Significant Classification Accuracy Research Articles

Roles of the Default Mode Network in Different Aspects of Self-representation When Remembering Social Autobiographical Memories.

Autobiographical memory (AM) is episodic memory for personally experienced events, in which self-representation is more important than that in laboratory-based memory. Theoretically, self-representation in a social context is categorized as the interpersonal self (IS) referred to in a social interaction with a person or the social-valued self (SS) based on the reputation of the self in the surrounding society. Although functional neuroimaging studies have demonstrated the involvement of the default mode network (DMN) in self-representation, little is known about how the DMN subsystems contribute differentially to IS-related and SS-related AMs. To elucidate this issue, we used fMRI to scan healthy young adults during the recollection of AMs. We performed multivariate pattern analysis (MVPA) and assessed functional connectivity in the DMN subsystems: the midline core, medial temporal lobe (MTL), and dorsomedial pFC (dmPFC) subsystems. The study yielded two main sets of findings. First, MVPA revealed that all DMN subsystems showed significant classification accuracy between IS-related and nonsocial-self-related AMs, and IS-related functional connectivity of the midline core regions with the retrosplenial cortex of the MTL subsystem and the dmPFC of the dmPFC subsystem was significant. Second, MVPA significantly distinguished between SS-related and nonsocial-self-related AMs in the midline core and dmPFC subsystems but not in the MTL subsystem, and SS-related functional connectivity with the midline core regions was significant in the temporal pole and TPJ of the dmPFC subsystem. Thus, dissociable neural mechanisms in the DMN could contribute to different aspects of self-representation in social AMs.

Human-robot collaborative disassembly enabled by brainwaves and improved generative adversarial network

Human-robot collaboration (HRC) can greatly facilitate the disassembling processes of end-of-life (EoL) products. For the robot in HRC, an intuitive control function enabled by a brain-machine interface (BMI) and human brainwaves will be useful to support dynamic decision-making under various disassembly conditions. However, a major challenge in developing a BMI-enabled HRC disassembly system is that there are usually a limited number of brainwave signals available for system training. To address the challenge, a novel model, namely TCN-D2GAN, was designed in this study. The model was used to augment acquired brainwave signals and support the establishment of the BMI-enabled HRC disassembly system. The innovations of the TCN-D2GAN model come from the following aspects: (i) the strengths of a temporal convolutional network (TCN) and a dual discrimination generative adversarial network (D2GAN) were combined to accomplish high-performance signal augmentation, and (ii) new loss functions were defined in the model to avoid the use of hyperparameters, so the training effect of the model can be optimised. Based on two public electroencephalography (EEG) datasets, experiments for signal augmentation and system training were carried out. The TCN-D2GAN model was compared with several mainstream generative adversarial network (GAN) models. Results indicated that the TCN-D2GAN model exhibited better performance in terms of signal classification accuracy and training efficiency. Finally, disassembly experiments were presented to validate the applicability of the system.

Automatic detection of epileptic seizure using machine learning-based IANFIS-LightGBM system

A sudden increase in electrical activity in the brain is a defining feature of one of the severe neurological diseases known as epilepsy. This abnormality appears as a seizure, and identifying seizures is an important field of research. An essential technique for examining the features of neurological issues brain activities, and epileptic seizures is electroencephalography (EEG). In EEG data, analyzing epileptic irregularities visually requires a lot of time from neurologists. For accurate detection of epileptic seizures, numerous scientific techniques have been used with EEG data, and most of these techniques have produced promising results. For EEG signal classification with a high classification accuracy rate, the present research proposes an enhanced machine learning-based epileptic seizure detection model. The present research provides a hybrid Improved Adaptive Neuro-Fuzzy Inference System (IANFIS)-Light Gradient Boosting Machine (LightGBM) technique for automatically detecting and diagnosing epilepsy from EEG data. The experimental findings were supported by EEG records made available by the German University of Bonn and scalp EEG data acquired at Children’s Hospital Boston. The suggested IANFIS-LightGBM, according to the results, offers the most significant classification accuracy ratings in both situations.

