MCI Detection Using Kernel Eigen-Relative-Power Features of EEG Signals

Classification between individuals with mild cognitive impairment (MCI) and healthy controls (HC) based on electroencephalography (EEG) has been considered a challenging task to be addressed for the purpose of its early detection. In this study, we proposed a novel EEG feature, the kernel eigen-relative-power (KERP) feature, for achieving high classification accuracy of MCI versus HC. First, we introduced the relative powers (RPs) between pairs of electrodes across 21 different subbands of 2-Hz width as the features, which have not yet been used in previous MCI-HC classification studies. Next, the Fisher’s class separability criterion was applied to determine the best electrode pairs (five electrodes) as well as the frequency subbands for extracting the most sensitive RP features. The kernel principal component analysis (kernel PCA) algorithm was further performed to extract a few more discriminating nonlinear principal components from the optimal RPs, and these components form a KERP feature vector. Results carried out on 51 participants (24 MCI and 27 HC) show that the newly introduced subband RP feature showed superior classification performance to commonly used spectral power features, including the band power, single-electrode relative power, and also the RP based on the conventional frequency bands. A high leave-one-participant-out cross-validation (LOPO-CV) classification accuracy 86.27% was achieved by the RP feature, using a simple linear discriminant analysis (LDA) classifier. Moreover, with the same classifier, the proposed KERP further improved the accuracy to 88.24%. Finally, cascading the KERP feature to a nonlinear classifier, the support vector machine (SVM), yields a high MCI-HC classification accuracy of 90.20% (sensitivity = 87.50% and specificity = 92.59%). The proposed method demonstrated a high accuracy and a high usability (only five electrodes are required), and therefore, has great potential to further develop an EEG-based computer-aided diagnosis system that can be applied for the early detection of MCI.

Neonatal Seizure Detection Using Deep Convolutional Neural Networks

Identifying a core set of features is one of the most important steps in the development of an automated seizure detector. In most of the published studies describing features and seizure classifiers, the features were hand-engineered, which may not be optimal. The main goal of the present paper is using deep convolutional neural networks (CNNs) and random forest to automatically optimize feature selection and classification. The input of the proposed classifier is raw multi-channel EEG and the output is the class label: seizure/nonseizure. By training this network, the required features are optimized, while fitting a nonlinear classifier on the features. After training the network with EEG recordings of 26 neonates, five end layers performing the classification were replaced with a random forest classifier in order to improve the performance. This resulted in a false alarm rate of 0.9 per hour and seizure detection rate of 77% using a test set of EEG recordings of 22 neonates that also included dubious seizures. The newly proposed CNN classifier outperformed three data-driven feature-based approaches and performed similar to a previously developed heuristic method.

A Topical Word Embeddings for Text Classification

This paper presents an approach that uses topic models based on LDA to represent documents in text categorization problems. The document representation is achieved through the cosine similarity between document embeddings and embeddings of topic words, creating a Bag-of-Topics (BoT) variant. The performance of this approach is compared against those of two other representations: BoW (Bag-of-Words) and Topic Model, both based on standard tf-idf. Also, to reveal the effect of the classifier, we compared the performance of the nonlinear classifier SVM against that of the linear classifier Naive Bayes, taken as baseline. To evaluate the approach we use two bases, one multi-label (RCV-1) and another single-label (20 Newsgroup). The model presents significant results with low dimensionality when compared to the state of the art.

A Deep Feature Learning Method for Drill Bits Monitoring Using the Spectral Analysis of the Acoustic Signals

Machine fault diagnosis (MFD) has gained an important enthusiasm since the unfolding of the pattern recognition techniques in the last three decades. It refers to all of the studies that aim to automatically detect the faults on the machines using various kinds of signals that they can generate. The present work proposes a MFD system for the drilling machines that is based on the sounds they produce. The first key contribution of this paper is to present a system specifically designed for the drills, by attempting not only to detect the faulty drills but also to detect whether the sounds were generated during the active or the idling stage of the whole machinery system, in order to provide a complete remote control. The second key contribution of the work is to represent the power spectrum of the sounds as images and apply some transformations on them in order to reveal, expose, and emphasize the health patterns that are hidden inside them. The created images, the so-called power spectrum density (PSD)-images, are then given to a deep convolutional autoencoder (DCAE) for a high-level feature extraction process. The final step of the scheme consists of adopting the proposed PSD-images + DCAE features as the final representation of the original sounds and utilize them as the inputs of a nonlinear classifier whose outputs will represent the final diagnosis decision. The results of the experiments demonstrate the high discrimination potential afforded by the proposed PSD-images + DCAE features. They were also tested on a noisy dataset and the results show their robustness against noises.

A SOLUTION FOR IMPROVEMENT OF ECG ARRHYTHMIA RECOGNITION USING RESPIRATION INFORMATION

Electrocardiogram (ECG) and respiration signals are two basic and important and valuable biomedical signals as source of information used to determine a person's health status. However, ECG signals are usually of small amplitude and are susceptible to various noises such as: the 50Hz grid noise, poor electrodes’ contacts with the patient's skin, the patient's emotional variations, the respiration and movement of the patient... The idea in this paper by filtering out the effect of the respiration in the ECG signal or by incorporating the information of breathing stage into the ECG signal classification the we can improve the reliability and accuracy of the arrythmia classification. This paper proposes a solution, which uses wavelet filter to reduce the effect of respiration in the ECG signals and will use additional information from the breathing rhythm (when available) to help better classifying the arrythmias. As the main nonlinear classifier we use the classical neuro-fuzzy TSK network. The proposed solution will be tested with data from the MIT-BIH and the MGH/MF databases.

TBSTM: A Novel and Fast Nonlinear Classification Method for Image Data

A new classifier for image data classification named as linear twin bounded support tensor machine (linear TBSTM) is proposed by adding regularization terms in objective functions, which results in the realization of structural risk minimization avoids of the singularity of matrices. We know that up to now nonlinear classifiers based on STM for image data classification are not seen more. In order to remedy this limitation, a new matrix kernel function is introduced and based on which the nonlinear version of TBSTM is studied with a detailed theoretical derivation, and then a nonlinear classifier called as nonlinear TBSTM is suggested. In order to examine the effectiveness of the proposed classifiers, a series of comparative experiments with three linear classifiers STM, TSTM and PSTM are performed on 15 binary image classification problems taken from ORL, YALE and AR datasets. Experiment results show that the proposed classifiers are effective and efficient.