Neural Network-based Estimation Research Articles

Neural network-based estimation of lower limb joint kinematics: A minimally intrusive approach for gait analysis

The establishment of a quantitative gait analysis system holds paramount importance, particularly in the context of functional rehabilitation of the lower limbs. Clinicians emphasize the imperative for sensors to be portable, compact, integrated, and non-intrusive, crucial characteristics in the rehabilitation field to facilitate their use and ensure optimal integration into care protocols. This study investigates an innovative approach aimed at reducing the reliance on body-fixed sensors by harnessing their data within a neural network, thus concentrating on the joint kinematics of the lower limbs. The primary objective is to estimate the flexion-extension angles of the hip, knee, and ankle during walking, utilizing data collected by two sensors positioned on the subject's legs. Initially, the neural network undergoes training with calculated data (leg tilt angles and angular velocities) sourced from the OpenSim database, followed by further refinement with experimental data obtained from a subject walking on a treadmill, wherein leg tilt angles and angular velocities are measured. The significance of this research is underscored by the demonstrated capability, through conducted tests, of the implemented networks to efficiently fuse data from a minimal set of sensors. Consequently, the proposed approach emerges as both practical and minimally intrusive, facilitating a robust evaluation of gait kinematic parameters.

An algorithm for estimating the ultimate speed of vehicles on wet roads

Real-time estimation of a vehicle’s ultimate hydroplaning speed is crucial for ensuring its safety and stability, especially in rainy weather conditions. This paper proposes an algorithm for estimating the ultimate hydroplaning speed of vehicles on wet and slippery road surfaces based on finite element analysis of tire. Using the ABAQUS simulation software, a finite element model of a 205/55 R16 radial tire is established. Subsequently, a fluid-structure coupling model between the tire, water layer, and road surface is developed. The finite element method is employed to simulate and analyze the effects of tire inflation pressure, water layer thickness, vertical load, and tread wear on the vehicle’s ultimate hydroplaning speed. Simulation results indicate that an increase in tire inflation pressure and vertical load leads to an increase in the ultimate hydroplaning speed, while tread wear and an increase in water layer thickness result in a decrease. Finally, based on the simulation analysis data, a BP neural network-based estimation algorithm for the vehicle’s ultimate hydroplaning speed is proposed by combining the relationship between tire inflation pressure, tire vertical load, tread wear, water layer thickness, and ultimate hydroplaning speed. The results show that the estimated ultimate hydroplaning velocity profile has a high degree of overlap with the actual ultimate hydroplaning velocity profile, with a maximum error of no more than 5 km/h and an average percentage error of 2.31416%.

Statistical Spatially Inhomogeneous Diffusion Inference

Inferring a diffusion equation from discretely observed measurements is a statistical challenge of significant importance in a variety of fields, from single-molecule tracking in biophysical systems to modeling financial instruments. Assuming that the underlying dynamical process obeys a d-dimensional stochastic differential equation of the form dx_t = b(x_t)dt + \Sigma(x_t)dw_t, we propose neural network-based estimators of both the drift b and the spatially-inhomogeneous diffusion tensor D = \Sigma\Sigma^T/2 and provide statistical convergence guarantees when b and D are s-Hölder continuous. Notably, our bound aligns with the minimax optimal rate N^{-\frac{2s}{2s+d}} for nonparametric function estimation even in the presence of correlation within observational data, which necessitates careful handling when establishing fast-rate generalization bounds. Our theoretical results are bolstered by numerical experiments demonstrating accurate inference of spatially-inhomogeneous diffusion tensors.

Neural network-based estimation of biomechanical vocal fold parameters.