A twin convolutional neural network with hybrid binary optimizer for multimodal breast cancer digital image classification

There is a wide application of deep learning technique to unimodal medical image analysis with significant classification accuracy performance observed. However, real-world diagnosis of some chronic diseases such as breast cancer often require multimodal data streams with different modalities of visual and textual content. Mammography, magnetic resonance imaging (MRI) and image-guided breast biopsy represent a few of multimodal visual streams considered by physicians in isolating cases of breast cancer. Unfortunately, most studies applying deep learning techniques to solving classification problems in digital breast images have often narrowed their study to unimodal samples. This is understood considering the challenging nature of multimodal image abnormality classification where the fusion of high dimension heterogeneous features learned needs to be projected into a common representation space. This paper presents a novel deep learning approach combining a dual/twin convolutional neural network (TwinCNN) framework to address the challenge of breast cancer image classification from multi-modalities. First, modality-based feature learning was achieved by extracting both low and high levels features using the networks embedded with TwinCNN. Secondly, to address the notorious problem of high dimensionality associated with the extracted features, binary optimization method is adapted to effectively eliminate non-discriminant features in the search space. Furthermore, a novel method for feature fusion is applied to computationally leverage the ground-truth and predicted labels for each sample to enable multimodality classification. To evaluate the proposed method, digital mammography images and digital histopathology breast biopsy samples from benchmark datasets namely MIAS and BreakHis respectively. Experimental results obtained showed that the classification accuracy and area under the curve (AUC) for the single modalities yielded 0.755 and 0.861871 for histology, and 0.791 and 0.638 for mammography. Furthermore, the study investigated classification accuracy resulting from the fused feature method, and the result obtained showed that 0.977, 0.913, and 0.667 for histology, mammography, and multimodality respectively. The findings from the study confirmed that multimodal image classification based on combination of image features and predicted label improves performance. In addition, the contribution of the study shows that feature dimensionality reduction based on binary optimizer supports the elimination of non-discriminant features capable of bottle-necking the classifier.

Model and Data Dual-Driven Double-Point Observation Network for Ultra-Short MI EEG Classification.

Although deep networks have succeeded in various signal classification tasks, the time sequence samples used to train the deep models are usually required to reach a certain length. Especially, in brain computer interface (BCI) research, around 3.5s-long motor imagery (MI) Electroencephalography (EEG) samples are needed to obtain satisfactory classification performance. This time-span requirement of the training samples makes real-time MI BCI systems impossible to implement based on deep networks, which restricts the related researches within laboratory and makes practical application hard to accomplish. To address this issue, a double-point observation deep network (DoNet) is developed to classify ultra-short samples buried in noise. First, an analytical solution is developed theoretically to perform ultra-short signal classification based on double-point couples. Then, a signal-noise model is constructed to study the interference of noise on classification based on double-point couples. Based on which, an independent identical distribution condition is utilized to improve the classification accuracy in a data-driven manner. Combining the theoretical model and data-driven mechanism, DoNet can construct a steady data-distribution for the double-point couples of the samples with the same label. Therefore, the conditional probability of each double-point couple of a test sample can be obtained. With a voting strategy, the samples can be accurately classified by fusing these conditional probabilities. Meanwhile, the noise interference can be suppressed. DoNet has been evaluated on two public EEG datasets. Compared to most state-of-the-art methods, the 1s-long EEG signal classification accuracy has been improved by more than 3%.

An Energy-Efficient ECG Processor with Ultra-Low-Parameter Multi-Stage Neural Network and Optimized Power-of-Two Quantization.

This work presents an energy-efficient ECG processor designed for Cardiac Arrhythmia Classification. The processor integrates a pre-processing and neural network accelerator, achieved through algorithm-hardware co-design to optimize hardware resources. We propose a lightweight two-stage neural network architecture, where the first stage includes discrete wavelet transformation and an ultra-low-parameter multilayer perceptron (MLP) network, and the second stage utilizes group convolution and channel shuffle. Both stages leverage neural networks for hardware resource reuse and feature a reconfigurable processing element array and memory blocks adapted to the proposed two-stage structure to efficiently handle various convolution and MLP layers operations in the two-stage network. Additionally, an optimized power-of-two (OPOT) quantization technique is proposed to enhance accuracy in low-bit quantization, and a multiplier-less processing element structure tailored for the OPOT weight quantization is introduced. The ECG processor was implemented on a 65nm CMOS process technology with 4KB of SRAM memory, achieving an energy consumption per interference of 0.15 uJ with a power supply of 1V, 64% energy saving compared to the recent state-of-the-art work. Under 4-bit weight precision, the 5-class ECG signal classification accuracy reached 98.59% on the MIT-BIH arrhythmia dataset.

