Data Mining - Concepts and Techniques.

Data Mining Practical Machine Learning Tools and Techniques

Context Electroencephalography (EEG) is a complex signal and can require several years of training, as well as advanced signal processing and feature extraction methodologies to be correctly interpreted. Recently, deep learning (DL) has shown great promise in helping make sense of EEG signals due to its capacity to learn good feature representations from raw data. Whether DL truly presents advantages as compared to more traditional EEG processing approaches, however, remains an open question. Objective In this work, we review 154 papers that apply DL to EEG, published between January 2010 and July 2018, and spanning different application domains such as epilepsy, sleep, brain-computer interfacing, and cognitive and affective monitoring. We extract trends and highlight interesting approaches from this large body of literature in order to inform future research and formulate recommendations. Methods Major databases spanning the fields of science and engineering were queried to identify relevant studies published in scientific journals, conferences, and electronic preprint repositories. Various data items were extracted for each study pertaining to (1) the data, (2) the preprocessing methodology, (3) the DL design choices, (4) the results, and (5) the reproducibility of the experiments. These items were then analyzed one by one to uncover trends. Results Our analysis reveals that the amount of EEG data used across studies varies from less than ten minutes to thousands of hours, while the number of samples seen during training by a network varies from a few dozens to several millions, depending on how epochs are extracted. Interestingly, we saw that more than half the studies used publicly available data and that there has also been a clear shift from intra-subject to inter-subject approaches over the last few years. About [Formula: see text] of the studies used convolutional neural networks (CNNs), while [Formula: see text] used recurrent neural networks (RNNs), most often with a total of 3-10 layers. Moreover, almost one-half of the studies trained their models on raw or preprocessed EEG time series. Finally, the median gain in accuracy of DL approaches over traditional baselines was [Formula: see text] across all relevant studies. More importantly, however, we noticed studies often suffer from poor reproducibility: a majority of papers would be hard or impossible to reproduce given the unavailability of their data and code. Significance To help the community progress and share work more effectively, we provide a list of recommendations for future studies and emphasize the need for more reproducible research. We also make our summary table of DL and EEG papers available and invite authors of published work to contribute to it directly. A planned follow-up to this work will be an online public benchmarking portal listing reproducible results.

https://iopscience.iop.org/article/10.1088/1741-2552/ab260c/pdf

Deep learning-based electroencephalography analysis: a systematic review.

Sleep stage classification constitutes an important preliminary exam in the diagnosis of sleep disorders. It is traditionally performed by a sleep expert who assigns to each 30 s of the signal of a sleep stage, based on the visual inspection of signals such as electroencephalograms (EEGs), electrooculograms (EOGs), electrocardiograms, and electromyograms (EMGs). We introduce here the first deep learning approach for sleep stage classification that learns end-to-end without computing spectrograms or extracting handcrafted features, that exploits all multivariate and multimodal polysomnography (PSG) signals (EEG, EMG, and EOG), and that can exploit the temporal context of each 30-s window of data. For each modality, the first layer learns linear spatial filters that exploit the array of sensors to increase the signal-to-noise ratio, and the last layer feeds the learnt representation to a softmax classifier. Our model is compared to alternative automatic approaches based on convolutional networks or decisions trees. Results obtained on 61 publicly available PSG records with up to 20 EEG channels demonstrate that our network architecture yields the state-of-the-art performance. Our study reveals a number of insights on the spatiotemporal distribution of the signal of interest: a good tradeoff for optimal classification performance measured with balanced accuracy is to use 6 EEG with 2 EOG (left and right) and 3 EMG chin channels. Also exploiting 1 min of data before and after each data segment offers the strongest improvement when a limited number of channels are available. As sleep experts, our system exploits the multivariate and multimodal nature of PSG signals in order to deliver the state-of-the-art classification performance with a small computational cost.

https://hal.archives-ouvertes.fr/hal-01810436/document

A Deep Learning Architecture for Temporal Sleep Stage Classification Using Multivariate and Multimodal Time Series

