Sleep State Classification Using Power Spectral Density and Residual Neural Network with Multichannel EEG Signals

Multiscale-temporal Feature Extraction and boundary confusion alleviation for VAG-based fine-grained multi-grade osteoarthritis deterioration monitoring

BACKGROUND AND OBJECTIVE

Multi-grade osteoarthritis (OA) deterioration monitoring in the daily paradigm using Vibroarthrography (VAG) is very challenging due to two difficulties: (1) the composition of VAG signals is complex in the daily paradigm where friction is intensified because of weight-bearing movements. (2) VAG signal samples near the decision boundary of adjacent deterioration grades are easy to be misclassified. The majority of existing works only focus on the binary classification of OA, providing inadequate assistance in instructing physicians to develop treatment plans based on the presence or absence of OA. Thus, we propose a novel framework for fine-grained multi-grade OA deterioration monitoring in the daily paradigm.

METHODS

We propose an end-to-end deep learning framework termed Fine-grained Multi-grade OA Deterioration Monitor (FMOADM), which consists of Multiscale-temporal Feature Extraction (MTFE) and Confusion-Free Master-Slave (CF-MS) Classification. Specifically, MTFE is adopted to extract multiscale-temporal discriminative features from the complicated VAG signals. And center loss is introduced by CF-MS to alleviate confusion at the boundary of adjacent deterioration grades in the feature space. Meanwhile, a master-slave structure is proposed for further fine-grained classification, where the master classifier integrates a channel attention mechanism and the slave classifier is designed to update MTFE parameters. As a result, the proposed method ensures fine-grained multi-grade OA monitoring performance via multiscale-temporal discriminative features and boundary confusion alleviation.

RESULTS

Experimental results on the VAG-OA dataset demonstrate that our framework outperforms counterpart methods in the daily paradigm. The proposed framework achieved 78% in precision, obtaining an 8% improvement over the state-of-the-art method.

CONCLUSION

The proposed framework benefits efficient multi-grade OA deterioration monitoring, empowering physicians to develop treatment plans based on fine-grained monitoring results. It takes knee joint health monitoring in daily activities a step further toward feasible.

Evaluating the depression state during perinatal period by non-invasive scalp EEG

Perinatal depression, with a prevalence of 10 to 20% in United States, is usually missed as multiple symptoms of perinatal depression are common in pregnant women. Worse, the diagnosis of perinatal depression still largely relies on questionnaires, leaving the objective biomarker being unveiled yet. This study suggested a safe and non-invasive technique to diagnose perinatal depression and further explore its underlying mechanism. Considering the non-invasiveness and clinical convenience of electroencephalogram for mothers-to-be and fetuses, we collected the resting-state electroencephalogram of pregnant women at the 38th week of gestation. Subsequently, the difference in network topology between perinatal depression patients and healthy mothers-to-be was explored, with related spatial patterns being adopted to achieve the classification of pregnant women with perinatal depression from those healthy ones. We found that the perinatal depression patients had decreased brain network connectivity, which indexed impaired efficiency of information processing. By adopting the spatial patterns, the perinatal depression could be accurately recognized with an accuracy of 87.88%; meanwhile, the depression severity at the individual level was effectively predicted, as well. These findings consistently illustrated that the resting-state electroencephalogram network could be a reliable tool for investigating the depression state across pregnant women, and will further facilitate the clinical diagnosis of perinatal depression.

Temporal-spatial convolutional residual network for decoding attempted movement related EEG signals of subjects with spinal cord injury