Vocal fold (VF) vibrations are the primary source of human phonation. High-speed video (HSV) endoscopy enables the computation of descriptive VF parameters for assessment of physiological properties of laryngeal dynamics, i.e., the vibration of the VFs. However, underlying biomechanical factors responsible for physiological and disordered VF vibrations cannot be accessed. In contrast, physically based numerical VF models reveal insights into the organ's oscillations, which remain inaccessible through endoscopy. To estimate biomechanical properties, previous research has fitted subglottal pressure-driven mass-spring-damper systems, as inverse problem to the HSV-recorded VF trajectories, by global optimization of the numerical model. A neural network trained on the numerical model may be used as a substitute for computationally expensive optimization, yielding a fast evaluating surrogate of the biomechanical inverse problem. This paper proposes a convolutional recurrent neural network (CRNN)-based architecture trained on regression of a physiological-based biomechanical six-mass model (6 MM). To compare with previous research, the underlying biomechanical factor "subglottal pressure" prediction was tested against 288 HSV ex vivo porcine recordings. The contributions of this work are two-fold: first, the presented CRNN with the 6 MM handles multiple trajectories along the VFs, which allows for investigations on local changes in VF characteristics. Second, the network was trained to reproduce further important biomechanical model parameters like VF mass and stiffness on synthetic data. Unlike in a previous work, the network in this study is therefore an entire surrogate of the inverse problem, which allowed for explicit computation of the fitted model using our approach. The presented approach achieves a best-case mean absolute error (MAE) of 133 Pa (13.9%) in subglottal pressure prediction with 76.6% correlation on experimental data and a re-estimated fundamental frequency MAE of 15.9 Hz (9.9%). In-detail training analysis revealed subglottal pressure as the most learnable parameter. With the physiological-based model design and advances in fast parameter prediction, this work is a next step in biomechanical VF model fitting and the estimation of laryngeal kinematics.

Denoising autoencoder genetic programming: strategies to control exploration and exploitation in search

Denoising autoencoder genetic programming (DAE-GP) is a novel neural network-based estimation of distribution genetic programming approach that uses denoising autoencoder long short-term memory networks as a probabilistic model to replace the standard mutation and recombination operators of genetic programming. At each generation, the idea is to capture promising properties of the parent population in a probabilistic model and to use corruption to transfer variations of these properties to the offspring. This work studies the influence of corruption and sampling steps on search. Corruption partially mutates candidate solutions that are used as input to the model, whereas the number of sampling steps defines how often we re-use the output during model sampling as input to the model. We study the generalization of the royal tree problem, the Airfoil problem, and the Pagie-1 problem, and find that both corruption strength and the number of sampling steps influence exploration and exploitation in search and affect performance: exploration increases with stronger corruption and lower number of sampling steps. The results indicate that both corruption and sampling steps are key to the success of the DAE-GP: it permits us to balance the exploration and exploitation behavior in search, resulting in an improved search quality. However, also selection is important for exploration and exploitation and should be chosen wisely.

Adaptive Neural Coordinated Control for Multiple Euler-Lagrange Systems With Periodic Event-Triggered Sampling.

This article addresses the event-triggered coordinated control problem for multiple Euler-Lagrange systems subject to parameter uncertainties and external disturbances. Based on the event-triggered technique, a distributed coordinated control scheme is first proposed, where the neural network-based estimation method is incorporated to compensate for parameter uncertainties. Then, an input-based continuous event-triggered (CET) mechanism is developed to schedule the triggering instants, which ensures that the control command is activated only when some specific events occur. After that, by analyzing the possible finite-time escape behavior of the triggering function, the real-time data sampling and event monitoring requirement in the CET strategy is tactfully ruled out, and the CET policy is further transformed into a periodic event-triggered (PET) one. In doing so, each agent only needs to monitor the triggering function at the preset periodic sampling instants, and accordingly, frequent control updating is further relieved. Besides, a parameter selection criterion is provided to specify the relationship between the control performance and the sampling period. Finally, a numerical example of attitude synchronization for multiple satellites is performed to show the effectiveness and superiority of the proposed coordinated control scheme.

