A personalized FEM model for reproducible measurement of anti-inflammatory drugs in transdermal administration to knee

A personalized model of the human knee for enhancing the inter-individual reproducibility of a measurement method for monitoring Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) after transdermal delivery is proposed. The model is based on the solution of Maxwell Equations in the electric-quasi-stationary limit via Finite Element Analysis. The dimensions of the custom geometry are estimated on the basis of knee circumference at the patella, body mass index, and sex of each individual. An optimization algorithm allows to find out the electrical parameters of each subject by experimental impedance spectroscopy data. Muscular tissues were characterized anisotropically, by extracting Cole–Cole equation parameters from experimental data acquired with twofold excitation, both transversal and parallel to tissue fibers. A sensitivity and optimization analysis aiming at reducing computational burden in model customization achieved a worst-case reconstruction error lower than 5%. The personalized knee model and the optimization algorithm were validated in vivo by an experimental campaign on thirty volunteers, 67% healthy and 33% affected by knee osteoarthritis (Kellgren–Lawrence grade ranging in [1,4]), with an average error of 3%.

Reconstructive Reservoir Computing to Detect Anomaly in Time-series Signals

We propose reconstructive reservoir computing (RRC) for anomaly detection working for time-series signals. This paper investigates its fundamental properties with experiments employing echo state networks (ESNs). The RRC model is a reconstructor to replicate a normal input time-series signal with no delay or a certain delay (delay ≥ 0). In its anomaly detection process, we evaluate instantaneous reconstruction error defined as the difference between input and output signals at each time. Experiments with a sound dataset from industrial machines demonstrate that the error is low for normal signals while it becomes higher for abnormal ones, showing successful anomaly detection. It is notable that the RRC models’ behavior is very different from that of conventional anomaly detection models, that is, those based on forecasting (delay < 0). The error of the proposed reconstructor is explicitly lower than that of a forecaster, resulting in superior distinction between normal and abnormal states. We show that the RRC model is effective over a large range of reservoir parameters. We also illustrate the distribution of the output weights optimized through a training to discuss their roles in the reconstruction. Then, we investigate the influence of the neuronal leaking rate and the delay time shift amount on the transient response and the reconstruction error, showing high effectiveness of the reconstructor in anomaly detection. The proposed RRC will play a significant role for anomaly detection in the present and future sensor network society

Reconstructive Reservoir Computing to Detect Anomaly in Time-series Signals

We propose reconstructive reservoir computing (RRC) for anomaly detection working for time-series signals. This paper investigates its fundamental properties with experiments employing echo state networks (ESNs). The RRC model is a reconstructor to replicate a normal input time-series signal with no delay or a certain delay (delay ≥ 0). In its anomaly detection process, we evaluate instantaneous reconstruction error defined as the difference between input and output signals at each time. Experiments with a sound dataset from industrial machines demonstrate that the error is low for normal signals while it becomes higher for abnormal ones, showing successful anomaly detection. It is notable that the RRC models’ behavior is very different from that of conventional anomaly detection models, that is, those based on forecasting (delay < 0). The error of the proposed reconstructor is explicitly lower than that of a forecaster, resulting in superior distinction between normal and abnormal states. We show that the RRC model is effective over a large range of reservoir parameters. We also illustrate the distribution of the output weights optimized through a training to discuss their roles in the reconstruction. Then, we investigate the influence of the neuronal leaking rate and the delay time shift amount on the transient response and the reconstruction error, showing high effectiveness of the reconstructor in anomaly detection. The proposed RRC will play a significant role for anomaly detection in the present and future sensor network society

Consistency of Bayesian inference with Gaussian process priors for a parabolic inverse problem

We consider the statistical non-linear inverse problem of recovering the absorption term f>0 in the heat equation with given sufficiently smooth functions describing boundary and initial values respectively. The data consists of N discrete noisy point evaluations of the solution u_f. We study the statistical performance of Bayesian nonparametric procedures based on a large class of Gaussian process priors. We show that, as the number of measurements increases, the resulting posterior distributions concentrate around the true parameter generating the data, and derive a convergence rate for the reconstruction error of the associated posterior means. We also consider the optimality of the contraction rates and prove a lower bound for the minimax convergence rate for inferring f from the data, and show that optimal rates can be achieved with truncated Gaussian priors.

