Search

Despite the existence of evidence-based rehabilitation strategies that address biomechanical deficits, the persistence of recurrent ankle problems in 70% of patients with acute ankle sprains highlights the unresolved nature of this issue. Artificial intelligence (AI) emerges as a promising tool to identify definitive predictors for ankle sprains. This paper aims to summarize the use of AI in investigating the ankle biomechanics of healthy and subjects with ankle sprains. Articles published between 2010 and 2023 were searched from five electronic databases. 59 papers were included for analysis with regards to: i). types of motion tested (functional vs. purposeful ankle movement); ii) types of biomechanical parameters measured (kinetic vs kinematic); iii) types of sensor systems used (lab-based vs field-based); and, iv) AI techniques used. Most studies (83.1%) examined biomechanics during functional motion. Single kinematic parameter, specifically ankle range of motion, could obtain accuracy up to 100% in identifying injury status. Wearable sensor exhibited high reliability for use in both laboratory and on-field/clinical settings. AI algorithms primarily utilized electromyography and joint angle information as input data. Support vector machine was the most used supervised learning algorithm (18.64%), while artificial neural network demonstrated the highest accuracy in eight studies. The potential for remote patient monitoring is evident with the adoption of field-based devices. Nevertheless, AI-based sensors are underutilized in detecting ankle motions at risk of sprain. We identify three key challenges: sensor designs, the controllability of AI models, and the integration of AI-sensor models, providing valuable insights for future research. [Display omitted] • Application of sensors and artificial intelligence to analyse ankle biomechanics. • Lab sensors predict parameters, paving the way for post-modern rehabilitation. • Field devices show potential for ankle injury prediction. • Focus on streamlined smart sensors and effective data management in the future. [ABSTRACT FROM AUTHOR]

Gradient-based algorithms are commonly used for training radial basis function neural network (RBFNN). However, it is still difficult to avoid vanishing gradient to improve the learning performance in the training process. For this reason, in this paper, an accelerated second-order learning (ASOL) algorithm is developed to train RBFNN. First, an adaptive expansion and pruning mechanism (AEPM) of gradient space, based on the integrity and orthogonality of hidden neurons, is designed. Then, the effective gradient information is constantly added to gradient space and the redundant gradient information is eliminated from gradient space. Second, with AEPM, the neurons are generated or pruned accordingly. In this way, a self-organizing RBFNN (SORBFNN) which reduces the structure complexity and improves the generalization ability is obtained. Then, the structure and parameters in the learning process can be optimized by the proposed ASOL-based SORBFNN (ASOL-SORBFNN). Third, some theoretical analyses including the efficiency of the proposed AEPM on avoiding the vanishing gradient and the stability of SORBFNN in the process of structural adjustment are given, then the successful application of the proposed ASOL-SORBFNN is guaranteed. Finally, to illustrate the advantages of the proposed ASOL-SORBFNN, several experimental studies are examined. By comparing with other existing approaches, the results show that ASOL-SORBFNN performs well in terms of both learning speed and prediction accuracy. [ABSTRACT FROM AUTHOR]

We use a recently developed synchronous Spiking Neural Network (SNN) model to study the problem of learning hierarchically-structured concepts. We introduce an abstract data model that describes simple hierarchical concepts. We define a feed-forward layered SNN model, with learning modeled using Oja's local learning rule, a well known biologically-plausible rule for adjusting synapse weights. We define what it means for such a network to recognize hierarchical concepts; our notion of recognition is robust, in that it tolerates a bounded amount of noise. Then, we present a learning algorithm by which a layered network may learn to recognize hierarchical concepts according to our robust definition. We analyze correctness and performance rigorously; the amount of time required to learn each concept, after learning all of the sub-concepts, is approximately O 1 η k ℓmax log (k) + 1 ɛ + b log (k) , where k is the number of sub-concepts per concept, ℓmax is the maximum hierarchical depth, η is the learning rate, ɛ describes the amount of uncertainty allowed in robust recognition, and b describes the amount of weight decrease for "irrelevant" edges. An interesting feature of this algorithm is that it allows the network to learn sub-concepts in a highly interleaved manner. This algorithm assumes that the concepts are presented in a noise-free way; we also extend these results to accommodate noise in the learning process. Finally, we give a simple lower bound saying that, in order to recognize concepts with hierarchical depth two with noise-tolerance, a neural network should have at least two layers. The results in this paper represent first steps in the theoretical study of hierarchical concepts using SNNs. The cases studied here are basic, but they suggest many directions for extensions to more elaborate and realistic cases. [ABSTRACT FROM AUTHOR]