Automatic sleep staging has been often treated as a simple classification problem that aims at determining the label of individual target polysomnography epochs one at a time. In this paper, we tackle the task as a sequence-to-sequence classification problem that receives a sequence of multiple epochs as input and classifies all of their labels at once. For this purpose, we propose a hierarchical recurrent neural network named SeqSleepNet (source code is available at http://github.com/pquochuy/SeqSleepNet ). At the epoch processing level, the network consists of a filterbank layer tailored to learn frequency-domain filters for preprocessing and an attention-based recurrent layer designed for short-term sequential modeling. At the sequence processing level, a recurrent layer placed on top of the learned epoch-wise features for long-term modeling of sequential epochs. The classification is then carried out on the output vectors at every time step of the top recurrent layer to produce the sequence of output labels. Despite being hierarchical, we present a strategy to train the network in an end-to-end fashion. We show that the proposed network outperforms the state-of-the-art approaches, achieving an overall accuracy, macro F1-score, and Cohen’s kappa of 87.1%, 83.3%, and 0.815 on a publicly available dataset with 200 subjects.

/pdf/seqsleepnet-end-to-end-hierarchical-recurrent-neural-network-2x1la2lrui.pdf

SeqSleepNet: End-to-End Hierarchical Recurrent Neural Network for Sequence-to-Sequence Automatic Sleep Staging

Sleep specialists often conduct manual sleep stage scoring by visually inspecting the patient’s neurophysiological signals collected at sleep labs. This is, generally, a very difficult, tedious and time-consuming task. The limitations of manual sleep stage scoring have escalated the demand for developing Automatic Sleep Stage Classification (ASSC) systems. Sleep stage classification refers to identifying the various stages of sleep and is a critical step in an effort to assist physicians in the diagnosis and treatment of related sleep disorders. The aim of this paper is to survey the progress and challenges in various existing Electroencephalogram (EEG) signal-based methods used for sleep stage identification at each phase; including pre-processing, feature extraction and classification; in an attempt to find the research gaps and possibly introduce a reasonable solution. Many of the prior and current related studies use multiple EEG channels, and are based on 30 s or 20 s epoch lengths which affect the feasibility and speed of ASSC for real-time applications. Thus, in this paper, we also present a novel and efficient technique that can be implemented in an embedded hardware device to identify sleep stages using new statistical features applied to 10 s epochs of single-channel EEG signals. In this study, the PhysioNet Sleep European Data Format (EDF) Database was used. The proposed methodology achieves an average classification sensitivity, specificity and accuracy of 89.06%, 98.61% and 93.13%, respectively, when the decision tree classifier is applied. Finally, our new method is compared with those in recently published studies, which reiterates the high classification accuracy performance.

Sleep Stage Classification Using EEG Signal Analysis: A Comprehensive Survey and New Investigation

Currently, sleep disorders are considered as one of the major human life issues. There are several stable physiological stages that the human brain goes through during sleep. Nowadays, many biomedical signals such as EEG, ECG, EMG, and EOG offer useful details for clinical setups that are used in identifying sleep disorders. In this work, we propose an efficient technique that could be implemented in hardware to differentiate sleep stages which will assist physicians in the diagnosis and treatment of related sleep disorders. This study depends on different EEG datasets from PhysioNet using the Sleep-EDF [Expanded] Database that were acquired and described by scientists for the analysis and diagnosis of sleep stages. Generally, the EEG signal is decomposed into five bands: delta, theta, alpha, beta, and gamma to define the change in brain state. In this work, Butterworth band-pass filters are designed to filter and decompose EEG into the above frequency sub-bands. In addition, various discriminating features including energy, standard deviation and entropy are computed and extracted from each δ, □, α, β and γ sub-band. Furthermore, the extracted features are then fed to a supervised learning classifier; support vector machine (SVM) to be able to recognize the sleep stages state and identify if the acquired signal is corresponding to wake or stage 1 of sleep, according to the purpose of this research. The key novelty of this work is to identify the sleep stages from a publicly available EEG signal dataset by using a feasible set of features, easily implementable filters in any microcontroller device, and an efficient classification method. Therefore, physicians can track these sleep stages to identify certain patterns such as detecting fatigue, drowsiness, and/or various sleep disorders such as sleep apnea. The experimental results on a variety of subjects verify 92.5% of classification accuracy of the proposed work.