Brain Computer Interface (BCI) offers a promising approach to restoring hand functionality for people with cervical spinal cord injury (SCI). A reliable classification of brain activities based on appropriate flexibility in feature extraction could enhance BCI systems performance. In the present study, based on convolutional layers with temporal-spatial, Separable and Depthwise structures, we develop Temporal-Spatial Convolutional Residual Network)TSCR-Net(and Temporal-Spatial Convolutional Iterative Residual Network)TSCIR-Net(structures to classify electroencephalogram (EEG) signals. Using EEG signals in five different hand movement classes of SCI people, we compare the effectiveness of TSCIR-Net and TSCR-Net models with some competitive methods. We use the bayesian hyperparameter optimization algorithm to tune the hyperparameters of compact convolutional neural networks. In order to show the high generalizability of the proposed models, we compare the results of the models in different frequency ranges. Our proposed models decoded distinctive characteristics of different movement efforts and obtained higher classification accuracy than previous deep neural networks. Our findings indicate that TSCIR-Net and TSCR-Net models fulfills a better classification accuracy of 71.11%, and 64.55% for EEG_All and 57.74%, and 67.87% for EEG_Low frequency data sets than the compared methods in the literature.

Temporal Feature Extraction and Machine Learning for Classification of Sleep Stages Using Telemetry Polysomnography

Accurate sleep stage detection is crucial for diagnosing sleep disorders and tailoring treatment plans. Polysomnography (PSG) is considered the gold standard for sleep assessment since it captures a diverse set of physiological signals. While various studies have employed complex neural networks for sleep staging using PSG, our research emphasises the efficacy of a simpler and more efficient architecture. We aimed to integrate a diverse set of feature extraction measures with straightforward machine learning, potentially offering a more efficient avenue for sleep staging. We also aimed to conduct a comprehensive comparative analysis of feature extraction measures, including the power spectral density, Higuchi fractal dimension, singular value decomposition entropy, permutation entropy, and detrended fluctuation analysis, coupled with several machine-learning models, including XGBoost, Extra Trees, Random Forest, and LightGBM. Furthermore, data augmentation methods like the Synthetic Minority Oversampling Technique were also employed to rectify the inherent class imbalance in sleep data. The subsequent results highlighted that the XGBoost classifier, when used with a combination of all feature extraction measures as an ensemble, achieved the highest performance, with accuracies of 87%, 90%, 93%, 96%, and 97% and average F1-scores of 84.6%, 89%, 90.33%, 93.5%, and 93.5% for distinguishing between five-stage, four-stage, three-stage, and two distinct two-stage sleep configurations, respectively. This combined feature extraction technique represents a novel addition to the body of research since it achieves higher performance than many recently developed deep neural networks by utilising simpler machine-learning models.

Current status and prospects of automatic sleep stages scoring: Review

The scoring of sleep stages is one of the essential tasks in sleep analysis. Since a manual procedure requires considerable human and financial resources, and incorporates some subjectivity, an automated approach could result in several advantages. There have been many developments in this area, and in order to provide a comprehensive overview, it is essential to review relevant recent works and summarise the characteristics of the approaches, which is the main aim of this article. To achieve it, we examined articles published between 2018 and 2022 that dealt with the automated scoring of sleep stages. In the final selection for in-depth analysis, 125 articles were included after reviewing a total of 515 publications. The results revealed that automatic scoring demonstrates good quality (with Cohen's kappa up to over 0.80 and accuracy up to over 90%) in analysing EEG/EEG + EOG + EMG signals. At the same time, it should be noted that there has been no breakthrough in the quality of results using these signals in recent years. Systems involving other signals that could potentially be acquired more conveniently for the user (e.g. respiratory, cardiac or movement signals) remain more challenging in the implementation with a high level of reliability but have considerable innovation capability. In general, automatic sleep stage scoring has excellent potential to assist medical professionals while providing an objective assessment.