Find the global optimal solution of the model is one promising research topic in computational intelligent community. Dependent on analogies to natural processes, the evolutionary swarm intelligent algorithms are widely used for solving global optimization problems which directed by the fitness values. In this paper, we propose one efficient fractional global learning machine (Fragmachine) which includes two stages (descending and ascending) to determine the optimal search path. The neural network model is used to approach the given fitness value. Specifically, for the descending stage, the integer gradient of the network output with respect the current location is employed to find the next descending point, while for the ascending stage, the fractional gradient is implemented to climb and escape from the local optimal point. We further propose one adaptive learning rate during training which relies on both the current gradient (integer or fractional) information and the fitness value. Finally, a series of numerical experiments verify the effectiveness of the proposed algorithm, Fragmachine. [ABSTRACT FROM AUTHOR]

The paper presents an approach to machine learning involving the use of a hybrid method in Neural ODE. The authors consider a way of the Neural ODE application with an embedded hybrid method for intelligent manipulator control in a non-deterministic environment. To reduce computational resources, the number of layers of the neural network using ODE is reduced, the solution of which is carried out by a hybrid transformation method. The hybrid method includes normalizing the system, linear transformation, and polynomial transformation. Using the hybrid method in Neural ODE can increase the learning and computation speed. The transformation method has high accuracy with low computational complexity of the algorithm. As a result of applying the hybrid method, the original nonlinear system is transformed to a simpler form. To solve the simplified transformed system, the scheme of the numerical Runge-Kutta method of the third order with an adaptive step is used. The results of the computational experiment show an increase in the speed and accuracy of the calculation. The presented approach to implementing the hybrid method in Neural ODE allows faster training of the neural network of the world model on limited data obtained from the real environment. After training on the data generated by the model of the world, a robotic arm always successfully reaches the object during subsequent interaction with the real environment. This solves the problem of the limited ability of the agent to interact when learning with the real environment in robotic systems. [ABSTRACT FROM AUTHOR]

As an effective data pattern extraction method, machine learning is widely used in the field of nuclear power industry control, and has a great application. In order to solve the problem that there is a serious lack of operation data set for analyzing various variable conditions or accident conditions, this paper uses the maximum likelihood estimation and the least square estimation to prove that, when the simulation data of nuclear power plant and the operation data with different noise terms are used as the training data samples to train the machine learning algorithm models respectively, the difference between the two algorithms at each moment is only related to the mean of noise distribution. Especially when the mean of noise distribution is 0, they are equivalent. The conclusion shows that when the machine learning algorithm model is trained, the simulation data can supplement the operation data set. When the algorithm models are trained with the supplementary data set, the supplementary data will not change the calculations of the original algorithm model. Therefore, the simulation data can be used to expand the operation data set of variable conditions or accident conditions to improve the calculation accuracy of the algorithm models. In order to verify the conclusion, this paper uses simulation data and operation data with different noise terms to train the neural network algorithm models with L2 regularization term, and uses the algorithm models to calculate the steam mass flow rate at the outlet of main steam pipe of steam generator and the water temperature at the bottom of pressurizer. It is further proved that the calculations of thermal hydraulic transient operation parameters of the algorithm models trained by the operation data set with different noise terms are only equivalent to the translation of the calculations of the algorithm models trained by the simulation data set. And the value of translation is the mean of the noise distribution, which proves that the conclusion is effective. [ABSTRACT FROM AUTHOR]

• This paper used ANN hybridized with swarm intelligence algorithms to predict the carbonation resistance of RAC. • A comparison between the performance of proposed prediction models was performed. • The variable importance analysis and parametric analysis of input variables were studied. This paper investigates the prediction of carbonation depth for recycled aggregate concrete (RAC) with machine learning models. Nine parameters including RAC intrinsic properties and environmental conditions were considered as input variables. A dataset comprising 593 test data was used to train, validate, and test machine learning models. Results show that the Random forest model shows superior performance than the Gaussian progress regression model and standalone artificial neural network (ANN) model. All ANN models hybridized with swarm intelligence algorithms outperform the standalone ANN model, especially for the ANN model hybridized with the whale optimization algorithm. All machine learning models show higher accuracy than the existing code models and statistical models. The variable importance analysis shows that the carbonation resistance for RAC was determined by both internal and external factors. Based on the parametric analysis, the robustness of the proposed machine learning models was further confirmed. [ABSTRACT FROM AUTHOR]