/pdf/efficient-sleep-stage-classification-based-on-eeg-signals-4shqrm6vtf.pdf

Efficient sleep stage classification based on EEG signals

Nowadays, analyzing EEG signals has made it easy to diagnose many sleep-related breathing disorders such as Obstructive Sleep Apnea (OSA), which is a potentially serious sleep disorder that affects the quality of human life This paper introduces an efficient methodology that could be implemented in hardware to differentiate OSA patients from normal controls, based on the Electroencephalogram (EEG) signals For this purpose, first, the EEG recorded datasets that were obtained from the Phsyionet website are filtered and decomposed into delta, theta alpha, beta and gamma sub-bands using Infinite Impulse Response (IIR) Butterworth band-pass filters Second, descriptive features such as energy and variance are extracted from each frequency band that are used as input parameters for classification Finally, several machine learning algorithms including Support Vector Machines (SVM), Artificial Neural Networks (ANN), Linear Discriminant Analysis (LDA) and Naive Bayes (NB) are employed in order to identify if the OSA exists or not, according to the objective of this study The results that are obtained from these classifiers are then compared in terms of accuracy, sensitivity and specificity The experimental results show that the SVM attained the best classification accuracy of 9714% as compared to the others

Efficient obstructive sleep apnea classification based on EEG signals

In recent years, the estimation of human sleep disorders from Electroencephalogram (EEG) signals have played an important role in developing automatic detection of sleep stages. A few methods exist in the market presently towards this aim. However, sleep physicians may not have full assurance and consideration in such methods due to concerns associated with systems accuracy, sensitivity and specificity. This paper presents a novel and efficient technique that can be implemented in a microcontroller device to identify sleep stages in an effort to assist physicians in the diagnosis and treatment of related sleep disorders by enhancing the accuracy of the developed algorithm using a single channel of EEG signals. First, the dataset of EEG signal is filtered and decomposed into delta, theta, alpha, beta and gamma subbands using Butterworth band-pass filters. Second, a set of sample statistical discriminating features are derived from each frequency band. Finally, sleep stages consisting of Wakefulness, Rapid Eye Movement (REM) and Non-Rapid Eye Movement (NREM) are classified using several supervised machine learning classifiers including multi-class Support Vector Machines (SVM), Decision Trees (DT), Neural Networks (NN), K-Nearest Neighbors (KNN) and Naive Bayes (NB). This paper combines REM with Stage 1 NREM due to data similarities. Performance is then compared based on single channel EEG signals that were obtained from 20 healthy subjects. The results show that the proposed technique using DT classifier efficiently achieves high accuracy of 97.30% in differentiating sleeps stages. Also, a comparison of our method with some recent available works in the literature reiterates the high classification accuracy performance.

A comparison of different machine learning algorithms using single channel EEG signal for classifying human sleep stages

In this work, an efficient digital system is designed using hardware to filter the Electrocardiogram (ECG) signal and to detect the QRS complex (beats). The system implementation has been done by using a Field Programmable Gate Array (FPGA). In the first phase of the hardware system implementation, Finite Impulse Response (FIR) filters are designed for preprocessing and denoising the ECG signal. The filtered signal is then used as the input of the second phase of the hardware implementation to detect and classify the ECG beats. The entire system has been implemented on ALTERA DE II FPGA by designing synthesizable finite state machines. Quartus II tool has been used to simulate and test the system. The designed system has been tested on ECG waves from the MIT-BIH Arrhythmia database by windowing the signal and applying adaptive signal and noise thresholds in each window of processing. The hardware system has achieved an overall accuracy of 98% in the beat detection phase, while providing the detected beats and the classification of irregular heart beat rates in real time.

Khald A. I. Aboalayon

Papers

Sleep Stage Classification Using EEG Signal Analysis: A Comprehensive Survey and New Investigation

Efficient sleep stage classification based on EEG signals

Efficient obstructive sleep apnea classification based on EEG signals

A comparison of different machine learning algorithms using single channel EEG signal for classifying human sleep stages

FPGA-based denoising and beat detection of the ECG signal