LDDNet: A Deep Learning Framework for the Diagnosis of Infectious Lung Diseases

This paper proposes a new deep learning (DL) framework for the analysis of lung diseases, including COVID-19 and pneumonia, from chest CT scans and X-ray (CXR) images. This framework is termed optimized DenseNet201 for lung diseases (LDDNet). The proposed LDDNet was developed using additional layers of 2D global average pooling, dense and dropout layers, and batch normalization to the base DenseNet201 model. There are 1024 Relu-activated dense layers and 256 dense layers using the sigmoid activation method. The hyper-parameters of the model, including the learning rate, batch size, epochs, and dropout rate, were tuned for the model. Next, three datasets of lung diseases were formed from separate open-access sources. One was a CT scan dataset containing 1043 images. Two X-ray datasets comprising images of COVID-19-affected lungs, pneumonia-affected lungs, and healthy lungs exist, with one being an imbalanced dataset with 5935 images and the other being a balanced dataset with 5002 images. The performance of each model was analyzed using the Adam, Nadam, and SGD optimizers. The best results have been obtained for both the CT scan and CXR datasets using the Nadam optimizer. For the CT scan images, LDDNet showed a COVID-19-positive classification accuracy of 99.36%, a 100% precision recall of 98%, and an F1 score of 99%. For the X-ray dataset of 5935 images, LDDNet provides a 99.55% accuracy, 73% recall, 100% precision, and 85% F1 score using the Nadam optimizer in detecting COVID-19-affected patients. For the balanced X-ray dataset, LDDNet provides a 97.07% classification accuracy. For a given set of parameters, the performance results of LDDNet are better than the existing algorithms of ResNet152V2 and XceptionNet.

Automated Analysis of Sleep Study Parameters Using Signal Processing and Artificial Intelligence

An automated sleep stage categorization can readily face noise-contaminated EEG recordings, just as other signal processing applications. Therefore, the denoising of the contaminated signals is inevitable to ensure a reliable analysis of the EEG signals. In this research work, an empirical mode decomposition is used in combination with stacked autoencoders to conduct automatic sleep stage classification with reliable analytical performance. Due to the decomposition of the composite signal into several intrinsic mode functions, empirical mode decomposition offers an effective solution for denoising non-stationary signals such as EEG. Preliminary results showed that through these intrinsic modes, a signal with a high signal-to-noise ratio can be obtained, which can be used for further analysis with confidence. Therefore, later, when statistical features were extracted from the denoised signals and were classified using stacked autoencoders, improved results were obtained for Stage 1, Stage 2, Stage 3, Stage 4, and REM stage EEG signals using this combination.

A Depression Diagnosis Method Based on the Hybrid Neural Network and Attention Mechanism

Depression is a common but easily misdiagnosed disease when using a self-assessment scale. Electroencephalograms (EEGs) provide an important reference and objective basis for the identification and diagnosis of depression. In order to improve the accuracy of the diagnosis of depression by using mainstream algorithms, a high-performance hybrid neural network depression detection method is proposed in this paper combined with deep learning technology. Firstly, a concatenating one-dimensional convolutional neural network (1D-CNN) and gated recurrent unit (GRU) are employed to extract the local features and to determine the global features of the EEG signal. Secondly, the attention mechanism is introduced to form the hybrid neural network. The attention mechanism assigns different weights to the multi-dimensional features extracted by the network, so as to screen out more representative features, which can reduce the computational complexity of the network and save the training time of the model while ensuring high precision. Moreover, dropout is applied to accelerate network training and address the over-fitting problem. Experiments reveal that the 1D-CNN-GRU-ATTN model has more effectiveness and a better generalization ability compared with traditional algorithms. The accuracy of the proposed method in this paper reaches 99.33% in a public dataset and 97.98% in a private dataset, respectively.

Convolutional Neural Network for Drowsiness Detection Using EEG Signals

Drowsiness detection (DD) has become a relevant area of active research in biomedical signal processing. Recently, various deep learning (DL) researches based on the EEG signals have been proposed to detect fatigue conditions. The research presented in this paper proposes an EEG classification system for DD based on DL networks. However, the proposed DD system is mainly realized into two procedures; (i) data acquisition and (ii) model analysis. For the data acquisition procedure, two key steps are considered, which are the signal collection using a wearable Emotiv EPOC+ headset to record 14 channels of EEG, and the signal annotation. Furthermore, a data augmentation (DA) step has been added to the proposed system to overcome the problem of over-fitting and to improve accuracy. As regards the model analysis, a comparative study is also introduced in this paper to argue the choice of DL architecture and frameworks used in our DD system. In this sense, The proposed DD protocol makes use of a convolutional neural network (CNN) architecture implemented using the Keras library. The results showed a high accuracy value (90.42%) in drowsy/awake discrimination and revealed the efficiency of the proposed DD system compared to other research works.