Network Intrusion Detection with Nonsymmetric Deep Autoencoding Feature Extraction

To address the problems of high reconstruction error and long training time when using Stack Nonsymmetric Deep Autoencoder (SNDAE) feature extraction technology for intrusion detection, Adam Nonsymmetric Deep Autoencoder (ANDAE) is proposed based on SNDAE. The Adam optimization algorithm is used to update network parameters during training so that the loss function can quickly converge to the ideal value. Under the premise of not affecting the effect of feature extraction, the network structure is simplified, and the training time of the network is reduced to realize the efficient extraction of the rapid growth of high-dimension and nonlinear network traffic features. For the low-dimensional prominent features extracted by ANDAE, Random Forest is used for classification to detect intrusion action, and a network intrusion detection model based on ANDAE feature extraction is implemented. The experimental results on the NSL-KDD and the CIC-IDS2017 datasets show that, compared to the SNDAE-based intrusion detection model, the ANDAE model has an average increase of 6.78% in accuracy, an average of 13.06% in recall, and an average of 14.9% in F1 scores. Feature extraction time is reduced by 23.1% on average. Thus, the ANDAE model is an intrusion detection solution, which can simultaneously improve detection accuracy and time efficiency.

Design and Measurement of a Dual FBG High-Precision Shape Sensor for Wing Shape Reconstruction

FBG shape sensors based on soft substrates are currently one of the research focuses of wing shape reconstruction, where soft substrates and torque are two important factors affecting the performance of shape sensors, but the related analysis is not common. A high-precision soft substrates shape sensor based on dual FBGs is designed. First, the FBG soft substrate shape sensor model is established to optimize the sensor size parameters and get the optimal solution. The two FBG cross-laying method is adopted to effectively reduce the influence of torque, the crossover angle between the FBGs is 2α, and α = 30° is selected as the most sensitive angle to the torquer response. Second, the calibration test platform of this shape sensor is built to obtain the linear relationship among the FBG wavelength drift and curvature, rotation radian loaded vertical force and torque. Finally, by using the test specimen shape reconstruction test, it is verified that this shape sensor can improve the shape reconstruction accuracy, and that its reconstruction error is 6.13%, which greatly improves the fit of shape reconstruction. The research results show that the dual FBG high-precision shape sensor successfully achieves high accuracy and reliability in shape reconstruction.

Recurrent Neural Network-Based Temperature Control System Weight Pruning Based on Nonlinear Reconstruction Error

Recurrent Neural Networks (RNNs) have been widely applied in various fields. However, in real-world application, because most devices like mobile phones are limited to the storage capacity when processing real-time information, an over-parameterized model always slows down the system speed and is not suitable to be employed. In our proposed temperature control system, the RNN-based control model processes the real-time temperature signals. It is necessary to compress the trained model with acceptable loss of control performance for further implementation in the actual controller when the system resource is limited. Inspired by the layer-wise neuron pruning method, in this paper, we apply the nonlinear reconstruction error (NRE) guided layer-wise weight pruning method on the RNN-based temperature control system. The control system is established based on MATLAB/Simulink. In order to compress the model size to save the memory capacity of temperature controller devices, we first prove the validity of the proposed reference-model (ref-model) guided RNN model for real-time online data processing on an actual temperature object; relative experiments are implemented based on a digital signal processor. On this basis, we then verified the NRE guided layer-wise weight pruning method on the well-trained temperature control model. Compared with the classical pruning method, experiment results indicate that the pruned control model based on NRE guided layer-wise weight pruning can effectively achieve the high accuracy at targeted sparsity of the network.