Strain Gauge Neural Network-Based Estimation as an Alternative for Force and Torque Sensor Measurements in Robot Manipulators

When a robotic manipulator interacts with its environment, the end-effector forces need to be measured to assess if a task has been completed successfully and for safety reasons. Traditionally, these forces are either measured directly by a 6-dimensional (6D) force–torque sensor (mounted on a robot’s wrist) or by estimation methods based on observers, which require knowledge of the robot’s exact model. Contrary to this, the proposed approach is based on using an array of low-cost 1-dimensional (1D) strain gauge sensors mounted beneath the robot’s base in conjunction with time series neural networks, to estimate both the end-effector 3-dimensional (3D) interaction forces as well as robot joint torques. The method does not require knowledge of robot dynamics. For comparison reasons, the same approach was used but with 6D force sensor measurements mounted beneath the robot’s base. The trained networks showed reasonably good performance, using the long-short term memory (LSTM) architecture, with a root mean squared error (RMSE) of 1.945 N (vs. 2.004 N; 6D force–torque sensor-based) for end-effector force estimation and 3.006 Nm (vs. 3.043 Nm; 6D force–torque sensor-based) for robot joint torque estimation. The obtained results for an array of 1D strain gauges were comparable with those obtained with a robot’s built-in sensor, demonstrating the validity of the proposed approach.

RDRGSE: A Framework for Noncoding RNA-Drug Resistance Discovery by Incorporating Graph Skeleton Extraction and Attentional Feature Fusion.

Identifying noncoding RNAs (ncRNAs)-drug resistance association computationally would have a marked effect on understanding ncRNA molecular function and drug target mechanisms and alleviating the screening cost of corresponding biological wet experiments. Although graph neural network-based methods have been developed and facilitated the detection of ncRNAs related to drug resistance, it remains a challenge to explore a highly trusty ncRNA-drug resistance association prediction framework, due to inevitable noise edges originating from the batch effect and experimental errors. Herein, we proposed a framework, referred to as RDRGSE (RDR association prediction by using graph skeleton extraction and attentional feature fusion), for detecting ncRNA-drug resistance association. Specifically, starting with the construction of the original ncRNA-drug resistance association as a bipartite graph, RDRGSE took advantage of a bi-view skeleton extraction strategy to obtain two types of skeleton views, followed by a graph neural network-based estimator for iteratively optimizing skeleton views aimed at learning high-quality ncRNA-drug resistance edge embedding and optimal graph skeleton structure, jointly. Then, RDRGSE adopted adaptive attentional feature fusion to obtain final edge embedding and identified potential RDRAs under an end-to-end pattern. Comprehensive experiments were conducted, and experimental results indicated the significant advantage of a skeleton structure for ncRNA-drug resistance association discovery. Compared with state-of-the-art approaches, RDRGSE improved the prediction performance by 6.7% in terms of AUC and 6.1% in terms of AUPR. Also, ablation-like analysis and independent case studies corroborated RDRGSE generalization ability and robustness. Overall, RDRGSE provides a powerful computational method for ncRNA-drug resistance association prediction, which can also serve as a screening tool for drug resistance biomarkers.

EEG-based brain connectivity analysis in autism spectrum disorder: Unraveling the effects of bumetanide treatment

Emerging evidence suggests that cognitive impairment associated with brain network disorders in people with autism could be improved with medications such as bumetanide. However, the extent to which bumetanide is effective in improving brain function in these individuals has not been adequately studied. The main purpose of this study is to investigate the nonlinear brain connectivity and topological changes in brain networks of people with autism spectrum disorders (ASD) after a three-month course of bumetanide treatment. We used electroencephalography (EEG) data of nine participants recorded during the face emotion recognition activity in two stages before and after bumetanide treatment. Brain connectivity matrix was calculated using a neural network-based estimator. Graph criteria and statistical tests have been used to determine the effects of bumetanide treatment on children and adolescents with autism. Bumetanide treatment significantly alters the brain connectivity networks based on stimuli type. Differences in brain connectivity related to the sad stimuli are more significant. The most of the significant changes of the strength graph metric was in the occipital electrodes and electrodes related to the right hemisphere. These results suggest that bumetanide may affect effective connectivity and be used a promising treatment for improving social interactions in patients with autism. It also suggests that brain connectivity patterns can be considered as a neural marker to be used in the development of new therapies.