Digital holographic microscopy (DHM) is a rapidly developing imaging technique that allows for the reconstruction of three-dimensional images from holographic data. In this review paper, we present an overview of the state-of-the-art in on-axis DHM reconstruction methods. We classify these methods into three categories: direct, iterative, and machine learning based algorithms. In the direct category, we summarize methods that directly apply the holographic data to reconstruct the sample without any iterative optimization. These methods are fast and efficient, but they often suffer from noise and low contrast. The iterative category contains methods that use an iterative optimization process to improve the quality of the reconstructed image. This can lead to higher accuracy and noise reduction, but at the cost of increased computational complexity. Within the iterative category, we also discuss methods that assume sparsity in the reconstruction and other domains, which can further improve the convergence of the reconstruction algorithm. Finally, we describe machine learning methods for DHM reconstruction. These methods use deep neural networks trained on large datasets to learn the relationship between 2D holographic data and their reconstructed fields. Machine learning based methods can achieve similar results to the iterative methods at a fraction of the time, however, they require large amounts of computational resources and training data. We provide a summary of these methods in a table (Table B.1) and a map (Fig. 1), and unify the naming convention of all methods, allowing readers to easily compare and contrast the different approaches. • Review of current State-of-the-Art in on-axis digital holographic microscopy. • Compares methods, experimental setups, pros, and cons of each method. • Unifies all methods under single notation. • Provides an easy-to-read map of methods for easy comparison. [ABSTRACT FROM AUTHOR]

• A machine-learning (ML) method was developed to predict sooting tendency (YSI) of fuel mixtures called soot-specialized artificial neural network. • The YSI database was established from previous experiment including 151 fuel mixtures and 21 pure components. • Synergistic effects of "n-cetane& iso -cetane" and " iso -cetane&butylcyclohexane" were predicted by ML and validated by experiment and simulation. • The ML model provided an accurate and low-cost approach to supplement experimental data with mean relative error less than 7%. Interaction between components is an important characteristic in a fuel mixture, which has significant influence on sooting tendency and is strongly related to chemical structures of components. It is time-consuming to study mixing effects between multiple fuels on soot formation through experimental methods, so it is essential to develop a simple and reliable statistical method to supplement experimental data. All sooting tendency data was transformed into a dimensionless index called Yield Sooting Index (YSI). A machine learning algorithm, named soot-specialized artificial neural network (S-ANN), was developed in this paper to predict sooting tendency of fuel surrogates. The mean relative error of S-ANN prediction results can be reduced to about 6%, which is better than that of linear method (13.72%). As a widely used index, YSI can characterize the relative scale of fuel sooting tendency but cannot clearly represent the synergistic or neutralization effects of mixture components. Therefore, this work proposes Soot Synergistic/Neutralization Index (SSNI) using the deviation between predicted value and linear fitting value. SSNIs are transformed from the prediction results of S-ANN and can indicate the interactions between fuels in terms of sooting tendency, which can also provide important information for component selection and sooting tendency adjustment during formulation process of surrogates. Two groups of binary mixtures were selected to validate the algorithm and index. From validation experiment, it is indicated that negative synergistic effect exists between n-cetane and iso -cetane; positive synergistic effect exists between iso -cetane and butylcyclohexane. These two synergistic effects are simulated and analyzed through previously developed chemical mechanism. The results reveal that soot-specialized artificial neural network algorithm have a strong ability to predict sooting tendency of fuel mixtures, which can be used as an accurate and low-cost approach to supplement experimental data. SSNI can help to understand the interactions of fuel mixtures and provide valuable guidance for the construction of surrogates. [ABSTRACT FROM AUTHOR]

• Deep learning algorithm for identification of type and flight mode of detected drones. • Novel deep learning algorithm for detection of multiple drones in radio frequency domain. • Results of three indoor experiments concerning collecting and analyzing radio signals from drones. • Public available database with radio signals of drones from two different radio frequency bands. Unmanned aerial systems, especially drones have gone through remarkable improvement and expansion in recent years. Drones have been widely utilized in many applications and scenarios, due to their low price and ease of use. However, in some applications drones can pose a malicious threat. To diminish risks to public security and personal privacy, it is necessary to deploy an effective and affordable anti-drone system in sensitive areas to detect, localize, identify, and defend against intruding malicious drones. This research article presents a new publicly available radio frequency drone dataset and investigates detection and identification methodologies to detect single or multiple drones and identify a single detected drone's type. Moreover, special attention in this paper has been underlined to examine the possibility of using deep learning algorithms, particularly fully connected deep neural networks as an anti-drone solution within two different radio frequency bands. We proposed a supervised deep learning algorithm with fully-connected deep neural network models that use raw drone signals rather than features. Regarding the research results, the proposed algorithm shows a lot of potentials. The probability of detecting a single drone is 99.8%, and the probability of type identification is 96.1%. Moreover, the results of multiple drones detection demonstrate an average accuracy of 97.3%. There have not been such comprehensive publications, to this time, in the open literature that have presented and enlightened the problem of multiple drones detection in the radio frequency domain. [ABSTRACT FROM AUTHOR]

