paving the way for designing portable, cost-effective, scalable, and harmless ubiquitous self-powered acceleration sensors.

We show that the accelerating expansion phase of the universe can emerge from the group field theory formalism, a candidate theory of quantum gravity. The cosmological evolution can be extracted from condensate states using mean field approximation, in a form of modified FLRW equations. By introducing an effective equation of state w, we can reveal the relevant features of the evolution, and show that with proper choice of parameters, w will approach to −1, corresponds to the behaviour of cosmological constant, results in a late time acceleration and leads to de Sitter spacetime asymptotically.

Consider a robot that is navigating in a space modeled by a graph, and that wants to know its current location. It can send a signal to determine how far it is from each landmark among a set of fixed landmarks. We study the problem of computing the minimum required number of landmarks, and where they should be placed so that the robot can always determine its location. Since the problem is an NP-complete problem, the robot's responses to the actions are slow. To accelerate this response, we can use the parallel version of this problem. In this work, we introduce a new parallel implementation for determining the metric dimension of a given graph. We run the proposed algorithm on a symmetric multi-processing (SMP) cluster using C programming language and the Message Passing Interface (MPI) library. Finally, we run our implementation on four categories of graphs (the tracks in which the robot moves): a cycle graph Cn, a path graph Pn, a triangular snake graph and a ladder graph Ln. Preliminary computational results indicate that the metric dimension problem is an NP-complete problem and prove the ability of the proposed algorithm to achieve a speedup of 6 for 8 processors.

Emotion detection holds significant importance in facilitating human-computer interaction, enhancing the depth of engagement. By integrating this capability, we pave the way for forthcoming AI technologies to possess a blend of cognitive and emotional understanding, bridging the divide between machine functionality and human emotional complexity. This progress has the potential to reshape how machines perceive and respond to human emotions, ushering in an era of empathetic and intuitive artificial systems. This paper introduces a novel approach to multimodal emotion recognition, seamlessly integrating speech and text modalities to accurately infer emotional states. Employing CNNs, we meticulously analyze speech using Mel spectrograms, while a BERT-based model processes the textual component, leveraging its bidirectional layers for profound semantic comprehension. The outputs from both modalities are combined using an attention-based fusion mechanism that optimally weighs their contributions. The proposed method undergoes meticulous testing on two distinct datasets: Carnegie Mellon University’s Multimodal Opinion Sentiment and Emotion Intensity (CMU-MOSEI) dataset and the Multimodal Emotion Lines Dataset (MELD). The results demonstrate superior efficacy compared to existing frameworks, achieving an accuracy of 88.4% and an F1-score of 87.9% on the CMU-MOSEI dataset, and a notable weighted accuracy (WA) of 67.81% and a weighted F1 (WF1) score of 66.32% on the MELD dataset. This comprehensive system offers precise emotion detection and introduces several significant advancements in the field.

Stock price prediction is of great significance but faces numerous challenges. In this study, a hybrid CNN-LSTM model was utilized. The stock data were processed and feature engineering was conducted. The model architecture and training strategies were expounded. Experimental results demonstrated that this model outperformed traditional methods and benchmark models on the test set, featuring strong capabilities in feature extraction and handling long-term dependencies. Key roles of data preprocessing, hyperparameter adjustment, and model fusion were also summarized, providing valuable references for stock price prediction.

Speaker recognition methods based on convolutional neural networks (CNN) have been widely used in the security field and smart wearable devices. However, the traditional CNN has a large number of hyperparameters that are difficult to be determined, which makes the model easy to fall into local optimum or even fail to converge during the training process. Intelligent algorithms such as particle swarm optimization and genetic algorithm are used to solve the above problems. However, these algorithms have poor performance compared with the current emerging meta-heuristic algorithms. In this study, the dung beetle optimized convolution neural network (DBO-CNN) is proposed to identify the speakers, which is helpful in finding suitable hyperparameters for training. By testing the dataset of 50 people, it was demonstrated that the accuracy of the model was significantly improved by using this approach. Compared with the traditional CNN and CNN optimized by other intelligent algorithms, the accuracy of DBO-CNN has increased by 0.6%~4.8%, and reached 98.3%.

An application of CO2/HCO3--free solution (Zero-CO2) did not increase intracellular pH (pHi) in ciliated human nasal epithelial cells (c-hNECs), leading to no increase in the beating frequency (CBF) or bend distance (CBD, an index of amplitude). This study demonstrated that the pHi of the c-hNECs expressing carbonic anhydrase IV (CAIV) is high (7.64), while that of the ciliated human bronchial epithelial cells (c-hBECs) expressing no CAIV is low (7.10). Upon applying Zero-CO2, an extremely high level of pHi caused the pHi, CBF and CBD to decrease, while a low level of pHi caused them to increase. The high pHi was generated by a high rate of HCO3- influx via interac-tions between CAIV and Na+/HCO3- cotransport (NBC) in c-hNECs. An NBC inhibitor (S0859) decreased the pHi and CBF, and Zero-CO2 application increased the CBF in S0859-treated c-hNECs. In conclusion, the interactions of CAIV and NBC maximize the HCO3- influx to increase the pHi in c-hNECs. This novel mechanism causes the pHi to decrease, leading to no increase in CBF and CBD in c-hNECs upon Zero-CO2 application, and appears to play a crucial role in maintaining pHi, CBF and CBD in c-hNECs periodically exposed to air (0.04% CO2) with respiration.

In recent years, Edge Intelligence (EI) has emerged, combining edge computing with AI, specifically deep learning, to run AI algorithms directly on edge devices. In practical applications, EI faces challenges related to computational power, power consumption, size, and cost, with the primary challenge being the trade-off between computational power and power consumption. This has rendered traditional computing platforms unsustainable, making heterogeneous parallel computing platforms a crucial pathway for implementing EI. In our research, we leveraged the Xilinx Zynq 7000 heterogeneous computing platform, employed High-Level Synthesis (HLS) for design, and implemented two different accelerators for LeNet-5 using loop unrolling and pipelining optimization techniques. The experimental results show that when running at a clock speed of 100 MHz, the PIPELINE accelerator, compared to the UNROLL accelerator, experiences an 8.09% increase in power consumption but speeds up by 14.972 times, making the PIPELINE accelerator superior in performance. Compared to the CPU, the PIPELINE accelerator reduces power consumption by 91.37% and speeds up by 70.387 times, while compared to the GPU, it reduces power consumption by 93.35%.This study provides two different optimization schemes for edge intelligence applications through design and experimentation, and demonstrates the impact of different quantization methods on FPGA resource consumption. These experimental results can provide a reference for practical applications, thereby providing a reference hardware acceleration scheme for edge intelligence applications.

The design of curved surface sliders (CSS) based on the elastic response spectrum is done by iteration to find the combination of friction coefficient and displacement capacity which satisfies the condition that the maximum horizontal CSS force is equal to the horizontal force of the structure. Although this CSS design is valid it does not necessarily minimize structural acceleration. This paper therefore describes the optimum CSS design for minimum structural acceleration. All valid CSS designs and the optimum CSS design are represented by their associated trajectory in the elastic response spectrum plane which visualizes the optimization problem. The results demonstrate that the optimum CSS design is not obtained at maximum tolerated effective damping ratio. The subsequent sensitivity analysis describes how much the structural acceleration increases if the actual friction coefficient of the real CSS deviates from its optimum design value. The analysis points out that the increase in structural acceleration is approximately one order of magnitude smaller than the deviation in friction. The sensitivity data may be used by structural engineer to determine tolerable deviations in friction coefficient which still results in acceptable structural accelerations.

Recently, SMR (Small Modular Reactor) is expected to be able to operate flexibly in conjunction with a high renewable energy penetration grid as a result of improved safety and easy power control compared to large nuclear power plants. SMRs such as South Korea’s System-integrated Modular Advanced ReacTor (SMART) are designed to use canned motors (CMs) for their reactor coolant pumps (RCPs) to enhance safety. The CMs passively enhance the reactor’s safety by preventing leakage of radioactive reactor coolant. However, with motor sizes designed to be as small as possible, and the increased air gap of CMs between stator and rotor, starting torque may be insufficiently small compared to that of similar sized induction motors (IMs). Thusly, CMs may require variable frequency drives (VFDs) to start. This paper compares the torque characteristics of CMs with those of IMs for SMART’s RCPs. ETAP is then used to perform motor starting analysis for CMs started with and without VFDs. The results are presented and analyzed to find out if the VFDs deal with the CM starting torque issue.