Artificial neural network-based estimation for rotor-flux model reference adaptive system

At the start, this paper focuses on the function of a rotor-flux model reference adaptive system (RF-MRAS) and in the following part on the realization and application of artificial neural networks (ANN) in a sensorless induction motor drive. Afterwards, a data collection and usage process for the training of ANN is described. In the final part, experimental results of ANN's ability to estimate rotor flux are presented. According to simulations, ANN estimations are accurate and its application as a part of a control scheme looks promising.

Multiarea Inertia Estimation Using Convolutional Neural Networks and Federated Learning

With the increase in penetration of renewable energy sources (RES), traditional inertia estimation techniques based purely on the number of online synchronous generators are increasingly unsuitable, ultimately leading towards suboptimal frequency control in the electric power grid. The stochastic nature of RES additionally makes the system inertia a time-varying quantity. Furthermore, the frequency and inertial response of power systems change drastically in multiarea power systems with interconnected tie-lines. Hence, it is important for state/parameter estimation (e.g., inertia) in multiarea systems, while ensuring communication between each of the areas. In this article, a client–server-based federated learning framework is used to estimate power system inertia in a multiarea system. Federated learning is a machine learning technique where multiple decentralized devices are trained with local data, and a global model is updated and redistributed by a central server by aggregating the trained weights of the decentralized devices, without exchanging the local data. Using local frequency measurements, obtained from the phase-locked loop of an energy storage system, the inertia at each of the areas can be estimated locally via offline training using convolutional neural networks (CNNs), whereas the CNN weights update in an online fashion. The framework, tested on a two-area power system, accurately estimated the inertia constant for both independent and identically distributed (IID) and non-IID data. Furthermore, the CNN-based method outperformed conventional neural network-based estimation techniques in terms of number of communication rounds and estimation accuracy.

Neural network-based event-triggered data-driven control of disturbed nonlinear systems with quantized input

This paper is devoted to design an event-triggered data-driven control for a class of disturbed nonlinear systems with quantized input. A uniform quantizer reconstructed with decreasing quantization intervals is employed to reduce the quantization error. A neural network-based estimation strategy is proposed to estimate both the pseudo partial derivative and disturbances. Consequently, an input triggering rule for single-input single-output systems is provided by incorporating the estimated disturbances, the quantization error bound and tracking errors. Resorting to the Lyapunov method, sufficient conditions for synthesized error systems to be uniformly ultimately bounded are presented. The validity of the proposed scheme is demonstrated via a simulation example.

Neural Network-based Estimation of Microbubbles Generated in Cardiopulmonary Bypass Circuit: A Clinical Application Study.

The cardiopulmonary bypass system used in cardiac surgery can generate microbubbles (MBs) that may cause complications, such as neurocognitive dysfunction, when delivered into the blood vessel. Estimating the number of MBs generated, thus, is necessary to enable the surgeons to deal with it. To this end, we previously proposed a neural network-based model for estimating the number of MBs from four factors measurable from the cardiopulmonary bypass system: suction flow rate, venous reservoir level, blood viscosity, and perfusion flow rate. However, the model has not been adapted to the data collected from actual surgery cases. In this study, the accuracy of MBs estimated by the proposed model was examined in four clinical cases. The results showed that the coefficient of determination between estimated MBs and the measured MBs throughout the surgeries was R2=0.558 (p<0.001). We found that the surgical treatments, such as administration of drugs, fluids and blood transfusions, increased the number of measured MBs. The coefficient of determination increased to R2= 0.8762 (p<0.001) by excluding the duration of these treatments. This result indicates that the model can estimate the number of MBs with high accuracy under the clinical environment.

Portfolio Optimization Using Mean-Semi Variance approach with Artificial Neural Networks: Empirical Evidence from Pakistan

Purpose: The challenge of managing a portfolio effectively is allocating capital among numerous stock holdings to achieve maximum profit. Therefore, the purpose of this study is to guide investors in developing optimal portfolios in the stock market of Pakistan.&#x0D; Design/Methodology/Approach: To pick and optimize a portfolio in the most effective way possible, we used the daily closing stock prices of a sample of listed firms at the Pakistan stock exchange. The study applied the mean semi-variance approach and compared the performance of portfolios with equally weighted portfolios under artificial neural networks and historical-based return estimation in Pakistan.&#x0D; Findings: The result shows that artificial neural network-based estimation of the expected return vector has outperformed the historical return estimation under mean semi-variance portfolio optimization and constrained mean semi-variance portfolios based on the Sharp ratio in Pakistan.&#x0D; Implications/Originality/Value: The study suggests that investors, fund managers, and portfolio analysts should focus on the more sophisticated neural network-based choice for the development of portfolios in the equity market of Pakistan.