Temporal Electroencephalography Traits Dissociating Tactile Information and Cross-Modal Congruence Effects.

To explore whether temporal electroencephalography (EEG) traits can dissociate the physical properties of touching objects and the congruence effects of cross-modal stimuli, we applied a machine learning approach to two major temporal domain EEG traits, event-related potential (ERP) and somatosensory evoked potential (SEP), for each anatomical brain region. During a task in which participants had to identify one of two material surfaces as a tactile stimulus, a photo image that matched ('congruent') or mismatched ('incongruent') the material they were touching was given as a visual stimulus. Electrical stimulation was applied to the median nerve of the right wrist to evoke SEP while the participants touched the material. The classification accuracies using ERP extracted in reference to the tactile/visual stimulus onsets were significantly higher than chance levels in several regions in both congruent and incongruent conditions, whereas SEP extracted in reference to the electrical stimulus onsets resulted in no significant classification accuracies. Further analysis based on current source signals estimated using EEG revealed brain regions showing significant accuracy across conditions, suggesting that tactile-based object recognition information is encoded in the temporal domain EEG trait and broader brain regions, including the premotor, parietal, and somatosensory areas.

Classification of hyper-scale multimodal imaging datasets

Algorithms that classify hyper-scale multi-modal datasets, comprising of millions of images, into constituent modality types can help researchers quickly retrieve and classify diagnostic imaging data, accelerating clinical outcomes. This research aims to demonstrate that a deep neural network that is trained on a hyper-scale dataset (4.5 million images) composed of heterogeneous multi-modal data can be used to obtain significant modality classification accuracy (96%). By combining 102 medical imaging datasets, a dataset of 4.5 million images was created. A ResNet-50, ResNet-18, and VGG16 were trained to classify these images by the imaging modality used to capture them (Computed Tomography (CT), Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), and X-ray) across many body locations. The classification accuracy of the models was then tested on unseen data. The best performing model achieved classification accuracy of 96% on unseen data, which is on-par, or exceeds the accuracy of more complex implementations using EfficientNets or Vision Transformers (ViTs). The model achieved a balanced accuracy of 86%. This research shows it is possible to train Deep Learning (DL) Convolutional Neural Networks (CNNs) with hyper-scale multimodal datasets, composed of millions of images. Such models can find use in real-world applications with volumes of image data in the hyper-scale range, such as medical imaging repositories, or national healthcare institutions. Further research can expand this classification capability to include 3D-scans.

Robust CNN architecture for classification of reach and grasp actions from neural correlates: an edge device perspective

Brain–computer interfaces (BCIs) systems traditionally use machine learning (ML) algorithms that require extensive signal processing and feature extraction. Deep learning (DL)-based convolutional neural networks (CNNs) recently achieved state-of-the-art electroencephalogram (EEG) signal classification accuracy. CNN models are complex and computationally intensive, making them difficult to port to edge devices for mobile and efficient BCI systems. For addressing the problem, a lightweight CNN architecture for efficient EEG signal classification is proposed. In the proposed model, a combination of a convolution layer for spatial feature extraction from the signal and a separable convolution layer to extract spatial features from each channel. For evaluation, the performance of the proposed model along with the other three models from the literature referred to as EEGNet, DeepConvNet, and EffNet on two different embedded devices, the Nvidia Jetson Xavier NX and Jetson Nano. The results of the Multivariant 2-way ANOVA (MANOVA) show a significant difference between the accuracies of ML and the proposed model. In a comparison of DL models, the proposed models, EEGNet, DeepConvNet, and EffNet, achieved 92.44 ± 4.30, 90.76 ± 4.06, 92.89 ± 4.23, and 81.69 ± 4.22 average accuracy with standard deviation, respectively. In terms of inference time, the proposed model performs better as compared to other models on both the Nvidia Jetson Xavier NX and Jetson Nano, achieving 1.9 sec and 16.1 sec, respectively. In the case of power consumption, the proposed model shows significant values on MANOVA (p < 0.05) on Jetson Nano and Xavier. Results show that the proposed model provides improved classification results with less power consumption and inference time on embedded platforms.