Background: Fetal overgrowth is related to many perinatal complications including stillbirth, cesarean section, maternal and neonatal injuries, and shoulder dystocia. It is related to maternal diabetes, obesity, and gestational weight gain, but also happens in low-risk pregnancies. There is ongoing discussion regarding definitions, methods of detection, and classification. The method used for detection is of crucial importance as it draws a line between those at risk and low-risk populations. Methods: In this a narrative review of relevant evidence identified through PubMed search with one of the general terms (macrosomia, large-for-gestational-age) combined with the outcome of interest. Results: This revive summarizes evidence on the relation of fetal overgrowth with stillbirth, cesarean sections, shoulder dystocia, anal sphincter injury, and hemorrhage. Customized growth charts help to detect mothers and fetuses at risk of those complications. Relations between fetal overgrowth and diabetes, maternal weight, and gestational weight gain. Conclusions: a substantial proportion of complications are an effect of the fetus growing above its potential and should be recognized as a new dangerous condition of Fetal Growth Acceleration

Temperature and pressure sensors currently encounter challenges such as slow response times, large sizes, and insufficient sensitivity. To address these issues, we developed tetraphenylethylene (TPE)-doped polyvinylidene fluoride (PVDF) nanofiber membranes using electrospinning, with process parameters optimized through a convolutional neural network (CNN). We systematically analyzed the effects of PVDF concentration, spinning voltage, tip-to-collector distance, and flow rate on fiber morphology and diameter. The CNN model achieved high predictive accuracy, resulting in uniform and smooth nanofibers under optimal conditions. Incorporating TPE enhanced the hydrophobicity and mechanical properties of the nanofibers. Additionally, the fluorescent properties of the TPE-doped nanofibers remained stable under UV exposure and exhibited significant linear responses to temperature and pressure variations. The nanofibers demonstrated a temperature sensitivity of -0.976 gray value/°C and pressure sensitivity with an increase in fluorescence intensity from 537 a.u. to 649 a.u. under 600 g pressure. These findings highlight the potential of TPE-doped PVDF nanofiber membranes for advanced temperature and pressure sensing applications.

This article comprehensively explores multiple aspects of cryptocurrencies and their price forecasting. Firstly, the article introduces the definition of cryptocurrency and its development process on a global scale, especially focusing on the launch of Facebook Libra and China's central bank digital currency, highlighting the importance and influence of digital currency in the global financial market. Subsequently, the article analyzes the advantages of digital currencies over traditional currencies, including improving economic transaction efficiency, reducing transaction costs and enhancing transaction transparency. Meanwhile, the article also explores the challenges and risks potentially brought by the development of digital currencies, such as regulatory uncertainty and market volatility. In this context, the article raises the importance of cryptocurrency price forecasting and introduces the forecasting models and techniques commonly used today. Finally, through specific experimental analysis, the effectiveness of using the deep learning model CNN-LSTM to predict the price of Bitcoin is demonstrated, and the directions of future research and optimization strategy are proposed. In summary, this paper comprehensively presents the research status and prospects of cryptocurrency and its price prediction field through systematic introduction and analysis.

Multi-floor environments are usually ignored while designing an autonomous robot for indoor cleaning applications. However, for efficient operation in such environments, the ability of a robotic platform to traverse staircases is crucial. Staircase detection and localization is highly important for planning the traversal on staircases. This paper describes a deep learning approach using Convolutional Neural Networks (CNNs) based Robot Operation System (ROS) to staircase detection and localization. We use an object detection network to detect staircases in images. We also localize these staircases using a contour detection algorithm to detect the target point, a point close to the center of the first step, and the angle of approach to the target point. Experiments are performed with data obtained from images captured on different types of staircases at different view points/angles. Results show that the approach is very accurate in identifying the presence of the staircase in the working environment and is also able to locate the target point with good accuracy.

This research investigates the application of deep learning in sentiment analysis of Canadian maritime case law. It offers a framework for improving maritime law and legal analytics policy-making procedures. The automation of legal document extraction takes center stage, underscoring the vital role sentiment analysis plays at the document level. Therefore, this study introduces a novel strategy for sentiment analysis in Canadian maritime case law, combining sentiment case law approaches with state-of-the-art deep learning techniques. The overarching goal is to systematically unearth hidden biases within case law and investigate their impact on legal outcomes. Employing Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) based models, this research achieves a remarkable accuracy of 98.05% for categorizing instances. In contrast, Conventional Machine Learning techniques such as Support Vector Machines (SVM) yield an accuracy rate of 52.57%, Naive Bayes at 57.44 %, and Logistic Regression at 61.86%. The superior accuracy of the CNN and STM model combination underscores its usefulness in legal sentiment analysis, offering promising future applications in diverse fields like legal analytics and policy design. These findings mark a significant choice for AI-powered legal tools, presenting more sophisticated and sentiment-aware options for the legal profession.

Partial differential equations (PDEs) are used to describe a wide range of phenomena, such as heat transfer, fluid dynamics, and quantum mechanics. By solving PDEs, we can ob- tain insights into the behavior of the system and make predictions about its future evolution. Conventional numerical methods for obtaining the approximate solutions of PDEs may re- quire extensive computational resources and time, especially for complex PDEs and large- scale problems. The recently developed physics-informed neural network (PINN) has shown promise in a variety of scientific and engineering fields by incorporating physical laws into the loss functions of the neural network (NN). In addition, the training of PINN does not require ground truth data, but it has poor generalization ability to unseen domains. On the other hand, a convolutional neural network (CNN) has fast inference and better generalization ability, but it requires a large amount of training data. Taking the advantages of PINN and CNN by using Legendre multiwavelets (LMWs) as basis functions, we introduce a new method to approach the PDEs in this paper, namely Physics-Informed Legendre Multiwavelets CNN (PiLMWs-CNN), in order to continuously approximate a grid-based state representation that can be handled by a CNN. PiLMWs-CNN enable us to train our models using only physics-informed loss functions without any pre- computed training data, simultaneously providing fast and continuous solutions that gener- alize to previously unknown domains. In particular, the LMWs can simultaneously possess compact support, orthogonality, symmetry, high smoothness, and high approximation order. Compared to orthonormal polynomial (OP) bases, the approximation accuracy can be greatly increased and computation costs can be significantly reduced by using LMWs. We applied PiLMWs-CNN to approximate the damped wave equation, incompressible Navier-Stokes (N- S) equation, and two-dimensional heat conduction equation. The experimental results show that this method provides more accurate, efficient, and fast convergence with better stability when approximating the solution of PDEs.

Accurate fuel mapping plays a crucial role in fire detection and management strategies. This paper presents a method for discriminating between wildfire fuel types by exploiting together remote sensing data and Convolutional Neural Networks (CNN). Specially, a CNN-based classification approach that leverages Sentinel-2 imagery is exploited to accurately classify fuel types into seven preliminary main classes (conifers, broadleaf, shrubs, grass, bare soil, urban areas, and water bodies) with an high accuracy of 0.99$\%$. To further refine the fuel mapping results, subclasses were generated from the seven principals by using biomass and bioclimatic maps. These additional maps provide complementary information about vegetation density and climatic conditions, respectively. By incorporating this information, we align our fuel type classification with the widely used Scott/Burgan fuel classification system. This refinement step allows for a more detailed and comprehensive assessment of fuel types, enhancing the accuracy and effectiveness of fire management efforts, which can be utilized by fire management agencies, policymakers, and researchers for improved fire behavior prediction and mitigation practices. The proposed approach presents a valuable tool for enhancing fire management, contributing to more effective wildfire prevention and mitigation efforts.

Deep convolutional neural networks (CNN) are favored methods widely used in medical image processing due to their assured shown performance. Recently, the emergence of new lung diseases and the possibility of early detection of their symptoms has attracted many researchers to classify diseases by training deep CNNs on lung CT images. The trained networks are expected to distinguish between lung indications in different diseases, especially at the early stages of them. With the hope of achieving this purpose, we proposed an efficient deep CNN called AFEX-Net with adaptive feature extraction layers that successfully extract distinguishing features and classify chest CT images. The efficiency of the proposed network has two aspects: it is a lightweight network with low number of parameters and fast training and it has adaptive pooling layers and adaptive activation functions to increase its level of compatibility to the input data. The proposed network has been evaluated on a dataset with more than 10K chest CT slices, while an efficient pre-processing method is developed to remove any bias from the images. Additionally, we evaluated the performance of the proposed model on the public COVID-CTset dataset to prove the generalisability of our model. The obtained results confirm the competence of the proposed network in confronting medical images, where prompt and accurate learning is required.

