Automatic detection of absence seizures with compressive sensing EEG

Automatic recognition of epileptic EEG patterns via Extreme Learning Machine and multiresolution feature extraction

In this paper the new model is proposed to automatically detect and predict absence seizure using hybrid deep learning algorithm [convolutional recurrent neural network (CRNN)] with single channel electroencephalography (EEG) only as input. This model comprises of four steps: 1) single channel segmentation process; 2) extraction of relevant features using convolution network; 3) recurrent network for detection and early detection; 4) SVM have been used as last layer to obtain a result with respect to time. This model enhances the feature extraction by feeding the raw input into convolutional layer, improves the detection with gated recurrent unit (GRU) and reduces the early detection rate with support vector machine (SVM). Our proposed model achieves 100% overall accuracy on seizure detection as normal and absence seizure and detect within three seconds of the overall seizure duration. Also this model can be act as a generic model for classification task with detection and early detection of bio-signal (EEG, ECG and EMG).

Dynamic characteristics of absence EEG recordings with multiscale permutation entropy analysis

n/a

The automated epileptic seizure detection has emerged as an important field in the recent years; this involves analyzing the Electroencephalogram (EEG) signals instead of the traditional visual inspection performed by expert neurologists. In this paper, a model has been introduced that integrates Weighted Permutation Entropy (WPE) as input feature to a Support Vector Machine (SVM) learning model to enhance the sensitivity and precision of the detection process. WPE is a modified statistical parameter of the permutation entropy (PE) that measures the complexity and irregularity of a time series. It incorporates both the mapped ordinal pattern of the time series and the information contained in the amplitude of its sample points. The proposed system utilizes the fact that entropy based measures for the EEG segments during epileptic seizure are lower than in normal EEG. Experiments have been conducted to demonstrate the sensitivity and precision of the proposed model. Results showed that the model can considerably discriminate between EEG brain signal into seizure and no-seizure classes.

Epilepsy is a chronic neurological disorder that affects more than 50 million people world wide, characterized by recurrent seizures (World Health Organization [WHO], 2006). An epileptic seizure is a transient occurrence of signs and/or symptoms due to abnormal excessive or synchronous neuronal activity in the brain (Fisher et al., 2005 & Berg et al., 2010). This electrical hyperactivity can have its source in different parts of the brain and produces physical symptoms such as short periods of inattention and loose of memory, a sensory hallucination, or a whole-body convulsion. The frequency of these events can vary from one in a year to several in a day. The majority of the patients suffer from unpredictable, persistent and frequent seizures which limit the independence of an individual, increase the risk of serious injury and mobility, and result in both social isolation and economic hardship (Friedman & Gilliam, 2010). In addition, the patients with epilepsy have an increased mortality risk of approximately 2 to 3 times that of the general population (Ficker, 2000). The first line of treatment for epilepsy is with multiple anti-epileptic drugs and it is effective in about 70% of the cases. From the 30% remaining affected individuals only less than 10% could benefit from surgical therapy leaving a 20% of the total of people with epilepsy who will continue suffering sudden, incontrollable seizures and for whom other forms of treatment are being investigated (Theodore & Ficker, 2004; WHO, 2006). For any of the reasons exposed before the seizure detection is an important component in the diagnosis of epilepsy and for the seizures control. In the clinical practice this detection basically involves visual scanning of Electroencephalogram (EEG) long recordings by the physicians in order to detect and classify the seizure activity present in the EEG signal. Usually these are multichannel records of 24 to 72 hours length which implies a very time consuming task and it is also kwon that the conclusions are very subjective so disagreement between physicians are not rare. The seek here is to detect automatically in long term EEG records those segments of the signal that present epileptic seizures for the shake of reducing the high amount of information to be analyzed by the neurologists. Thus them could focus their attention in these part of the information so a more precisely and quick diagnosis can be made. Seizure detection is also a useful tool for treatments such us timely drug delivery, electrical stimulation and seizure alert systems. Automated seizure detection, quantification and recognition have been of interest of the biomedical community researchers since the 1970s. In some initial works a number of

Electroencephalography (EEG) is considered the output of a brain and it is a bioelectrical signal with multiscale and nonlinear properties. Motor Imagery EEG (MI-EEG) not only has a close correlation with the human imagination and movement intention but also contains a large amount of physiological or disease information. As a result, it has been fully studied in the field of rehabilitation. To correctly interpret and accurately extract the features of MI-EEG signals, many nonlinear dynamic methods based on entropy, such as Approximate Entropy (ApEn), Sample Entropy (SampEn), Fuzzy Entropy (FE), and Permutation Entropy (PE), have been proposed and exploited continuously in recent years. However, these entropy-based methods can only measure the complexity of MI-EEG based on a single scale and therefore fail to account for the multiscale property inherent in MI-EEG. To solve this problem, Multiscale Sample Entropy (MSE), Multiscale Permutation Entropy (MPE), and Multiscale Fuzzy Entropy (MFE) are developed by introducing scale factor. However, MFE has not been widely used in analysis of MI-EEG, and the same parameter values are employed when the MFE method is used to calculate the fuzzy entropy values on multiple scales. Actually, each coarse-grained MI-EEG carries the characteristic information of the original signal on different scale factors. It is necessary to optimize MFE parameters to discover more feature information. In this paper, the parameters of MFE are optimized independently for each scale factor, and the improved MFE (IMFE) is applied to the feature extraction of MI-EEG. Based on the event-related desynchronization (ERD)/event-related synchronization (ERS) phenomenon, IMFE features from multi channels are fused organically to construct the feature vector. Experiments are conducted on a public dataset by using Support Vector Machine (SVM) as a classifier. The experiment results of 10-fold cross-validation show that the proposed method yields relatively high classification accuracy compared with other entropy-based and classical time–frequency–space feature extraction methods. The t-test is used to prove the correctness of the improved MFE.