Brief Visual Deprivation Effects on Brain Oscillations During Kinesthetic and Visual-motor Imagery.

It is widely recognized that opening and closing the eyes can direct attention to external or internal stimuli processing. This has been confirmed by studies showing the effects of changes in visual stimulation changes on cerebral activity during different tasks, e.g., motor imagery and execution. However, an essential aspect of creating a mental representation of motion, such as imagery perspective, has not yet been investigated in the present context. Our study aimed to verify the effect of brief visual deprivation (under eyes open [EO] and eyes closed [EC] conditions) on brain wave oscillations and behavioral performance during kinesthetic imagery (KMI) and visual-motor imagery (VMI) tasks. We focused on the alpha and beta rhythms from visual- and motor-related EEG activity sources. Additionally, we used machine learning algorithms to establish whether the registered differences in brain oscillations might affect motor imagery brain-computer interface (MI-BCI) performance. The results showed that the occipital areas in the EC condition presented significantly stronger desynchronization during VMI tasks, which is typical for enhanced visual stimuli processing. Furthermore, the stronger desynchronization of alpha rhythms from motor areas in the EO, than EC condition confirmed previous effects obtained during real movements. It was also found that simulating movement under EC/EO conditions affected signal classification accuracy, which has practical implications for MI-BCI effectiveness. These findings suggest that shifting processing toward external or internal stimuli modulates brain rhythm oscillations associated with different perspectives on the mental representation of movement.

A Novel Acoustic Evaluation Method for the Diagnosis of Adductor Spasmodic Dysphonia

Introduction: Adductor spasmodic dysphonia (ADSD) is characterized by involuntary laryngeal muscle spasms. Due to the lack of a quantitative evaluation method, most measurements have demonstrated difficulty in validity and reliability for diagnosing ADSD. This study aimed to establish a novel indicator for ADSD and determine its diagnostic effects. Methods: We investigated 98 voice samples from 49 patients with ADSD and 49 healthy participants. A sustained vowel was recorded by a high-definition audio recorder. Voice samples underwent regular acoustic evaluation and a novel global dimension (GD) method. GD, Jitter, Shimmer, HNR, Frequency shift, and CPPS were measured for both groups. Results: Statistical analysis revealed that the GD method effectively differentiated ADSD patients from healthy participants (p < 0.001, D′>0.8). Subsequent multiclass receiver operating characteristic analysis demonstrated that GD possessed the most significant classification accuracy (area under the curve = 0.988) compared with other acoustic parameters. Conclusion: GD was an effective metric for objective differentiation between ADSD patients and healthy participants. This metric could assist clinicians in the diagnosis of ADSD patients.

Automated Computer-Aided Detection and Classification of Intracranial Hemorrhage Using Ensemble Deep Learning Techniques.