We present a Machian model of Quantum Cosmology with full dark matter and light speed expansion and rotation. During galaxy formation and evolution, fraction of dark matter transforms to visual matter with a relation of the form, m_vis = constant * (m_dark)^2/3. Using this relation and replacing MOND’s ‘critical acceleration’ with “current cosmic maximum angular acceleration”, galactic flat rotation speed range of (50 to 500) km/sec can be fitted well. Estimated flat rotation speeds of DD168, Milky Way and UGC12591 are 49.96 km/sec, 199.66 km/sec and 521.75 km/sec respectively. Based on these striking coincidences, it is possible to say that, MOND’s approach is implicitly connected with cosmological estimation of 95% invisible matter. Considering galactic total matter and current cosmic maximum angular acceleration, galactic working radii, angular velocity and visual matter density can be estimated. Even though, this model is free from ‘big bang’, ‘inflation’, ‘dark energy’, ‘flatness’ and ‘red shift’ issues, at 2.722 K, estimated present Hubble parameter is 66.24 km/sec/Mpc, cosmic radius is 146.3 times the Hubble radius, angular velocity is 146.3 times lower than the Hubble parameter and cosmic age is 146.3 times the Hubble age. With future observations and advanced telescopes, it may be possible to see far distant galaxies and very old stars far beyond the current observable cosmic radius.

Conceiving vacuum energy as gravitational particles subject to Heisenberg’s energy-time uncertainty, modelled as dipoles in a fluidal space at thermodynamic equilibrium, and interpreting the Bekenstein-Hawking entropy as the effective amount of spins of those dipoles enclosed within the event horizon of the universe, allows the calculation of Milgrom’s acceleration constant. The result is a quantum mechanical interpretation of gravity, and dark matter in particular.

Evaluation of the effects of a high-frequency acceleration (HFA) device on patient pain response to orthodontic forces. A multi-centered trial investigating pain sensitivity to orthodontic forces on 75 subjects at 4 study centers. Subjects underwent clear aligner treatment, with or without adjunctive HFA and documented their pain intensity using the validated NRS10 numeric rating scale. In-Office and At-Home ratings were measured separately for each subject for immediate and extended effect evaluations. Use of HFA devices in conjunction with clear aligner orthodontic treatment demonstrated significant reduction in subjects’ recorded pain ratings vs controls within 5 minutes of aligner exchange, (p = 0.006) and significant reduction in recorded pain ratings vs controls over a 7-day period following aligner exchange (p = 0.018). A 99.6% daily compliance rate with at home use of the HFA device was recorded for all subjects in the study. HFA significantly reduces pain attributed to orthodontic force. HFA delivers clinically significant immediate pain relief, and clinically significant extended pain relief over the 7 days following adjustment.

This research proposes a near-infrared hyperspectral imaging/two-branch convolutional neural network (NIR-HSI/2B-CNN) algorithmic scheme to detect mango anthracnose of the species Colletotrichum asianum at the early stages of disease development. In the algorithmic model development, root mean square propagation was used as the solver to train the neural network, given 150 epochs. In addition, spectral raw data was preprocessed to transform it into an understandable and efficient format. The optimal classification model was the 2B-CNN model with 1st-derivative preprocessing, achieving an accuracy of 0.94 for the calibration set and 0.71 for the prediction set. The proposed NIR-HSI/2B-CNN scheme could detect anthracnose mangoes since the the first day of inoculation of the spore suspension (i.e., day 0) through to day 3, achieving a moderate classification accuracy. Meanwhile, the accuracy of conventional convolutional neural networks (CNN) were within a range of 0.66-0.67 for the calibration set and 0.55-0.57 for the prediction set. The results indicated that incorporating spatial features in the 2B-CNN modeling enhanced the prediction performance of the algorithm. The proposed NIR-HSI/2B-CNN algorithmic scheme needs refinements to be able to reliably sort mango fruits into those suitable for premium fresh consumption and export without anthracnose and those for domestic consumption or processing. The novelty of this research lies in the use of NIR-HSI and 2B-CNN algorithm to detect plant pathogens at the early stages of disease development. In addition, the new method of natural simulation to deposit the fungal spores onto the mango surface by spraying spore suspension onto the mango surface where the conidia penetrated unaided into the underneath of the mango peel is proposed.

Deep Learning (DL) has made significant advances in computer vision with the advent of Vision Transformers (ViT). Unlike Convolutional Neural Networks (CNNs), ViTs use self-attention to extract both local and global features from image data, and then use residual connections to feed these features directly into a fully networked multilayer perceptron head. In hospitals, hematologists prepare peripheral blood smears (PBSs) and read them under a medical microscope to detect abnormalities in blood counts such as leukemia. However, this task is time-consuming and prone to human error. This study investigates the transfer learning process of Google ViT and ImageNet CNNs to automate the reading of PBSs. The study used two online PBS datasets, PBC and BCCD, and transferred them into balanced datasets to investigate the influence of data amount and noise immunity on both neural networks. The PBC results show that Google ViT is an excellent DL neural solution for data scarcity. The BCCD results show that Google ViT is superior to ImageNet CNNs in dealing with unclean, noisy image data because it is able to extract both global and local features and use residual connections, despite the additional time and computational overhead.

Accurate detection of players and teams in ice hockey games is crucial to the tracking of individual players on court and team tactical decisions, which is therefore becoming an important task for coaches and other analysts. However, hockey is a fluid sport due to complex dynamics and frequent substitutions by both teams, resulting in various body positions of players. Few traditional player detection models from other team sports take these characteristics into account, especially for the detection of teams without prior annotations. Here, we design a two-phase cascaded Convolutional Neural Network (CNN) model coupling between individual players position information and team uniform colors to hierarchically detect players in ice hockey games. Our model filters most of disturbing information, such as audience and sideline advertising bars, in Phase I, and refines the detection of targeted players in Phase II, which is efficient and accurate with a precision rate of 91.30% and a recall rate of 95.60% for individual players detection, and an average accuracy of 93.05% in team classification from a self-built dataset of collected images in the 2018 Winter Olympics. Meanwhile, we also present results based on the images and real-time detection from broadcasting videos of 2019-20 NHL regular games covering all 31 teams to show the robustness of our model.

The heat and matter transfer during glucose catabolism in living systems and their relation with entropy production are a challenging subject of the classical thermodynamics applied to biology. In this respect, an analogy between mechanics and thermodynamics has been performed via the definition of the entropy density acceleration expressed by the time derivative of the rate of entropy density and related to heat and matter transfer in minimum living systems. Cells are regarded as open thermodynamic systems that exchange heat and matter resulting from irreversible processes with the intercellular environment. Prigogine’s minimum energy dissipation principle is reformulated using the notion of entropy density acceleration applied to glucose catabolism. It is shown that, for out-of-equilibrium states, the calculated entropy density acceleration is finite and negative and approaches as a function of time a zero value at global thermodynamic equilibrium for heat and matter transfer independently of the cell type and the metabolic pathway. These results could be important for a deeper understanding of entropy generation and its correlation with heat transfer in cell biology with special regard to glucose catabolism representing the prototype of irreversible reactions and a crucial metabolic pathway in stem cells and cancer stem cells.

Detecting people and technical objects in various situations, such as natural disasters and warfare, is critical to search and rescue operations and the safety of the civilian. A fast and accurate detection for people and equipment can significantly increase the effectiveness of search and rescue missions and provide timely assistance to people. Computer vision and deep learning technologies play a key role in detecting the required objects due to their ability to analyse big volumes of visual data in real time. The performance of the neural networks such as YOLOv4-v8, Faster R-CNN, SSD, and EfficientDet has been analysed using COCO2017, SARD, SeaDronesSee, and VisDrone2019 datasets. Main criteria for comparison were mAP, Precision, Recall, F1-Score, and the ability of the neural network to work in real time. The most important metrics for evaluating the efficiency and performance of models for a given task are accuracy (mAP), F1-Score, and processing speed (FPS). These metrics allow us to evaluate both the accuracy of object recognition and the ability to use the models in real-world environments where high processing speed is important. Although different neural networks perform better on certain types of metrics, YOLO outperforms them on all metrics, showing the best results of mAP-0.88, F1 - 0.88, and FPS - 48, so the focus was on these models.

