Automatic sleep staging: A computer assisted approach for optimal combination of features and polysomnographic channels

To improve applicability of automatic sleep staging an efficient subject-independent method is proposed with application in sleep–wake detection and in multiclass sleep staging (awake, non-rapid eye movement (NREM) sleep and rapid eye movement (REM) sleep). In turn, NREM is further divided into three stages denoted here by N1, N2, and N3. To assess the method, polysomnographic (PSG) records of 40 patients from our ISRUC-Sleep dataset, which was scored by an expert clinician in the central hospital of Coimbra, are used. To find the best combination of PSG signals for automatic sleep staging, six electroencephalographic (EEG), two electrooculographic (EOG), and one electromyographic (EMG) channels are analyzed. An extensive set of feature extraction techniques are applied, covering temporal, frequency and time–frequency domains. The maximum overlap wavelet transform (MODWT), a shift invariant transform, was used to extract the features in time–frequency domain. The extracted feature set is transformed and normalized to reduce the effect of extreme values of features. The most discriminative features are selected through a two-step method composed by a manual selection step based on features’ histogram analysis followed by an automatic feature selector. The selected feature set is classified using support vector machines (SVMs). The system achieved the best performance by combining 6 channels (C3, C4, O1, left EOG (LOC), right EOG (ROC) and chin EMG (X1)) for sleep–wake detection, and 9 channels (C3, C4, O1, O2, F3, F4, LOC, ROC, X1) for multiclass sleep staging.     

Multi-class grasping classifiers using EEG data and a common spatial pattern filter

This paper describes the classification of various human actions from brain activity. In particular, we focus on grasping movements and estimate grasping patterns from electroencephalogram (EEG) data. EEG data is converted to grasping features by using a common spatial pattern filter (CSP filter), and the features are subsequently classified into grasping categories by using the k-nearest neighbor method. We tested the pipeline of feature extraction and classification on the EEG dataset. The EEG data were acquired while participants grasped an object according to the Cutkosky’s grasping taxonomy, in which grasping movements are categorized into nine power-type grasping patterns and seven precision-type grasping patterns. The best classification rate for 9-class power-type grasping patterns was 48% and for 7-class precision-type grasping patterns was 40%.     

Improving cervical cancer classification with imbalanced datasets combining taming transformers with T2T-ViT

Cervical cell classification has important clinical significance in cervical cancer screening at early stages. However, there are fewer public cervical cancer smear cell datasets, the weights of each classes’ samples are unbalanced, the image quality is uneven, and the classification research results based on CNN tend to overfit. To solve the above problems, we propose a cervical cell image generation model based on taming transformers (CCG-taming transformers) to provide high-quality cervical cancer datasets with sufficient samples and balanced weights, we improve the encoder structure by introducing SE-block and MultiRes-block to improve the ability to extract information from cervical cancer cells images; we introduce Layer Normlization to standardize the data, which is convenient for the subsequent non-linear processing of the data by the ReLU activation function in feed forward; we also introduce SMOTE-Tomek Links to balance the source data set and the number of samples and weights of the images we use Tokens-to-Token Vision Transformers (T2T-ViT) combing transfer learning to classify the cervical cancer smear cell image dataset to improve the classification performance. Classification experiments using the model proposed in this paper are performed on three public cervical cancer datasets, the classification accuracy in the liquid-based cytology Pap smear dataset (4-class), SIPAKMeD (5-class), and Herlev (7-class) are 98.79%, 99.58%, and 99.88%, respectively. The quality of the images we generated on these three data sets is very close to the source data set, the final averaged inception score (IS), Fréchet inception distance (FID), Recall and Precision are 3.75, 0.71, 0.32 and 0.65 respectively. Our method improves the accuracy of cervical cancer smear cell classification, provides more cervical cell sample images for cervical cancer-related research, and assists gynecologists to judge and diagnose different types of cervical cancer cells and analyze cervical cancer cells at different stages, which are difficult to distinguish. This paper applies the transformer to the generation and recognition of cervical cancer cell images for the first time.