Diagnosing Intracranial Hemorrhage (ICH) at an early stage is difficult since it affects the blood vessels in the brain, often resulting in death. We propose an ensemble of Convolutional Neural Networks (CNNs) combining Squeeze and Excitation-based Residual Networks with the next dimension (SE-ResNeXT) and Long Short-Term Memory (LSTM) Networks in order to address this issue. This research work primarily used data from the Radiological Society of North America (RSNA) brain CT hemorrhage challenge dataset and the CQ500 dataset. Preprocessing and data augmentation are performed using the windowing technique in the proposed work. The ICH is then classified using ensembled CNN techniques after being preprocessed, followed by feature extraction in an automatic manner. ICH is classified into the following five types: epidural, intraventricular, subarachnoid, intra-parenchymal, and subdural. A gradient-weighted Class Activation Mapping method (Grad-CAM) is used for identifying the region of interest in an ICH image. A number of performance measures are used to compare the experimental results with various state-of-the-art algorithms. By achieving 99.79% accuracy with an F-score of 0.97, the proposed model proved its efficacy in detecting ICH compared to other deep learning models. The proposed ensembled model can classify epidural, intraventricular, subarachnoid, intra-parenchymal, and subdural hemorrhages with an accuracy of 99.89%, 99.65%, 98%, 99.75%, and 99.88%. Simulation results indicate that the suggested approach can categorize a variety of intracranial bleeding types. By implementing the ensemble deep learning technique using the SE-ResNeXT and LSTM models, we achieved significant classification accuracy and AUC scores.

AN EMD AND IMF ENERGY ENTROPY-BASED OPTIMIZED FEATURE EXTRACTION AND CLASSIFICATION SCHEME FOR SINGLE TRIAL EEG SIGNAL

For the purpose of improving the classification accuracy of single-trial electroencephalogram (EEG) signal during motor imagery (MI) process, this study proposed a classification method which combines intrinsic mode functions (IMFs) energy entropy and improved empirical mode decomposition (EMD) scheme. Singular value decomposition (SVD), Gaussian mixture model (GMM), EMD and IMF energy entropy were employed for the newly designed scheme. After removing noise and artifacts from acquired EEG signals in EEGLAB, SVD was applied, and the singular values were clustered by GMM. The insignificant characteristics indicated by the small SVD values were then removed, and the signals were reconstructed, feeding to EMD algorithm. Those IMFs mapping to [Formula: see text], [Formula: see text], [Formula: see text] and [Formula: see text] frequencies were selected as the major features of the Electroencephalogram signal. The SVM classifier with Radial Basis Function Neural Network, linear, and polynomial kernel functions and voting mechanism then kicked in for classification. The results were compared with that of the traditional EMD and EEMD through simulation, showing that the proposed scheme can eliminate mode mixing effectively and improve the single-trial EEG signal classification accuracy significantly, suggesting the probability of designing a more efficient EEG control system based on the proposed scheme.

DiRaC-I: Identifying Diverse and Rare Training Classes for Zero-Shot Learning

Zero-Shot Learning (ZSL) is an extreme form of transfer learning that aims at learning from a few “seen classes” to have an understanding about the “unseen classes” in the wild. Given a dataset in ZSL research, most existing works use a predetermined, disjoint set of seen-unseen classes to evaluate their methods. These seen (training) classes might be sub-optimal for ZSL methods to appreciate the diversity and rarity of an object domain. Inspired by strategies like active learning, it is intuitive that intelligently selecting the training classes can improve ZSL performance. In this work, we propose a framework called Diverse and Rare Class Identifier (DiRaC-I) which, given an attribute-based dataset, can intelligently yield the most suitable “seen classes” for training ZSL models. DiRaC-I has two main goals – constructing a diversified set of seed classes, and using them to initialize a visual-semantic mining algorithm for acquiring the classes capturing both diversity and rarity in the object domain adequately. These classes can then be used as “seen classes” to train ZSL models for image classification. We simulate a real-world scenario where visual samples of novel object classes in the wild are available to neither DiRaC-I nor the ZSL models during training and conducted extensive experiments on two benchmark data sets for zero-shot image classification — CUB and SUN. Our results demonstrate DiRaC-I helps ZSL models to achieve significant classification accuracy improvements – specifically, up to 8% for CUB and up to 5% for SUN dataset. Additionally, while recognizing classes exhibiting rare attributes we also observe a performance boost for ZSL models, which is up to 10% and 7% for CUB and SUN datasets, respectively.