Recently, a significant number of tools for detecting dangerous objects have been developed. Unfortunately, the performance offered by them is overestimated due to the poor quality of the datasets used (insufficiently numerous, contain items not strictly related to dangerous objects, insufficient range of presentation conditions). To fill in a gap in this area we have built an extensive dataset dedicated to detecting the objects most often used in various acts of breaching public security (baseball bat, gun, knife, machete, rifle). This collection contains images presenting the detected objects with different quality and under different environmental conditions. We believe that the results obtained from it are more reliable and give a better idea of the detection accuracy that can be achieved under real conditions. We used the Faster R-CNN with different backbone networks in the study. The best results were obtained for the ResNet152 backbone. The mAP value was 85%, while the AP level ranged from 80% to 91%, depending on the item detected. An average real-time detection speed was 11-13 FPS. Both the accuracy and speed of the Faster R-CNN model allow it to be recommended for use in public security monitoring systems aimed at detecting potentially dangerous objects.

Some recent studies show that filters in convolutional neural networks (CNNs) have low color selectivity in datasets of natural scenes such as Imagenet. CNNs, bio-inspired in the visual cortex, are characterized by their hierarchical learning structure, which appears to gradually transform the representation space. Inspired by the direct connection between the LGN and V4, which allows V4 to handle low-level information closer to the trichromatic input in addition to processed information that comes from V2/V3, we propose the addition of a long skip connection (LSC) between the first and last blocks of the feature extraction stage to allow deeper parts of the network to receive information from shallower layers. This type of connection improves classification accuracy by combining simple-visual and complex-abstract features to create more color-selective ones. We have applied this strategy to classic CNN architectures and quantitatively and qualitatively analyzed the improvement in accuracy while focusing on color selectivity. The results show that, in general, skip connections improve accuracy, but LSC improves it even more and enhances the color selectivity of the original CNN architectures. As a side result, we propose a new color representation procedure for organizing and filtering feature maps, making their visualization more manageable.

We introduce a novel hybrid Transformer-CNN architecture for robotic grasp detection, designed to enhance the accuracy of grasping unknown objects. Our proposed architecture has two key designs. Firstly, we develop a hierarchical transformer as the encoder, incorporating the external attention to effectively capture the correlation features across the data. Secondly, the decoder is constructed with cross-layer connections to efficiently fuse multi-scale features. Channel attention is introduced in the decoder to model the correlation between channels and to adaptively recalibrate the channel correlation feature response, thereby increasing the weight of the effective channels. Our method is evaluated on the Cornell and Jacquard public datasets, achieving an image-wise detection accuracy of 98.3% and 95.8% on each dataset, respectively. Additionally, we achieve object-wise detection accuracy of 96.9% and 92.4% on the same datasets. A physical experiment is also performed using the Elite 6Dof robot, with a grasping accuracy rate of 93.3%, demonstrating the proposed method's ability to grasp unknown objects in real-world scenarios. The results of this study show that our proposed method outperforms other state-of-the-art methods.

This paper introduces a fusion convolutional architecture for efficient learning of spatio-temporal features in video action recognition. Unlike 2D CNNs, 3D CNNs can be applied directly on consecutive frames to extract spatio-temporal features. The aim of this work is to fuse the convolution layers from 2D and 3D CNNs to allow temporal encoding with fewer parameters than 3D CNNs. We adopt transfer learning from pre-trained 2D CNNs for spatial extraction, followed by temporal encoding, before connecting to 3D convolution layers at the top of the architecture. We construct our fusion architecture, semi-CNN, based on three popular models: VGG-16, ResNets and DenseNets, and compare the performance with their corresponding 3D models. Our empirical results evaluated on the action recognition dataset UCF-101 demonstrate that our fusion of 1D, 2D and 3D convolutions outperforms its 3D model of the same depth, with fewer parameters and reduces overfitting. Our semi-CNN architecture achieved an average of 16 – 30% boost in the top-1 accuracy when evaluated on an input video of 16 frames.

Sensitivity to weather conditions, and specially to clouds, is a severe limiting factor to the use of optical remote sensing for Earth monitoring applications. A possible alternative, is to resort to weather-insensitive synthetic aperture radar (SAR) images. However, in many real-world applications, critical decisions are made based on some informative spectral features, such as water, vegetation or soil indices, which cannot be extracted from SAR images. In the absence of optical sources, these data must be estimated. The current practice is to perform linear interpolation between data available at temporally close time instants. In this work, we propose to estimate missing spectral features through data fusion and deep-learning. Several sources of information are taken into account - optical sequences, SAR sequences, DEM - so as to exploit both temporal and cross-sensor dependencies. Based on these data, and a tiny cloud-free fraction of the target image, a compact convolutional neural network (CNN) is trained to perform the desired estimation. To validate the proposed approach, we focus on the estimation of the normalized difference vegetation index (NDVI), using coupled Sentinel-1 and Sentinel-2 time-series acquired over an agricultural region of Burkina Faso from May to November 2016. Several fusion schemes are considered, causal and non-causal, single-sensor or joint-sensor, corresponding to different operating conditions. Experimental results are very promising, showing a significant gain over baselines methods according to all performance indicators.

Acceleration parameters in sit-to-stand (STS) movements are useful for measuring lower-limb function in older adults. The purpose of this study was to examine the relationship between acceleration in STS movements and physical function and the test-retest reliability of acceleration parameters in older adults. We performed cross-sectional analyses on 244 older adults including 107 men (mean age:77.4±4.7) and 137 women (mean age: 75.6±5.3). Four acceleration parameters were measured in STS movements: maximum acceleration (MA), maximum velocity (MV), maximum power (MP), and stand-up time (ST). Good intraclass correlation coefficients (ICC>0.70) were observed for all parameters. For the acceleration parameters, MA, MV, and MP were relatively strongly associated with the 5-time STS test (men: r=-0.36~-0.47; women: r=-0.37~-0.45) and the Timed up and Go test (men: r=-0.39~0.47, women: r=-0.43~-0.51): MP was also strongly associated with grip strength (men: r=0.48, women: r=0.43). All acceleration parameters were poorer in participants reporting mobility limitations than in those reporting no mobility limitations. These findings support the usefulness of sensor-based STS measurement. The system is expected to be useful in various settings where care prevention is addressed.

A Mw7.8 earthquake struck Turkey on February 6, 2023, causing severe casualties and economic losses. This paper investigates the characteristics of strong ground motion and seismogenic fault of the earthquake. We collected and processed the strong ground motion records of 379 stations using Matlab, SeismoSignal and Surfer software, and obtained the peak ground acceleration (PGA) contour map. We analyzed the near-fault effect, the fault locking segment effect and the trampoline effect of the earthquake based on the spatial distribution of PGA, the fault geometry and slip distribution. We found that the earthquake generated a very strong ground motion concentration effect in the near-fault area, with the maximum PGA exceeding 2000cm/s2. However, the presence of fault locking segments influenced the spatial distribution of ground motion, resulting in four significant PGA high-value concentration areas at a local dislocation, a turning point and the end of the East Anatolian Fault. We also revealed for the first time the typical manifestation of the trampoline effect in this earthquake, which was characterized by a large vertical acceleration with the positive direction significantly larger than the negative direction. This paper provides an important reference for understanding the seismogenic mechanism, damage mode, characteristics and strong earthquake law of Turkey earthquake.

With the urgent need of precise positioning faced by internet of things (IoT) applications, the universality and cost of indoor positioning devices become key factors. Since the 5G network has been widely deployed, new opportunity is brought by tightly fusing the traditional low-cost sensors, i.e., the magnetometer. In this study, using 5G channel state information (CSI) and geomagnetic data, a multi-input convolutional neural network (CNN) localization system was proposed. First, to generate a data tensor that is easy for CNN processing, the raw data was reconstructed individually. Then, to comprehensively incorporate the features of 5G CSI and geomagnetic strength data, the ReLU function was chosen as the activation function of the convolutional layer. After that, a multi-input CNN was trained using the incorporated geomagnetic strength and CSI amplitude in the off-line side, and the trained CNN was recorded as a location fingerprint, which can be used for the user position prediction. Finally, in the online side and using a multi-input CNN, the 2-D coordinates are estimated and tested indoors in a typical conference room scenario. The results showed that longer sampling time of fingerprint data result in better uniqueness of the reference point, while the data collection time of locating points does not need to be long. Taking the positioning efficiency and accuracy into consideration, a sampling time of 3s at the reference point and 0.2s at the locating point are recommended. The positioning accuracy using the proposed method was 1.41 m, with an improvement of 22.9% compared with the 5G positioning, and an improvement of 18.0% compared with the 5G and geomagnetic fusing positioning using single CNN.