     

基于双向LSTM卷积网络与注意力机制的自动睡眠分期模型

针对现阶段深度睡眠分期模型存在的梯度消失、对时序信息学习能力较弱等问题,提出一种基于双向长短时记忆卷积网络与注意力机制的自动睡眠分期模型。将少样本类别的睡眠脑电数据通过过采样方式进行数据增强后,利用带残差块的卷积神经网络学习数据特征表示,再通过带注意力层的双向长短时记忆网络挖掘深层时序信息,使用Softmax层实现睡眠分期的自动判别。实验使用Sleep-EDF数据集中19晚单通道脑电信号对模型进行交叉验证,取得了较高的分类准确率和宏平均F₁值,优于对比方法。该方法能够有效缓解睡眠分期判别中少数类分类性能较低的问题,并提高了深度睡眠分期模型的整体分类性能。     

基于CNN-BiLSTM的自动睡眠分期算法

睡眠分期是睡眠数据分析的基础,针对目前睡眠分期存在的依赖人工提取、人工判别效率低、自动睡眠分期准确率不高等问题,本文研究模型是基于卷积神经网络和双向长短时记忆神经网络2个深度学习神经网络相结合的,利用脑电信号来进行自动睡眠分期的模型方法.算法能提取得到原始脑电信号的梅尔频谱,利用卷积神经网络和双向长短时记忆神经网络进行...     

Detecting rapid eye movement sleep using a single EEG signal channel

Sleep stage scoring is generally determined in a polysomnographic (PSG) study where technologists use electroencephalogram (EEG), electromyogram (EMG), and electrooculogram (EOG) signals to determine the sleep stages. Such a process is time consuming and labor intensive. To reduce the workload and to improve the sleep stage scoring performance of sleep experts, this paper introduces an intelligent rapid eye movement (REM) sleep detection method that requires only a single EEG channel. The proposed approach distinguishes itself from previous automatic sleep staging methods by introducing two sets of auxiliary features to help resolve the difficulties caused by interpersonal EEG signal differences. In addition to adopting conventional time and frequency domain features, two empirical rules are introduced to enhance REM detection performance based on sleep being a continuous process. The approach was tested with 779,661 epochs obtained from 947 overnight PSG studies. The REM sleep detection results show a kappa coefficient at 0.752, an accuracy level of 0.930, a sensitivity score of 0.814, and a positive predictive value of 0.775. The results also show that the performance of the approach varies with the ratio of REM sleep and the severity of sleep apnea of the subjects. The experimental results also show that it is possible to improve the performance of an automatic sleep staging method by tailoring it to subgroups of persons that have similar sleep architecture and clinical characteristics.     

Automatic classification of sleep stages based on the time-frequency image of EEG signals

In this paper, a new method for automatic sleep stage classification based on time-frequency image (TFI) of electroencephalogram (EEG) signals is proposed. Automatic classification of sleep stages is an important part for diagnosis and treatment of sleep disorders. The smoothed pseudo Wigner–Ville distribution (SPWVD) based time-frequency representation (TFR) of EEG signal has been used to obtain the time-frequency image (TFI). The segmentation of TFI has been performed based on the frequency-bands of the rhythms of EEG signals. The features derived from the histogram of segmented TFI have been used as an input feature set to multiclass least squares support vector machines (MC-LS-SVM) together with the radial basis function (RBF), Mexican hat wavelet, and Morlet wavelet kernel functions for automatic classification of sleep stages from EEG signals. The experimental results are presented to show the effectiveness of the proposed method for classification of sleep stages from EEG signals.     