Statistical feature-based EEG signals classification using ANN and SVM classifiers for Parkinson’s disease detection

Parkinson's disease (PD) is a neurological disorder which is progressive in nature. Although there is no cure to this disease, symptomatic treatments are available. These treatments can slow the progressive development of the symptoms. Medications can treat some of the symptoms of the PD up to a great extent that in turn may help the patients to live a normal life. Besides these medications, the patients can also be provided with various therapies based on the types of their symptoms. But for providing any treatment, detection of its symptoms at an early stage is very important. This can reduce its future complexities considerably. Early diagnosis along with proper medications may treat the symptoms of PD significantly. This motivates to propose a new and effective methodology for detection and analysis of PD. In this work, an approach has been proposed for identification of PD patients by using Electroencephalogram (EEG) signals. Here, the EEG signals of normal persons and PD patients are processed in three stages. First, the raw EEG signals are pre-processed for removal of noises and artefacts present. Out of various techniques, Wavelet transform is used for this purpose. In the MATLAB environment, de-noising can be executed by using the in-built functions. Performances of the de-noising techniques are examined with the performance parameters namely Root Mean Square Error (RMSE) as well as Signal to Noise Ratio (SNR). In the second stage, statistical features are extracted from the pre-processed EEG signals. In this work, five statistical features are considered for performing the classification. In the final stage, the extracted features are classified using Artificial Neural Network (ANN) and Support Vector Machine (SVM) techniques. ANN is an efficient classifier that predicts the human brain's working manners. SVM on the other hand has been proven as one of the most prevailing classification algorithms that gives highly accurate and robust results. This is a novel approach of analyzing the performances of the classification techniques by evaluating the best performing feature. In both the classifiers accuracy, precision, sensitivity and specificity are calculated from the confusion matrix evaluated from the values of the statistical features. In ANN, results using six different training algorithms at different hidden layers are calculated and compared. This proves the training algorithm Levenberg-Marquardt back-propagation with hidden layer 20 as the best combination for performing the classification. From the results it is seen that both ANN and SVM classifiers provide significant classification accuracies of 94.7% and 96.5% respectively. Amongst the five considered features, Mean performs the best in terms of classification accuracy.

Multi-domain-fusion deep learning for automatic modulation recognition in spatial cognitive radio

Automatic modulation recognition (AMR) is a critical technology in spatial cognitive radio (SCR), and building high-performance AMR model can achieve high classification accuracy of signals. AMR is a classification problem essentially, and deep learning has achieved excellent performance in various classification tasks. In recent years, joint recognition of multiple networks has become increasingly popular. In complex wireless environments, there are multiple signal types and diversity of characteristics between different signals. Also, the existence of multiple interference in wireless environment makes the signal characteristics more complex. It is difficult for a single network to accurately extract the unique features of all signals and achieve accurate classification. So, this article proposes a time–frequency domain joint recognition model that combines two deep learning networks (DLNs), to achieve higher accuracy AMR. A DLN named MCLDNN (multi-channel convolutional long short-term deep neural network) is trained on samples composed of in-phase and quadrature component (IQ) signals, to distinguish modulation modes that are relatively easy to identify. This paper proposes a BiGRU3 (three-layer bidirectional gated recurrent unit) network based on FFT as the second DLN. For signals with significant similarity in the time domain and significant differences in the frequency domain that are difficult to distinguish by the former DLN, such as AM-DSB and WBFM, FFT (Fast Fourier Transform) is used to obtain frequency domain amplitude and phase (FDAP) information. Experiments have shown that the BiGUR3 network has superior extraction performance for amplitude spectrum and phase spectrum features. Experiments are conducted on two publicly available datasets, the RML2016.10a and RML2016.10b, and the results show that the overall recognition accuracy of the proposed joint model reaches 94.94% and 96.69%, respectively. Compared to a single network, the recognition accuracy is significantly improved. At the same time, the recognition accuracy of AM-DSB and WBFM signals has been improved by 17% and 18.2%, respectively.