Facial expression recognition is crucial for understanding human emotions and nonverbal communication. With the growing prevalence of facial recognition technology and its various applications, accurate and efficient facial expression recognition has become a significant research area. However, most previous methods have focused on designing unique deep-learning architectures while overlooking the loss function. This study presents a new loss function that allows simultaneous consideration of inter- and intra-class variations to be applied to CNN architecture for facial expression recognition. More concretely, this loss function reduces the intra-class variations by minimizing the distances between the deep features and their corresponding class centers. It also increases the inter-class variations by maximizing the distances between deep features and their non-corresponding class centers, and the distances between different class centers. Numerical results from several benchmark facial expression databases, such as Cohn-Kanade Plus, Oulu-Casia, MMI, and FER2013, are provided to prove the capability of the proposed loss function compared with existing ones.

It is important for food recognition to separate each ingredient within a food image at the pixel level. In this paper, we propose a new approach to segment ingredients by utilizing a CNN-based Single-Ingredient Classification Model. In detail, we firstly introduce a standardized biological-based hierarchical ingredient structure and construct a single-ingredient image dataset based on this structure. Then, we build a single-ingredient classification model based on a novel convolutional neural network (CNN) architecture that utilizes an attention mechanism. Afterwards, we propose a new framework for segmentation using the above single-ingredient classification model as the backbone. In this framework, two methods are involved in segmenting ingredients in the food images. We introduce five evaluation metrics (IoU, Dice, Purity, Entirety, Loss of GTs) to assess the performance of ingredient segmentation in terms of ingredient classification. Extensive experiments demonstrate the effectiveness of the proposed method, achieving an maximal mIoU of 0.65, mDice of 0.77, mPurity of 0.83, mEntirety of 0.80, and mLoGTs of 0.06 on the FoodSeg103 dataset. The results confirm that our CNN-based architecture achieves higher segmentation performance compared to ResNet18 and EfficientNet-B0 when used as the backbone for ingredient segmentation. We believe that our ingredient segmentation approach lays the foundation for subsequent ingredient recognition.

The physics of Hubble's law, motivated by the anomalous acceleration of Pioneer 10/11, suggests that the Hubble constant corresponds to a gravitational field of the Universe. The Hubble constant inferred from Pioneer 10/11 data is $\sim 69$ km/s/Mpc. The anomalous acceleration of Pioneer 10/11 can be interpreted as an observational evidence for Mach's principle. The redshift is an apparent value. Its intrinsic value has a limit resulting to $v

The use of sensors in monitoring of lateral accelerations in delivery van transport focuses on measuring lateral accelerations on routes with roundabouts and curves to increase road safety. Using microelectromechanical system (MEMS) sensors, it measures the lateral accelerations acting on the vehicle and the load being transported during the test drives to study vehicle dynamics for cargo securing. Using an accelerometer and position tracking, accelerations can be detected when traversing curves and roundabouts at selected locations on the vehicle. The level of accelerations on the roof of the vehicle was found to be like those occurring on a lashed load with limited movements, where regression analysis can be used to determine the relationship between lateral accelerations at different sensor positions. If we compare the mean values of the lateral accelerations of the individual events between the sensors, the sensor on the side of the vehicle body at the height of the sensor on the load had approximately 5% lower mean values than the sensor on the roof. The sensor on the load measured approximately 5% higher mean values than the sensor on the roof. Hence, the mean lateral accelerations of the individual events for the sensor on the load are 10% higher than for the sensor at the same height on the vehicle body. From the above testing, we can say that the values of the mean lateral accelerations of the van from the sensor on the roof of the vehicle, are closer to the values of the accelerations of the sensor on the load than to the values of the sensor on the body of the vehicle at the same height.

Nondipole terms in the atom-laser interaction arising due to the presence of a magnetic component in an electromagnetic wave and its inhomogeneity lead to nonseparability of the center-of-mass (CM) and electron variables in the neutral atom and, as a consequence, to its acceleration. We investigate this effect as well as the accompanying excitation and ionization processes for the hydrogen atom in strong (1012 - 2×1014 W/cm2) linearly polarized short-wavelength (5eV ≲ℏω≲ 27 eV) electromagnetic pulses of about 8 fs duration. The study has been carried out within the framework of a hybrid quantum-quasiclassical approach in which the coupled time-dependent Schrödinger equation for an electron and the classical Hamilton equations for the CM of an atom are simultaneously integrated. Optimal conditions with respect to the frequency and intensity of the electromagnetic wave for the acceleration of atoms without their noticeable ionization are found in the analyzed region.

The crystallization conditions from the solution play an important role in determining the mor-phology, phase composition, and photovoltaic properties of perovskite films. Post-processing of the obtained films can have a crucial role in increasing the grain size of perovskite and enhancing its crystallinity. It has been shown that the formation of crystal nuclei can be utilized to accelerate crystallization. In this case, crystallization occurs through the growth of seed crystals created in the solution, enabling the formation of relatively large crystals. For the deposition of CH3NH3PbI3 hybrid halide perovskite films from a solution of the perovskite in dimethylformamide the spin coating echnique was employed. Pre-crystallization was achieved by heating films at a temperature of 100 or 110 °C. The dissolution process involved adding a drop of dimethylformamide onto the substrate surface and allowing it to partially dis-solve the perovskite for 30-40 seconds. Subsequently, residual solvent was removed through spin coating. The recrystallization process was carried out by heating films at the temperature range of 80 to 130 °C. The morphological analysis of the perovskite film surfaces after recrystallization at tem-peratures ranging from 80 to 130 °C was performed. The infrared transmission spectra of the obtained perovskite films were investigated, and their light absorption characteristics were studied through transmission spectra. The perovskite structure in the obtained films was con-firmed by the peaks observed in the X-ray diffraction patterns.

In this paper, we implement a new method binary Particle Swarm Optimization (PSO) for solving the kidney exchange problem, which will improve the future decisions of kidney exchange programs. Because using a kidney exchange, we can help incompatible patient-donor couples to swap donors to receive a compatible kidney. Kidney paired donation programs provide an innovative approach for increasing the number of available kidneys. Further, we implementing binary particle swarm optimization in parallel with MATLAB with one, two, three and four threads and from the computations point of view, the authors compare the performance to reduce the running time for kidney exchange to match patients as fast as possible to help clinicians. Moreover, implementing binary particle swarm optimization in solving the kidney exchange problem is an effective method. The obtained results indicate that binary PSO outperforms other stochastic-based methods such as genetic algorithm, ant lion optimization, and efficient the number of resulting exchanges.

Evaluation of the effect of a high frequency acceleration device on clear aligner exchange intervals and treatment time required to achieve prescribed tooth movements. Sixteen subjects with similar Class I malocclusions, 5mm or less crowding, and treated with Invisalign were divided into two groups. Group 1 (experimental; N=8) underwent aligner treatment in conjunction with daily use of the high frequency acceleration device and exchanged aligners every 5 days. Group 2 (controls; N=8) underwent aligner treatment without use of the device and exchanged aligners every 14 days according to the manufacturer's recommended interval. All subjects were treated by one investigator, and results were evaluated by both; total number of aligners used, and number of refinements required, prior to final Vivera retention scan. A significant decrease in both treatment time, and number of aligners required to complete treatment was observed by experimental subjects as compared to controls. In addition, no refinements were required by experimental subjects, whereas 6 of 8 of the control subjects required 1 or more refinements. Use of the high frequency acceleration device in conjunction with Invisalign treatment resulted in more predictable tooth movement and a significant decrease in the length of treatment.