Automated Dataset Generation for Training Peer-to-Peer Machine Learning Classifiers

Peer-to-peer (P2P) classifications based on flow statistics have been proven accurate in detecting P2P traffic. A machine learning classification is affected by the quality and recency of the training dataset used. Hence, to classify P2P traffic on-line requires the removal of these limitations. In this paper, an automated training dataset generation for an on-line P2P traffic classification is proposed to allow frequent classifier retraining. A two-stage training dataset generator (TSTDG) is proposed by combining a 3-class heuristic and a 3-class statistical classification to automatically generate a training dataset. In the heuristic stage, traffic is classified as P2P, non-P2P, or unknown. In the statistical stage, a dual Decision Tree is built based on a dataset generated in the heuristic stage to reduce the amount of classified unknown traffic. The final training dataset is generated based on all flows that are classified in these two stages. The proposed system has been evaluated on traces captured from a campus network. The overall results show that the TSTDG can generate an accurate training dataset by classifying around 94 % of total flows with high accuracy (98.59 %) and a low false positive rate (1.27 %).     

Classification of sleep stages based on LSTAR model

Sleep study is very important in the health since sleep disorders affect the productivity of individuals. One of the important topics in sleep research is the classification of sleep stages using the electroencephalogram (EEG) signal. Electrical activities of brain are measured by EEG signal in the laboratory. In real-world environments, EEG signal is also used in portable monitoring devices to analyze sleep. In this study, we propose an efficient method for classification of sleep stages. EEG signals are examined by a new model from autoregressive (AR) family, namely logistic smooth transition autoregressive (LSTAR) to study sleep process. In contrast to the AR model, LSTAR is a non-linear one; therefore, it is suitable for modeling non-linear signals such as EEG. In the current research, at first, each 30-second epoch of EEG signal is decomposed into the time-frequency sub-bands using the double-density dual-tree discrete wavelet transform (D3TDWT). In the second step, LSTAR model is used for feature extraction from each sub-band. Next, the dimension of feature vector is reduced by tensor locality preserving projection (tensor LPP) method, and then the obtained features are given to classifier to determine the stage of each epoch based on the number of considered classes. After classifying sleep stages, some misclassified epochs can be corrected according to the smoothing rule. We consider different classifiers and evaluate their performance. The results indicate the efficiency of the proposed method in comparison with the recently introduced methods in terms of accuracy and Kappa coefficient.     

Adaptive neuro-fuzzy inference system employing wavelet coefficients for detection of alterations in sleep EEG activity during hypopnoea episodes

The Obstructive Sleep Apnoea Hypopnoea Syndrome (OSAH) means “cessation of breath” during the sleep hours and the sufferers often experience related changes in the electrical activity of the brain and heart. This paper describes the application of adaptive neuro-fuzzy inference system (ANFIS) model for automatic detection of alterations in the human electroencephalogram (EEG) activities during hypopnoea episodes. Decision making was performed in two stages: feature extraction by computation of wavelet coefficients and classification by the ANFIS trained with the backpropagation gradient descent method in combination with the least squares method. The EEG signals (pre and during hypopnoea) from three electrodes (C3, C4 and O2) were used as input patterns of the three ANFIS classifiers. To improve diagnostic accuracy, the fourth ANFIS classifier (combining ANFIS) was trained using the outputs of the three ANFIS classifiers as input data. The proposed ANFIS model combined the neural network adaptive capabilities and the fuzzy logic qualitative approach. Some conclusions concerning the saliency of features on detecting any possible changes in the human EEG activity due to hypopnoea (mild case of cessation of breath) occurrences were drawn through analysis of the ANFIS. The performance of the ANFIS model was evaluated in terms of training performance and classification accuracies and the results confirmed that the proposed ANFIS model has potential in detecting changes in the human EEG activity due to hypopnoea episodes.     