An adaptive cross-class transfer learning framework with two-level alignment

Because of insufficient data, cross-class transfer learning has potential prospects in motor imagery electroencephalogram (MI-EEG) based rehabilitation engineering, and it has not been effectively addressed to reduce cross-class variability between source and target domains and find the best class-to-class transitive correspondence (CCTC). In this paper, we propose an adaptive cross-class transfer learning (ACTL) framework with two-level alignment (TLA) for MI decoding. By using the fast partitioning around medoids method, the central samples of each class are acquired from the source domain and target domain respectively. Then, they are applied to construct the central alignment matrix to realize the 1st-level domain alignment for any possible CCTC case. The 2nd-level subject alignment is performed by Euclidean alignment for each class of aligned source domain. Furthermore, the central samples of the target domain together with the two-level aligned source domain are mapped into tangent space, the resulting features are fed to a teacher convolutional neural network (TCNN), it and the transferred student CNN (SCNN) are trained successively, and their parameters of distillation loss are optimized automatically by a scaling-based grid search method. Experiments are conducted on a public MI-EEG dataset with multiple subjects and MI-tasks, the proposed framework can find the optimal SCNN associated with the best CCTC, which achieves a statistically significant average classification accuracy of 86.37%. The results suggest that TLA is helpful for increasing the distribution similarity between different domains, and the knowledge distillation embedded in ACTL framework greatly simplifies the SCNN and outperforms the complicated TCNN.

Sum of Gaussian Feature-Based Symbolic Representations of Eddy Current Defect Signatures

ABSTRACT This study investigates a novel symbolic representation method based on Symbolic Aggregate Approximation (SAX) of time series that focuses on differential coil (D-coil) eddy current (EC) defect responses. The method uses the Sum of Gaussian (SoG) approximation of the defect response, Fuzzy C-means (FCM) clustering, and the extrema values in the approximation to provide a reduced representation that can effectively analyze possible fault conditions in the defect response. Comparisons to existing SAX methods are performed with the new method indicating significant classification accuracy improvement.

Recognizing Command Words using Deep Recurrent Neural Network for Both Acoustic and Throat Speech

The importance of speech command recognition in a human-machine interaction system is increased in recent years. In this study, we propose a deep neural network-based system for acoustic and throat command speech recognition. We apply a preprocessed pipeline to create the input of the deep learning model. Firstly, speech commands are decomposed into components using well-known signal decomposition techniques. The Mel-frequency cepstral coefficients (MFCC) feature extraction method is applied to each component of the speech commands to obtain the feature inputs for the recognition system. At this stage, we apply and compare performance using different speech decomposition techniques such as wavelet packet decomposition (WPD), continuous wavelet transform (CWT), and empirical mode decomposition (EMD) in order to find out the best technique for our model. We observe that WPD shows the best performance in terms of classification accuracy. This paper investigates long short-term memory (LSTM)-based recurrent neural network (RNN), which is trained using the extracted MFCC features. The proposed neural network is trained and tested using acoustic speech commands. Moreover, we also train and test the proposed model using a throat mic. speech commands as well. Lastly, the transfer learning technique is employed to increase the test accuracy for throat speech recognition. The weights of the model train with the acoustic signal are used to initialize the model used for throat speech recognition. Overall, we have found significant classification accuracy for both acoustic and throat command speech. We obtain LSTM is much better than the GMM-HMM model, convolutional neural networks such as CNN-tpool2 and residual networks such as res15 and res26 with an accuracy score of over 97% on Google’s Speech Commands dataset and we achieve 95.35% accuracy on our throat speech data set using the transfer learning technique.

Prediction and Analysis of Heart Diseases Using Heterogeneous Computing Platform

The work in this paper helps study cardiac rhythms and the electrical activity of the heart for two of the most critical cardiac arrhythmias. Various consumer devices exist, but implementation of an appropriate device at a certain position on the body at a certain pressure point containing an enormous number of blood vessels and developing filtering techniques for the most accurate signal extraction from the heart is a challenging task. In this paper, we provide evidence of prediction and analysis of Atrial Fibrillation (AF) and Ventricular Fibrillation (VF). Long-term monitoring of diseases such as AF and VF occurrences is very important, as these will lead to occurrence of ischemic stroke, cardiac arrest and complete heart failure. The AF and VF signal classification accuracy are much higher when processed on a Graphics Processor Unit (GPU) than Central Processing Unit (CPU) or traditional Holter machines. The classifier COMMA-Z filter is applied to the highly-sensitive industry certified Bio PPG sensor placed at the earlobe and computed on GPU.