Landslide is one of the most common geological disasters in China, which is characterized by suddenness and uncertainty, and it is difficult to realize accurate identification, early warning and forecasting of landslide disaster by conventional means. With the development of high-resolution remote sensing satellites and InSAR surface deformation monitoring technology, the traditional means of landslide monitoring data sources are limited, and there is a lack of effective methods to excavate the characteristics of the spatial distribution of landslide hazards and their triggering factors and other issues. In this paper, the area extending 10 km outside the VII isobar of the Gengma earthquake is taken as the study area, and 13 evaluation factors are screened out by integrating the factors of InSAR surface deformation, topography and geological environment, and the Bayesian Optimized Convolutional Neural Network (BO-CNN) is used for the evaluation of landslide susceptibility, and the BO-RF and PSO-SVM models are selected for the comparative analysis. The model accuracy evaluation was carried out by three indexes: ROC curve, AUC value and FR value, in which the ROC curves of PSO-SVM, BO-RF and BO-CNN were all close to the upper-left corner of the corner, and the AUC values were 0.9388, 0.9529, and 0.9535, respectively, and the FR value of landslide in the high susceptibility area of BO-CNN was as high as 14.9, and was higher than that of PSO-SVM and BO-RF, respectively. SVM and BO-RF model is 4.55 and 3.69 higher, the experimental results show that the BO-CNN model used in this paper has a better effect in landslide susceptibility evaluation, and the research results of the local government's disaster prevention and mitigation measures are of great significance.

it has been just over three years since SARS-CoV-2 viral infection was first known as causing severe acute and highly transmissible respiratory syndrome coronavirus. During one of the worst global pandemics of the last century, caused by the COVID-19 disease, under the persistent immune pressure exerted by the newest vaccines, the SARS-CoV-2 virus is fastly mutating and becoming less severe, but more contagious. On the other hand, some studies [1,2,3] indicate that the elderly COVID-19 patients or those who have reported a virulent episode of this infection, experience some long-lasting neurological complications, including encephalopathy, encephalitis, cerebrovascular disease, Parkinson’s disease (PD), Alzheimer’s disease (AD) and prion disease (PrD). So, there are several recent reports of patients [4,5] developing PrD following the SARS-CoV-2 infection, but, to the best of our knowledge, this is the first case reported in Italy on an association of the PRNP E200K somatic mutation with gCJD. The rising global prevalence neurodegenerative complications following COVID-19 disease adds urgency to the study of this potential relationship.

Human activities and natural events alter the earth’s surface and pose a constant threat to the environment. Thus, accurately monitoring and predicting these changes in a timely manner is crucial to provide solutions and mitigate the environmental consequences beforehand. This research introduces a novel framework, called HCD-Net, for change detection using bi-temporal hyperspectral images. The framework is based on double-stream deep feature extraction and an attention mechanism. The first stream focuses on deep feature extraction through 3D convolution layers and 3D Squeeze-and-Excitation (SE) blocks. The second stream focuses on deep feature extraction through 2D convolution and 2D SE blocks. The extracted deep features are then concatenated and fed to dense layers for decision-making. The efficiency of HCD-Net is compared to the state-of-the-art change detection methods. Additionally, the bi-temporal Airborne Visible InfraRed Imaging Spectrometer (AVIRIS) hyperspectral dataset was used for further evaluation of the change detection results. The results show that HCD-Net has a higher accuracy and the lowest false alarm rate, where the overall classification accuracy is above 96%, and the kappa coefficient is over 0.9.

The integration of nanoparticles in biomedical applications requires precise analysis of their photochemical properties. However, computational simulations of these properties are computationally intensive, hindering rapid advancement. This study explores the implementation of Graphics Processing Unit (GPU) acceleration techniques to enhance computational efficiency in analyzing nanoparticle photochemistry within bioinformatics frameworks.By leveraging GPU parallel processing capabilities, we developed optimized algorithms for photochemical property simulations, achieving significant speedup over traditional Central Processing Unit (CPU)-based methods. Our GPU-accelerated framework demonstrated substantial performance gains in computing photochemical properties, such as absorption spectra and energy transfer rates.

Spiking neural networks (SNNs) promise to perform tasks currently done by classical artificial neural networks (ANNs) faster, in smaller footprints and using less energy. Neuromorphic processors are set out to revolutionize computing at a large scale, but the move to edge-computing applications calls for finely-tuned custom implementations to keep pushing towards more efficient systems. To that end, we have examined the architectural design space for executing spiking neuron models on FPGA platforms, focusing on achieving ultra-low area and power consumption. This work presents an efficient clock-driven spiking neuron architecture used for the implementation of both fully-connected cores and 2D convolutional cores, which rely on deep pipelines for synaptic processing and distributed memory for weight and neuron states. With them, we have developed an accelerator for an SNN version of the LeNet-5 network trained on the MNIST dataset. At around 5.5 slices/neuron and only 348 mW, it is able to use 33% less area and 4 times as less power per neuron as current state-of-the-art implementations while keeping low simulation step times.

This paper is intended at highlighting the main mechanical parameters controlling the behavior of the so-called "acceleration-sensitive" Non-Structural Components (NSCs). In the first part a short review of the current state of knowledge and the critical issues related to the prediction of the seismic response of NSCs is reported. Then, the paper presents the results of a numerical parametric analysis intended to capture the key features of the dynamic response of a two-degree-of-freedom (2DOF) system which is supposed to be representative of both the main structure and the "non-structural" component (NSC). Particularly, it allows to simulate the coupled behaviour of both main structure and NSC and evaluating their response. The main parameters controlling the dynamic response of NSCs emerge from this study, which could pave the way towards formulating more mechanically consistent relationships for evaluating the maximum accelerations induced by seismic shakings on NSCs.

Smartphones, smartwatches, fitness trackers, and ad-hoc wearable devices are being increasingly used to monitor human activities. Data acquired by the hosted sensors are usually processed by machine-learning-based algorithms to classify human activities. The success of those algorithms mostly depends on the availability of training (labeled) data that, if made publicly available, would allow researchers to make objective comparisons between techniques. Nowadays, publicly available data sets are few, often contain samples from subjects with too similar characteristics, and very often lack of specific information so that is not possible to select subsets of samples according to specific criteria. In this article, we present a new smartphone accelerometer dataset designed for activity recognition. The dataset includes 11,771 activities performed by 30 subjects of ages ranging from 18 to 60 years. Activities are divided in 17 fine grained classes grouped in two coarse grained classes: 9 types of activities of daily living (ADL) and 8 types of falls. The dataset has been stored to include all the information useful to select samples according to different criteria, such as the type of ADL performed, the age, the gender, and so on. Finally, the dataset has been benchmarked with two different classifiers and with different configurations. The best results are achieved with k-NN classifying ADLs only, considering personalization, and with both windows of 51 and 151 samples.

The advent of Sixth Generation (6G) wireless technologies introduces challenges and opportunities for Mobile Ad Hoc Networks (MANETs) and Vehicular Ad Hoc Networks (VANETs), necessitating a reevaluation of traditional routing protocols. This paper introduces the Multi-Metric Scoring Dynamic Source Routing (MMS-DSR), a novel enhancement of the Dynamic Source Routing (DSR) protocol, designed to meet the demands of 6G-enabled MANETs and the dynamic environments of VANETs. MMS-DSR integrates advanced technologies and methodologies to enhance routing performance in dynamic scenarios. Key among these is the use of a CNN-LSTM based beamforming algorithm, which optimizes beamforming vectors dynamically, exploiting spatial-temporal variations characteristic of 6G channels. This enables MMS-DSR to adapt beam directions in real-time based on evolving network conditions, improving link reliability and throughput. Furthermore, MMS-DSR incorporates a multi-metric scoring mechanism that evaluates routes based on multiple QoS parameters, including latency, bandwidth, and reliability, enhanced by the capabilities of Massive MIMO and the IEEE 802.11ax standard. This ensures route selection is context-aware and adaptive to changing dynamics, making it effective in urban settings where vehicular and mobile nodes coexist. Additionally, the protocol uses machine learning techniques to predict future route performance, enabling proactive adjustments in routing decisions. The integration of dynamic beamforming and machine learning allows MMS-DSR to effectively handle the high mobility and variability of 6G networks, offering a robust solution for future wireless communications, particularly in smart cities.

Satellite clock bias (SCB) prediction is a crucial technology for satellite navigation systems, holding significant importance for Global Navigation Satellite System. This paper proposes a deep learning model for SCB prediction based on the fusion of the Beluga Whale Optimization (BWO), Convolutional Neural Network (CNN), Bidirectional Gated Recurrent Unit (BiGRU), and an attention mechanism. The CNN is utilized to extract the spatiotemporal characteristic information from the clock bias sequence, while the BiGRU fully extracts relevant features through forward and backward propagation. The introduction of an attention mechanism aims to preserve essential features within the clock bias sequence to enhance feature extraction by both CNN and BiGRU networks. Additionally, the BWO is employed to optimize parameter selection in order to improve model accuracy. Experimental verification demonstrates that for the BeiDou Navigation Satellite System's (BDS) hydrogen-maser atomic clocks, the predicted clock bias for 6 hours, 3 days, and 15 days are 0.078 ns, 0.475 ns, and 2.130 ns respectively - representing improvements of 31%, 45%, and 66% over CNN-BiGRU-Attention; 6%, 51%, and 56% over CNN-BiGRU; and 32%, 35%, and 73% over BiGRU, respectively.