Complex networks approach for EEG signal sleep stages classification

Sleep stage scoring is a challenging task. Most of existing sleep stage classification approaches rely on analysing electroencephalography (EEG) signals in time or frequency domain. A novel technique for EEG sleep stages classification is proposed in this paper. The statistical features and the similarities of complex networks are used to classify single channel EEG signals into six sleep stages. Firstly, each EEG segment of 30 s is divided into 75 sub-segments, and then different statistical features are extracted from each sub-segment. In this paper, feature extraction is important to reduce dimensionality of EEG data and the processing time in classification stage. Secondly, each vector of the extracted features, which represents one EEG segment, is transferred into a complex network. Thirdly, the similarity properties of the complex networks are extracted and classified into one of the six sleep stages using a k-means classifier. For further investigation, in the statistical features extraction phase two statistical features sets are tested and ranked based on the performance of the complex networks. To investigate the classification ability of complex networks combined with k-means, the extracted statistical features were also forwarded to a k-means and a support vector machine (SVM) for comparison. We also compare the proposed method with other existing methods in the literature. The experimental results show that the proposed method attains better classification results and a reasonable execution time compared with the SVM, k-means and the other existing methods. The research results in this paper indicate that the proposed method can assist neurologists and sleep specialists in diagnosing and monitoring sleep disorders.     

Automatic cervical cell segmentation and classification in Pap smears

Cervical cancer is one of the leading causes of cancer death in females worldwide. The disease can be cured if the patient is diagnosed in the pre-cancerous lesion stage or earlier. A common physical examination technique widely used in the screening is Papanicolaou test or Pap test. In this research, a method for automatic cervical cancer cell segmentation and classification is proposed. A single-cell image is segmented into nucleus, cytoplasm, and background, using the fuzzy C-means (FCM) clustering technique. Four cell classes in the ERUDIT and LCH datasets, i.e., normal, low grade squamous intraepithelial lesion (LSIL), high grade squamous intraepithelial lesion (HSIL), and squamous cell carcinoma (SCC), are considered. The 2-class problem can be achieved by grouping the last 3 classes as one abnormal class. Whereas, the Herlev dataset consists of 7 cell classes, i.e., superficial squamous, intermediate squamous, columnar, mild dysplasia, moderate dysplasia, severe dysplasia, and carcinoma in situ. These 7 classes can also be grouped to form a 2-class problem. These 3 datasets were tested on 5 classifiers including Bayesian classifier, linear discriminant analysis (LDA), K-nearest neighbor (KNN), artificial neural networks (ANN), and support vector machine (SVM). For the ERUDIT dataset, ANN with 5 nucleus-based features yielded the accuracies of 96.20% and 97.83% on the 4-class and 2-class problems, respectively. For the Herlev dataset, ANN with 9 cell-based features yielded the accuracies of 93.78% and 99.27% for the 7-class and 2-class problems, respectively. For the LCH dataset, ANN with 9 cell-based features yielded the accuracies of 95.00% and 97.00% for the 4-class and 2-class problems, respectively. The segmentation and classification performances of the proposed method were compared with that of the hard C-means clustering and watershed technique. The results show that the proposed automatic approach yields very good performance and is better than its counterparts.     

Road intersection detection and classification using hierarchical SVM classifier

In this paper, a hierarchical multi-classification approach using support vector machines (SVM) has been proposed for road intersection detection and classification. Our method has two main steps. The first involves the road detection. For this purpose, an edge-based approach has been developed using the bird’s eye view image which is mapped from the perspective view of the road scene. Then, the concept of vertical spoke has been introduced for road boundary form extraction. The second step deals with the problem of road intersection detection and classification. It consists on building a hierarchical SVM classifier of the extracted road forms using the unbalanced decision tree architecture. Many measures are incorporated for good evaluation of the proposed solution. The obtained results are compared to those of Choi et al. (2007).     

Classification of sleep stages using class-dependent sequential feature selection and artificial neural network

Several studies have been conducted for automatic classification of sleep stages to ease time-consuming manual scoring process that can involve a high degree of experience and subjectivity. But none of them has found a practical usage in medical area so far because of their under acceptable success rates. In this study, a different classification scheme is proposed to increase the success rate in automatic sleep stage scoring in which sleep stages were classified as Awake, Non-REM1, Non-REM2, Non-REM3 and REM stages. Using EEG, EMG and EOG recordings of five healthy subjects, a modified version of sequential feature selection method was applied to the sleep epochs in class by class basis and different artificial neural network (ANN) architectures were trained for each class. That is to say, sleep stages were classified with five ANN architectures each of which uses different features and different network parameters for classification. The highest classification accuracy was obtained for REM sleep as 95.13 % in addition to the lowest classification accuracy of 86.42 % for Non-REM3 sleep. The overall accuracy, on the other hand, was recorded as 90.93 %, which is a comparatively good result when the other studies using all stages are taken into account.     