Recently, artificial neural networks have been widely used for classification. It is important to optimize the weight parameters and topological structure of the neural network simultaneously. These two tasks are interdependent and should be solved at the same time to achieve a better result. However, existing works cannot balance the accuracy and diversity of neural networks very well. In this paper, a cooperative co-evolutionary algorithm is proposed to simultaneously evolve artificial neural network topology, neuron attributes, and connection weights. In the proposed algorithm, two effective strategies are proposed. First, the niche-based strategy is used in the evolutionary and cooperative process to refine the local search ability. In this way, a set of candidate networks with a higher level of output diversity is obtained. Second, a two-step comparison scheme is designed to acquire a compact ensemble network. Moreover, a fully connected weights matrix crossover scheme is used to avoid destroying the network structure. The proposed algorithm is tested on the benchmark classification problems in the UCI machine learning repository and compared with other state-of-the-art methods. The experimental results show that the proposed niche-based cooperative co-evolutionary ensemble neural network has a higher capability of generalization compared with other methods in six of nine kinds of classification problems. Furthermore, the proposed ensemble neural network has relatively low complexity. • A niche-based cooperative co-evolutionary algorithm is developed to evolve ANN. • A fully connected matrix crossover strategy is used to protect the network structure. • A set of networks with different structures can improve the performance of ensemble. [ABSTRACT FROM AUTHOR]

• Extreme learning machine is employed to predict data and filter noised data. • Data shortage and noised data obstacles for parameter extraction are solved. • Eight meta -heuristic algorithms are employed to two solid oxide fuel cell models. • Practical performance of the proposed method is validated in the 5 kW cell stack. • Two constructive recommendations are proposed for future development. A precise, fast, and robust parameter extraction technique of solid oxide fuel cell models is extremely crucial for optimal control and behavior analysis. In this paper, a novel extreme learning machine based method is proposed to extract unknown parameters of solid oxide fuel cell models including electrochemical model and simple electrochemical model. At first, extreme learning machine is applied to overcome two thorny obstacles (e.g., data shortage and noised data) via predicting additional data and updating noised data. Then, both original data collected from a 5 kW solid oxide fuel cell stack and processed data are transferred to effectively guide eight prominent meta-heuristic algorithms for effective parameter extraction. The performance of extreme learning machine is thoroughly investigated in two typical operation conditions through a comprehensive comparison based on various training data. Simulation results validate that the proposed approach can effectively contribute to searching efficient model parameters along with high accuracy, prominent stability, high speed, and great robustness. Particularly, the accuracy of parameter extraction for electrochemical model and simple electrochemical model can be improved by 49.3% and 65.6% at most, respectively. [ABSTRACT FROM AUTHOR]

• A novel explainable neural network is proposed for modeling main effects and structured interactions. • The GAMI-Net is a disentangled feedforward network with multiple additive subnetworks. • GAMI-Net takes into account three interpretability constraints: sparsity, heredity, marginal clarity. • An adaptive training algorithm is developed for training GAMI-Net efficiently. • GAMI-Net enjoys superior interpretability and outperforms benchmark methods. The lack of interpretability is an inevitable problem when using neural network models in real applications. In this paper, an explainable neural network based on generalized additive models with structured interactions (GAMI-Net) is proposed to pursue a good balance between prediction accuracy and model interpretability. GAMI-Net is a disentangled feedforward network with multiple additive subnetworks; each subnetwork consists of multiple hidden layers and is designed for capturing one main effect or one pairwise interaction. Three interpretability aspects are further considered, including a) sparsity, to select the most significant effects for parsimonious representations; b) heredity, a pairwise interaction could only be included when at least one of its parent main effects exists; and c) marginal clarity, to make main effects and pairwise interactions mutually distinguishable. An adaptive training algorithm is developed, where main effects are first trained and then pairwise interactions are fitted to the residuals. Numerical experiments on both synthetic functions and real-world datasets show that the proposed model enjoys superior interpretability and it maintains competitive prediction accuracy in comparison to the explainable boosting machine and other classic machine learning models. [ABSTRACT FROM AUTHOR]