The prompt and accurate diagnosis of breast lesions, including the distinction between cancer, non-cancer, and suspicious cancer, plays a crucial role in the prognosis of breast cancer. In this paper, we introduce a novel method based on feature extraction and reduction for detection of breast cancer in mammography images. First, we extract features from multiple pre-trained convolutional neural network (CNN) models, and then concatenate them. The most informative features are selected based on their mutual information with the target variable. Subsequently, the selected features can be classified using a machine learning algorithm. We evaluate our approach using four different machine learning algorithms, and our results demonstrate that the neural network-based classifier yields an accuracy as high as 92% for the RSNA dataset which is a new dataset that provides two views and additional features such as age. We compare our proposed algorithm with state-of-the-art methods and demonstrate its superiority, particularly in terms of accuracy and sensitivity. For the MIAS dataset, we achieve an accuracy as high as 94.5%, and for the DDSM dataset, an accuracy of 96% is attained. These results highlight the effectiveness of our method in accurately diagnosing breast lesions and surpassing existing approaches.

Panoramic and periapical radiograph tools help dentists diagnose the most common dental diseases. Generally, dentists identify dental caries manually by inspecting X-ray images. However, due to their heavy workload, or poor image quality, dentists may sometimes overlook some unnoticeable dental caries, which may ultimately hinder the patient treatment. The purpose of this study was to develop an algorithm that classifies the teeth X-Ray images into three categories of “Normal”, “Caries”, and “Filled”. Our study used a dataset of 116 patients and 3712 single teeth images for training, validation and testing. Images were pre-processed using a sharpening filter and an intensity color map. We used a pre-trained transfer learning model, the NASNetMobile, which served as the feature extractor and the Convolutional Neural Network (CNN) model served as the classifier. The training dataset had a “Recall” of 0.92, 0.90 and 0.91 for “Normal”, “Caries” and “Filled” respectively and the test dataset had a “Recall” of 0.86, 0.81 and 0.85 for “Normal”, “Caries” and “Filled” respectively. The classification of the teeth X-Rays was successful and can be valuable for dentist as the artificial intelligence algorithm can serve as a decision support tool to aid dentists when they need to diagnose dental treatment.

Advancements in deep learning and availability of medical imaging data have led to use of CNN based architectures in disease diagnostic assisted systems. In spite of the abundant use of reverse transcription-polymerase chain reaction (RT-PCR) based tests in COVID-19 diagnosis, CT images offer an applicable supplement with its high sensitivity rates. Here, we study classification of COVID-19 pneumonia (CP) and non-COVID-19 pneumonia (NCP) in chest CT scans using efficient deep learning methods to be readily implemented by any hospital. We report our deep network framework design that encompasses Convolutional Neural Networks (CNNs) and bidirectional Long Short Term Memory (biLSTM) architectures. Our study achieved high specificity (CP: 98.3%, NCP: 96.2% Healthy: 89.3%) and high sensitivity (CP: 84.0%, NCP: 93.9% Healthy: 94.9%) in classifying COVID-19 pneumonia, non-COVID-19 pneumonia and healthy patients. Next, we provide visual explanations for the CNN predictions with gradient-weighted class activation mapping (Grad-CAM). The results provided a model explainability by showing that Ground Glass Opacities (GGO), indicators of COVID-19 pneumonia disease, were captured by our CNN network. Finally, we have implemented our approach in three hospitals proving its compatibility and efficiency.

Early disease detection is crucial for maximizingcrop yields and minimizing financial losses in agriculture. Inorder to categorise leaf diseases in pepper, tomato, and potato plants, this study suggests a revolutionary CNN design. The algorithm collects disease-specific features from input photos by using data augmentation techniques to artificially enlarge the training dataset. ReLU-activated convolutional layers gradually capture features at different levels, from low to high, while max pooling layers minimise spatial dimensionality. After thecharacteristics are extracted, a fully-connected layers with a softmax activation function classify them into illness groups. A well chosen dataset containing labelled healthy and sick leaves is used to train and validate the model. Generalizability to untesteddata is ensured via performance evaluation on an independent testing set. The field of deep learning for agricultural applications benefits from this research. This strategy has the ability to rev- olutionise agricultural practices by enabling automated disease identification. This might result in early interventions, better crop health, and eventually sustainable agricultural production.

Pulmonary sarcoidosis is a multisystem granulomatous interstitial lung disease (ILD) with a variable presentation and prognosis. Early accurate detection of pulmonary sarcoidosis may prevent progression to pulmonary fibrosis, a serious and potentially life-threatening form of the disease. However, the lack of a gold-standard diagnostic test and specific radiographic findings pose challenges in diagnosing pulmonary sarcoidosis. Chest computed tomography (CT) imaging is commonly used but requires expert, chest-trained radiologists to differentiate pulmonary sarcoidosis from lung malignancies, infections, and other ILDs. In this work, we developed a multichannel, CT and radiomics-guided ensemble network (RadCT-CNNViT) with visual explainability for pulmonary sarcoidosis vs. lung cancer (LCa) classification using chest CT images. We leverage CT and hand-crafted radiomics features as input channels, and a 3D convolutional neural network (CNN) and vision transformer (ViT) ensemble network for feature extraction and fusion before a classification head. The 3D CNN sub-network captures localized spatial information of lesions, while the ViT sub-network captures long-range, global dependencies between features. Through multichannel input and feature fusion, our model achieved the highest performance with accuracy, sensitivity, specificity, and combined AUC of 0.93±0.04, 0.94±0.04, 0.93±0.08 and 0.97 respectively in a 5-fold cross-validation study with pulmonary sarcoidosis (n=126) and LCa (n=93) cases. The model offers promising potential for improving the diagnosis of pulmonary sarcoidosis. A detailed ablation study showing the impact of CNN+ViT compared to CNN or ViT alone, and CT+radiomics input, compared to CT or radiomics alone, is also presented in this work.

Reducing the costs of repairing concrete structures damaged due to the appearance of cracks and reducing the number of people involved in the process of their repair is the subject of a multitude of experimental studies. Special emphasis should be placed on research involving industrial by-products, the disposal of which has a negative environmental impact, as is the case in the research presented in this paper. The basic idea was to prepare a mortar with added granulated blast furnace slag from Smederevo Steel Mill and then treat artificialy produced cracks with a Sporosarcina pasteurii DSM 33 suspension under the conditions of sterile demineralized and water from Danube river in order to simulate natural conditions. The results show a bio-stimulated healing efficiency of 32.02% in sterile demineralized water and 42.74% in Danube water already after 14 days. The SEM images clearly show calcium carbonate crystals as the main compound that has started to fill the crack, and the crystals are much more developed under the Danube water conditions. As a special type of research, microscopic images of cracks were classified into those with and without the presence of bacterial culture. By applying convolutional neural networks (ResNet 50), the classification success rate was 91.55%.

Industrial process control systems commonly exhibit features of time-varying, strong coupling, and strong nonlinearity. Obtaining accurate mathematical models of these nonlinear systems and achieving satisfactory control performance is still a challenging task. In this paper, data-driven modeling techniques and deep learning methods are used to accurately capture a category of smooth nonlinear system’s spatiotemporal features. The operating point of these systems may change over time, and their nonlinear characteristics can be locally linearized. We established the LSTM-CNN-ARX model by utilizing a fusion of long short-term memory (LSTM) network and convolutional neural network (CNN) to fit the coefficients of the state-dependent exogenous variable autoregressive (SD-ARX) model. Compared to other models, the hybrid LSTM-CNN-ARX model is more effective in capturing the nonlinear system’s spatiotemporal characteristics due to its incorporating the strengths of LSTM for learning temporal characteristics and CNN for capturing spatial characteristics. The model-based predictive control (MPC) strategy, namely LSTM-CNN-ARX-MPC, is developed by utilizing the model's local linear and global nonlinear features. The control comparison experiments conducted on a water tank system (WTS) show the effectiveness of the developed models and MPC methods.