Optimal channel selection for analysis of EEG-sleep patterns of neonates

This paper extends our previous work on automated detection and classification of neonate EEG sleep stages. In [19] we adapted and integrated a range of computational, mathematical and statistical tools for the analysis of neonatal electroencephalogram (EEG) sleep recordings with the aim of facilitating the assessment of neonatal brain maturation and dismaturity by studying the structure and temporal patterns of their sleep. That work relied on algorithms using a single channel of EEG. The present paper builds on our previous work by incorporating a larger selection of EEG channels that capture both the spatial distribution and temporal patterns of EEG during sleep. Using a multivariate analysis approach, we obtain the "optimal" selection of the EEG channels and characteristics that are most suitable for EEG sleep state separation.     

一种基于脑电信号的眼动方向分类方法

为了提高基于眼电的眼动方向的识别准确性,文中利用包含眼电伪迹的脑电信号,提出了一种新的眼动方向分类方法。首先,在10-20国际标准导联配置下,通过脑电仪采集靠近人脑额叶处的AF7,F7,FT7,T7,AF8,F8,FT8,T8这8个通道的脑电信号;然后,通过基线移除、归一化、最小二乘法降噪等进行数据预处理;最后,采用支持向量机的方法进行眼动方向的多次二分类,并使用投票策略实现眼动方向的四分类识别。实验结果表明,所提方法进行眼动方向分类时,在上、下、左、右4个方向上的分类率分别达到了78.47%,72.22%,84.03%,79.86%,平均分类率达到了78.65%。与已有的分类方法相比,所提方法的分类准确率更高,分类算法的实现过程更简单,这进一步验证了利用脑电信号识别眼动方向的可行性和有效性。     

基于脑电模糊分析的睡眠分期方法研究

利用脑电信号模糊特征分类的方法对睡眠进行分期研究。首先对脑电信号进行预处理,滤除干扰噪声后使用模糊熵算法、多尺度熵算法以及复杂度算法对脑电信号进行特征参数提取,采用最小二乘支持向量机（the Least Squares Support Vector Machine,LS-SVM）对特征参数进行分类,并将睡眠过程分为清醒期、浅睡期、深睡期和快速眼动期（Rapid Eye Movement,REM）,获得分期正确率。最后通过上述方法对2?000组睡眠脑电样本进行睡眠分期测试,与专家人工分期结果进行比对,将复杂度输入到最小二乘支持向量机进行分类的平均正确率是92.65%,高于模糊熵和多尺度熵作为最小二乘向量机的输入时的准确率。基于模糊特征的复杂度提取的特征参数可以作为睡眠分期的有效依据,在保证准确度的前提下,降低人工成本。     

Deep learning analysis of mobile physiological,environmental and location sensor data for emotion detection

The detection and monitoring of emotions are important in various applications, e.g., to enable naturalistic and personalised human-robot interaction. Emotion detection often require modelling of various data inputs from multiple modalities, including physiological signals (e.g., EEG and GSR), environmental data (e.g., audio and weather), videos (e.g., for capturing facial expressions and gestures) and more recently motion and location data. Many traditional machine learning algorithms have been utilised to capture the diversity of multimodal data at the sensors and features levels for human emotion classification. While the feature engineering processes often embedded in these algorithms are beneficial for emotion modelling, they inherit some critical limitations which may hinder the development of reliable and accurate models. In this work, we adopt a deep learning approach for emotion classification through an iterative process by adding and removing large number of sensor signals from different modalities. Our dataset was collected in a real-world study from smart-phones and wearable devices. It merges local interaction of three sensor modalities: on-body, environmental and location into global model that represents signal dynamics along with the temporal relationships of each modality. Our approach employs a series of learning algorithms including a hybrid approach using Convolutional Neural Network and Long Short-term Memory Recurrent Neural Network (CNN-LSTM) on the raw sensor data, eliminating the needs for manual feature extraction and engineering. The results show that the adoption of deep-learning approaches is effective in human emotion classification when large number of sensors input is utilised (average accuracy 95% and F-Measure=%95) and the hybrid models outperform traditional fully connected deep neural network (average accuracy 73% and F-Measure=73%). Furthermore, the hybrid models outperform previously developed Ensemble algorithms that utilise feature engineering to train the model average accuracy 83% and F-Measure=82%)     