Analysis of the Vibration Characteristics of a Leaf Spring System Using Artificial Neural Networks.

The real-time vibrations occurring in a leaf spring system may cause undesirable effects, such as stresses, strains, deflections, and surface deformations over the system. In order to detect the most appropriate working conditions in which the leaf spring system will work more stably and also to design optimized leaf spring systems, these external effects have to be detected with high accuracy. In this work, artificial neural network-based estimators have been proposed to analyze the vibration effects on leaf spring systems. In the experimental studies carried out, the vibration effects of low, medium, and high-pressure values applied by a hydraulic piston on a steel leaf spring system have been analyzed by a 3-axial accelerometer. After the experimental studies, the Radial Basis Artificial Neural Network (RBANN) and Cascade-Forward Back-Propagation Artificial Neural Network (CFBANN) based nonlinear artificial neural network structures have been proposed to analyze the vibration data measured from the leaf spring system under relevant working conditions. The simulation results represent that the RBANN structure can estimate the real-time vibrations occurring on the leaf spring system with higher accuracy and reaches lower RMS error values when compared to the CFBANN structure. In general, it can be concluded that the RBANN and CFBANN network structures can successfully be used in the estimation of real-time vibration data.

A rapid neural network-based demand estimation for generic buildings considering the effect of soft/weak story

Recent earthquakes clarified that existing a soft/weak-story in a building could completely invert the failure mechanisms of the structure. Several studies were implemented to evaluate the potential risk subjected to the buildings under the earthquake hazard. However, these researchers discarded the effect of soft/weak-story on the demand parameters of their models. This paper presents a rapid demand estimation framework for generic buildings considering the effect of soft/weak-story. In this regard, the simplified model developed according to the HAZUS approach is rectified to apply the effect of soft/weak-story on the structural behavior of the generic buildings. Artificial neural networks are implemented to remove the required time-consuming nonlinear response history analyses from the post-earthquake actions. This research utilized a suite of 111 earthquake records, originally developed by the SAC project. These motions are uniformly scaled from to to obtain a generalized dataset with different intensities. Bayesian optimization algorithm is conducted to achieve a prediction model with optimized hyperparameters. Results clarify that the proposed method is reliable and computationally cost-effective in predicting the demand parameters of generic buildings. This framework can be used in the body of a risk assessment platform to facilitate the emergency response to the earthquake hazard.

Artificial neural network-based estimation of COVID-19 case numbers and effective reproduction rate using wastewater-based epidemiology

As a cost-effective and objective population-wide surveillance tool, wastewater-based epidemiology (WBE) has been widely implemented worldwide to monitor the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA concentration in wastewater. However, viral concentrations or loads in wastewater often correlate poorly with clinical case numbers. To date, there is no reliable method to back-estimate the coronavirus disease 2019 (COVID-19) case numbers from SARS-CoV-2 concentrations in wastewater. This greatly limits WBE in achieving its full potential in monitoring the unfolding pandemic. The exponentially growing SARS-CoV-2 WBE dataset, on the other hand, offers an opportunity to develop data-driven models for the estimation of COVID-19 case numbers (both incidence and prevalence) and transmission dynamics (effective reproduction rate). This study developed artificial neural network (ANN) models by innovatively expanding a conventional WBE dataset to include catchment, weather, clinical testing coverage and vaccination rate. The ANN models were trained and evaluated with a comprehensive state-wide wastewater monitoring dataset from Utah, USA during May 2020 to December 2021. In diverse sewer catchments, ANN models were found to accurately estimate the COVID-19 prevalence and incidence rates, with excellent precision for prevalence rates. Also, an ANN model was developed to estimate the effective reproduction number from both wastewater data and other pertinent factors affecting viral transmission and pandemic dynamics. The established ANN model was successfully validated for its transferability to other states or countries using the WBE dataset from Wisconsin, USA.