The problem of health status prediction of insulated-gate bipolar transistors (IGBTs) gains a lot of attention in the field of health management of power electronic equipment. The performance degradation of IGBT gate oxide layer is one of the important failure modes. In view of failure mechanism analysis and easy implementation of monitoring circuit, this paper selects the gate leakage current of IGBT as the precursor parameter of gate oxide degradation, and uses time domain characteristic analysis, gray correlation degree, Mahalanobis distance, Kalman filter and other methods to carry out feature selection and fusion, and finally obtains a health indicator characterizing the degradation of IGBT gate oxide. Based on the Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) network, this paper constructs an IGBT gate oxide degradation prediction model, and performs experimental analysis on the dataset released by NASA-Ames Laboratory, and the average absolute error of performance degradation prediction is as low as 0.0216. Compared with Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), Support Vector Regression (SVR) and CNN-LSTM models, CNN-LSTM network has the highest prediction accuracy. These results show the feasibility of gate leakage current as a precursor parameter of IGBT gate oxide layer failure, as well as the accuracy and reliability of the CNN-LSTM prediction model.

In this paper, we address the problem of car detection from aerial images using Convolutional Neural Networks (CNN). This problem presents additional challenges as compared to car (or any object) detection from ground images because features of vehicles from aerial images are more difficult to discern. To investigate this issue, we assess the performance of two state-of-the-art CNN algorithms, namely Faster R-CNN, which is the most popular region-based algorithm, and YOLOv3, which is known to be the fastest detection algorithm. We analyze two datasets with different characteristics to check the impact of various factors, such as UAV's altitude, camera resolution, and object size. The objective of this work is to conduct a robust comparison between these two cutting-edge algorithms. By using a variety of metrics, we show that YOLOv3 yields better performance in most configurations, except that it exhibits a lower recall and less confident detections when object sizes and scales in the testing dataset differ largely from those in the training dataset.

The automatic recognition of scanned Medieval manuscripts still represents a challenge due to degradation, non standard layouts, or notations. This paper focuses on the Medieval square notation developed around the 11th century which is composed of staff lines, clefs, accidentals, and neumes which are basically connected single notes. We present a novel approach to tackle the automatic transcription by applying CNN/LSTM networks that are trained using the segmentation-free CTC-loss-function which considerably facilitates the GT-production. For evaluation, we use three different manuscripts and achieve a dSAR of 86.0% on the most difficult book and 92.2% on the cleanest one. To further improve the results, we apply a neume dictionary during decoding which yields a relative improvement of about 5%.

In this paper, we address the problem of car detection from aerial images using Convolutional Neural Networks (CNN). This problem presents additional challenges as compared to car (or any object) detection from ground images because features of vehicles from aerial images are more difficult to discern. To investigate this issue, we assess the performance of two state-of-the-art CNN algorithms, namely Faster R-CNN, which is the most popular region-based algorithm, and YOLOv3, which is known to be the fastest detection algorithm. We analyze two datasets with different characteristics to check the impact of various factors, such as UAV’s altitude, camera resolution, and object size. The objective of this work is to conduct a robust comparison between these two cutting-edge algorithms. By using a variety of metrics, we show that none of the two algorithms outperforms the other in all cases.

In order to enhance the comfort of cycling, it is imperative to investigate the effects of vibration on non-motorized bicycle riding from the perspectives of road characteristics and traffic features. Through an analysis of the mechanisms by which road and traffic conditions influence cycling vibrations, 13 influencing factors were identified. Subsequently, the non-motorized bicycle lanes in Wuhan city were selected as the subject of empirical research, where three-axis accelerometers attached to the rider's torso were employed to measure and categorize vibration comfort levels. The experimental road segments were found to exhibit comfort levels falling between slightly uncomfortable and relatively uncomfortable. Further analysis of the influencing factors was conducted using the Random Forest algorithm and Logistic Regression. The results revealed that six factors significantly impact the comfort of cycling: the presence of dedicated non-motorized bicycle lanes, the absence of physical separation between non-motorized and motorized traffic, cycling speed, the number of road surface irregularities, the presence of parking areas within the non-motorized bicycle lane, and the type of non-motorized bicycle. This study provides valuable insights into the factors affecting non-motorized bicycle lane usage and contributes to the refined design of urban non-motorized bicycle infrastructure, thereby facilitating better support for sustainable urban transportation.

The paper briefly reviews the Clifford algebras of space Cl(3,0) and anti-space Cl(0,3) with a particular focus on the paravector representation, emphasizing the fact that both algebras have an isomorphic center represented just by two coordinates. Since the paravector representation allows assigning the scalar element of grade 0 to the time coordinate, we consider the relativity in such two-dimensional spacetime for a uniformly accelerated frame with the constant acceleration 3Hc. Using the Rindler coordinate transformations in two-dimensional spacetime and then applying it to Minkowski coordinates, we obtain the FLRW metric, which in the case of the Clifford algebra of space Cl(3,0) corresponds to the anti-de Sitter (AdS) flat (k=0) case, the negative cosmological term and an oscillating model of the universe. The approach with anti-Euclidean Clifford algebra Cl(0,3) leads to the de Sitter model with the positive cosmological term and the exact form of the scale factor used in modern cosmology.

Background: The prevalence of falls among the older population underscores the imperative of comprehending human adaptations to gait perturbations. Dual-belt treadmills offer a controlled setting for such investigations. The purpose of this study was to examine the effect of acceleration of one belt of the treadmill during three different phases of the gait cycle on kinematic and kinetic parameters and relate these changes to unperturbed gait. Methods: Twenty-one healthy young females walked on a treadmill in a virtual environment, in which five unexpected perturbations were applied to the left belt at the Initial Contact (IC), Mid Stance (MS), and Pre-Swing (PS) phase of the gait cycle. Data from the undisturbed gait and the first disturbance of each trial were extracted for analysis. Results: All perturbations significantly affected the gait pattern, mainly by decreasing the knee extension angle. The perturbation in the IC phase had the most significant effect, resulting in a 248.48% increase in knee flexion torque. The perturbation in the MS phase mainly affected plantar flexion torque, increasing it by 118.18%, while perturbation in the PS phase primarily increased the hip extension torque by 73.02%. Conclusions: The presence of perturbations in the IC and PC phases caused the most aggressive and significant changes in gait parameters.

Weight and age interact to modify cognitive effects of head acceleration events after two seasons of youth football Abstract This is a follow-up study of youth American football players in a second season of play. Season-one findings identified a relationship between a head acceleration measure (HITsp) and cognitive score changes with weight modifying the effect in 9 to 10 year-olds. Sixty-eight youth completed a second season of play wearing helmet-mounted sensors and were assessed with neuropsychological tests pre- and post-season. Regression analysis of the full sample demonstrated a small but significant negative effect of HAE on cognition as indexed by test score changes: R2 = .06, F = 4.06, p = .024. Outcome differences between those who started playing at ages nine to 10, compared to those who started between 11 and 13 were identified: t(66) = -3.39, p<.01, d = -.84, 95th CI -2.77 to -.72. Regression models including players’ weights found that the relationship of weight to outcome was different by group: greater negative cognitive effects were found in younger-heavier players (R2 = 0.21, F =3.21, p = 0.03) and older-lighter players (R2 = 0.18, F =4.26, p <.001). These findings confirmed a negative relationship of HAE and cognitive change and point to player weight as an important developmental factor to consider in understanding concussion biomechanics.

Ultrasonic sensors are inexpensive and can take measurements with high accuracy, even with simple hardware configurations, so they are used in various fields. When multiple ultrasonic sensors exist in the measurement space, crosstalk occurs due to other nodes, which causes measurement errors. Crosstalk includes not only receiving homogeneous signals from other nodes, but also overlap by other signals, and interference by heterogeneous signals. This paper proposes a method that uses frequency sweep keying modulation to be robust against overlap and a faster region-based convolutional neural networks (R-CNN)-based demodulator to reduce the interference caused by heterogeneous signals. The demodulator works by training faster R-CNN with the spectrograms of various received signals and classifying the received signals using a faster R-CNN. Experiments implementing an ultrasonic crosstalk environment showed that compared to on-off keying (OOK), phase-shift keying (PSK), and frequency-shift keying (FSK), the proposed method can implement CDMA even with shorter codes, and is robust against overlap. Compared to correlation-based frequency sweep keying, the time-of-flight error was reduced by approximately 75%. While existing demodulators did not consider heterogeneous signals, the proposed method ignored approximately 99% of the OOK and PSK signals, and approximately 79% of the FSK signals. The proposed method performs better than the existing methods and is expected to be used in various applications.