Error Minimization in ECG Signal Reconstruction Using Discrete Wavelet Transform

For clinical study and diagnosis, compression of Electro Cardio Gram (ECG) signal is a fundamental step for processing. However, the compression and reconstruction introduce errors in the signal. Therefore, error minimization is crucial before using these signals for analysis and diagnosis. This paper presents an efficient method to minimize the reconstruction error using the adaptive filtering technique. Better reconstruction was achieved based on higher value of Compression Ratio and lesser value of Percent Root mean squared difference. Daubechies Wavelet easily detects the signal spikes while keeping less error rate using Least Mean Squared Error algorithm. However, the percentage value of error appeared to be minimum when using Daubechies Wavelet because of its small coefficients other than Haar and Coiflet Wavelet. Therefore, it was concluded that Daubechies Wavelet should have been used for error minimization in the reconstructed signal.

Rapid Test Method for Multi-Beam Profile of Phased Array Antennas

The measurement of the phased array antenna (PAA) is completely different from the traditional antenna, due to its multi beam patterns. Usually, each beam pattern of the PAA needs a separate measurement, which makes the overall time extremely long. Thus, the traditional method can no longer meet the efficiency and cost requirements of new PAA measurement. In this paper, a pattern reconstruction method is proposed which significantly reduce the measurement time of multi-beam PAAs. With the known array element patterns (AEP) and theoretical weighted port excitation of the beams, any beam pattern can be predicted by measuring only a certain beam pattern, due to the element excitation coefficient (including the matching, mutual coupling, and manufacturing factors, etc.) of the specific PAA being calculated. The approach has low reconstruction error in term of beam pointing accuracy, side lobe, and co-polar and cross-polar patterns while being validated for large scanning range. Through theoretical derivation and experiments, the effectiveness of the method is verified, and the testing efficiency of the phased array antenna can be improved by 10 times or even more.

Comparison of Empirical Mode Decomposition, Wavelets, and Different Machine Learning Approaches for Patient-Specific Seizure Detection Using Signal-Derived Empirical Dictionary Approach

Analysis of long-term multichannel EEG signals for automatic seizure detection is an active area of research that has seen application of methods from different domains of signal processing and machine learning. The majority of approaches developed in this context consist of extraction of hand-crafted features that are used to train a classifier for eventual seizure detection. Approaches that are data-driven, do not use hand-crafted features, and use small amounts of patients' historical EEG data for classifier training are few in number. The approach presented in this paper falls in the latter category, and is based on a signal-derived empirical dictionary approach, which utilizes empirical mode decomposition (EMD) and discrete wavelet transform (DWT) based dictionaries learned using a framework inspired by traditional methods of dictionary learning. Three features associated with traditional dictionary learning approaches, namely projection coefficients, coefficient vector and reconstruction error, are extracted from both EMD and DWT based dictionaries for automated seizure detection. This is the first time these features have been applied for automatic seizure detection using an empirical dictionary approach. Small amounts of patients' historical multi-channel EEG data are used for classifier training, and multiple classifiers are used for seizure detection using newer data. In addition, the seizure detection results are validated using 5-fold cross-validation to rule out any bias in the results. The CHB-MIT benchmark database containing long-term EEG recordings of pediatric patients is used for validation of the approach, and seizure detection performance comparable to the state-of-the-art is obtained. Seizure detection is performed using five classifiers, thereby allowing a comparison of the dictionary approaches, features extracted, and classifiers used. The best seizure detection performance is obtained using EMD based dictionary and reconstruction error feature and support vector machine classifier, with accuracy, sensitivity and specificity values of 88.2, 90.3, and 88.1%, respectively. Comparison is also made with other recent studies using the same database. The methodology presented in this paper is shown to be computationally efficient and robust for patient-specific automatic seizure detection. A data-driven methodology utilizing a small amount of patients' historical data is hence demonstrated as a practical solution for automatic seizure detection.