Competitive Multi-Verse Optimization with Deep Learning Based Sleep Stage Classification

Sleep plays a vital role in optimum working of the brain and the body. Numerous people suffer from sleep-oriented illnesses like apnea, insomnia, etc. Sleep stage classification is a primary process in the quantitative examination of polysomnographic recording. Sleep stage scoring is mainly based on experts’ knowledge which is laborious and time consuming. Hence, it can be essential to design automated sleep stage classification model using machine learning (ML) and deep learning (DL) approaches. In this view, this study focuses on the design of Competitive Multi-verse Optimization with Deep Learning Based Sleep Stage Classification (CMVODL-SSC) model using Electroencephalogram (EEG) signals. The proposed CMVODL-SSC model intends to effectively categorize different sleep stages on EEG signals. Primarily, data pre-processing is performed to convert the actual data into useful format. Besides, a cascaded long short term memory (CLSTM) model is employed to perform classification process. At last, the CMVO algorithm is utilized for optimally tuning the hyperparameters involved in the CLSTM model. In order to report the enhancements of the CMVODL-SSC model, a wide range of simulations was carried out and the results ensured the better performance of the CMVODL-SSC model with average accuracy of 96.90%.     

自然光普通摄像头的眼部分割及特征点定位数据集ESLD

目的 眼部状态的变化可以作为反映用户真实心理状态及情感变化的依据。由于眼部区域面积较小,瞳孔与虹膜颜色接近,在自然光下利用普通摄像头捕捉瞳孔大小以及位置的变化信息是当前一项具有较大挑战的任务。同时,与现实应用环境类似的具有精细定位和分割信息的眼部结构数据集的欠缺也是制约该领域研究发展的原因之一。针对以上问题,本文利用在普通摄像头场景下采集眼部图像数据,捕捉瞳孔的变化信息并建立了一个眼部图像分割及特征点定位数据集(eye segment and landmark detection dataset,ESLD)。方法 收集、标注并公开发布一个包含多种眼部类型的图像数据集ESLD。采用3种方式采集图像：1)采集用户使用电脑时的面部图像;2)收集已经公开的数据集中满足在自然光下使用普通摄像机条件时采集到的面部图像;3)基于公开软件UnityEye合成的眼部图像。3种采集方式可分别得到1 386幅、804幅和1 600幅眼部图像。得到原始图像后,在原始图像中分割出眼部区域,将不同尺寸的眼部图像归一化为256×128像素。最后对眼部图像的特征点进行人工标记和眼部结构分割。结果 ESLD数据集包含多种类型的眼部图像,可满足研究人员的不同需求。因为实际采集和从公开数据集中获取真实眼部图像十分困难,所以本文利用UnityEye生成眼部图像以改善训练数据量少的问题。实验结果表明,合成的眼部图像可以有效地弥补数据量缺少的问题,F1值可达0.551。利用深度学习方法分别提供了眼部特征点定位和眼部结构分割任务的基线。采用ResNet101作为特征提取网络情况下,眼部特征点定位的误差为5.828,眼部结构分割的mAP (mean average precision)可达0.965。结论 ESLD数据集可为研究人员通过眼部图像研究用户情感变化以及心理状态提供数据支持。