Physics-Informed Disturbance Estimation and Nonlinear Controller Design for a Multi-Axis Gimbal System

The increasing demand for target tracking, environmental surveys, surveillance and mapping requires multi-axis gimbal systems with high tracking and stabilization performance. In this study, a new torque estimation structure is proposed to compute the complex disturbances negatively affecting the performance of the system. Two different control strategies based on Active Disturbance Rejection Control (ADRC) and Estimated Torque Model are implemented on a two-axis gimbal system. In the first strategy, the purpose is to improve the robustness, environmental adaptability and tracking accuracy of the system. The tuning effort of the ADRC is reduced by integrating a Neural Network (NN) based disturbance compensator (NN assisted ADRC). In the second strategy, NN is replaced with an Estimated Torque Model (ETM assisted ADRC), whose inputs come from plant outputs. The simulation results show that, both NN and ETM assisted control structures decrease the tracking errors. However, the improvements achieved by the physics-informed neural network-based estimator are more significant.

Wearable Sensor-Based Step Length Estimation During Overground Locomotion Using a Deep Convolutional Neural Network.

Step length is a critical gait parameter that allows a quantitative assessment of gait asymmetry. Gait asymmetry can lead to many potential health threats such as joint degeneration, difficult balance control, and gait inefficiency. Therefore, accurate step length estimation is essential to understand gait asymmetry and provide appropriate clinical interventions or gait training programs. The conventional method for step length measurement relies on using foot-mounted inertial measurement units (IMUs). However, this may not be suitable for real-world applications due to sensor signal drift and the potential obtrusiveness of using distal sensors. To overcome this challenge, we propose a deep convolutional neural network-based step length estimation using only proximal wearable sensors (hip goniometer, trunk IMU, and thigh IMU) capable of generalizing to various walking speeds. To evaluate this approach, we utilized treadmill data collected from sixteen able-bodied subjects at different walking speeds. We tested our optimized model on the overground walking data. Our CNN model estimated the step length with an average mean absolute error of 2.89 ± 0.89 cm across all subjects and walking speeds. Since wearable sensors and CNN models are easily deployable in real-time, our study findings can provide personalized real-time step length monitoring in wearable assistive devices and gait training programs.

Space fractional kinetic model for different types of suspension profiles in turbulent flows with a neural network-based estimation of fractional orders

Estimation of particle concentration distribution profile is important to understand the inherent mixing process, changes of river morphology, and nature of transportation of particles. In this study, the effects of non-locality are investigated to analyze, distinguish and predict different types of profiles of the suspended concentration distribution in turbulent flows through open channels, pipes, and mixing boxes. The effects of non-locality are captured through fractional derivatives and integrals. Starting from the fractional Liouville equation and considering the power-law jumps of particles in the carrier fluid, the fractional kinetic equation is derived through mathematical analysis. The kinetic model is solved and an analytical model is proposed using the Laplace transformation. The proposed solution contains the two parameters (as α the order of fractional model and μ the shape parameter) Mittag-Leffler function and is a more general one that contains several previous models as special cases. The proposed model can predict four different types of suspension concentration distributions namely type-I, type-II, plume-like, and convex type profiles under different conditions. Further, the model is validated with existing experimental data of dilute and dense flow in open channels and pipes, flows in square pipes, slurry pipelines, and fluid mud gravity currents. Satisfactory results are obtained. A detailed non-linear multivariate regression analysis is carried out with a wide range of selected data and models are proposed to compute model parmeters. To increase the efficiency of the model, an artificial neural network (ANN) model is proposed to predict the model parameters. The ANN-based models show better results than the regression analysis. Validation results show that non-locality has significant effects on the structure of the concentration profiles. From the analysis of the model parameters, it is found that type-I profile occurs under the subdiffusion process where α⩽1 and μ=1 and other three types (type-II, plume-like and convex) of profiles are associated with the superdiffusion process when α>1 and μ>1.