This study investigates the feasibility of using tear sample analysis, based on protein corona formation on gold nanoparticles combined with electrospray ionization mass spectrometry (ESI-MS) and machine learning techniques, as a non-invasive approach for the detection of choroidal melanoma. The aim is to assess whether protein-nanoparticle interactions can support early and reliable identification of this ocular condition. Tear samples were collected using Schirmer strips from six healthy individuals and six patients diagnosed with choroidal melanoma, with subsequent augmentation to 18 samples per group. Gold nanoparticles (AuNPs, ~ 20 nm) were synthesized via citrate reduction and incubated with tear samples to form protein coronas, which were analyzed using ESI-MS. Eight statistical and entropy-based features (mean, variance, skewness, kurtosis, Shannon entropy, approximate entropy, sample entropy, and permutation entropy) were extracted from spectral data. Additionally, Continuous Wavelet Transform (CWT) with Mexican hat wavelet was applied to convert mass spectrometry data into 128 × 128 RGB images for deep learning analysis. Classification was performed using traditional machine learning models (Random Forest, Support Vector Machine, Decision Tree, Deep Neural Network) and transfer learning with pre-trained CNNs (VGG16, ResNet50, Xception), evaluated through 5-fold cross-validation. Significant differences in spectral intensity parameters were observed between healthy individuals and choroidal melanoma patients (p < 0.001), with notably lower Mean_Intensity values in cancer patients (56.41 ± 46.06 vs. 111.02 ± 10.01, Cohen’s d = 1.64). While m/z parameters showed moderate differences that didn’t reach statistical significance (p = 0.082), entropy-based features demonstrated strong discriminative power. Among traditional machine learning models, Random Forest achieved the highest accuracy (0.959 ± 0.003) and ROC AUC (0.993 ± 0.000) with remarkable computational efficiency (3.90 s per fold). For deep learning approaches using CWT-generated images, VGG16 demonstrated superior performance (Accuracy: 0.976 ± 0.008, ROC AUC: 0.997 ± 0.002) despite requiring significantly higher computational resources (1349.52 s per fold). This study demonstrates that tear sample analysis using protein corona formation on gold nanoparticles with ESI-MS and advanced machine learning techniques offers a promising non-invasive approach for choroidal melanoma detection with performance metrics that compare favorably to existing methods. The significant differences in spectral intensity parameters between groups suggest distinctive proteomic signatures that can be leveraged for diagnostic purposes. While both traditional machine learning and deep learning approaches achieved exceptional performance, each offers distinct advantages in terms of computational efficiency and feature extraction capabilities.

Real-time seizure onset detection has been an active area of research in the study of epilepsy. Electroencephalography (EEG) measurements are widely used in seizure detection due to their high temporal resolution, cost-effective, and noninvasive nature. Various approaches based on machine learning are used for epileptic seizure detection, but these approaches do not explicitly reveal the underlying dynamics, require larger datasets for training, and are computationally demanding. Although high-frequency oscillations (HFOs) are the new biomarkers of epilepsy, they cannot be used with the existing data acquisition systems as they require high sampling rates and high cutoff frequency of the used filters. In this article, we present a novel approach for real-time seizure detection using high spatial frequencies. Since eigenvalues of the graph Laplacian represent spatial frequencies, we conjecture that higher eigenvalues and eigenvectors will contain the detailed information of seizure and non-seizure brain states. Hence, we have formed sub-band characteristic response vector (sub-band CRV) using weighted sum of eigenvectors corresponding to higher spatial frequencies and analyzed it over time. We have used a publicly available dataset to demonstrate the efficacy of the proposed approach. We observed that the proposed approach performs satisfactorily well in real-time automated seizure detection without requiring any kind of prior training. Moreover, our approach is not only accurate in seizure detection but also is independent of sampling rates, hence can be implemented easily in clinical realm for developing an automated seizure detection tool with the existing data acquisition systems operating at low sampling rates.

BackgroundTo address the limitation in gait assessment for patients with knee osteoarthritis (KOA) and after total knee arthroplasty (TKA), where it is difficult to simultaneously quantify joint dynamic coordination and movement complexity, a multidimensional gait feature fusion algorithm is proposed.MethodsSpatial motion data were collected from 70 participants (21 healthy controls, 24 KOA patients, and 25 post-TKA patients) using a 3D motion capture system. Hip-knee cyclograms were constructed to extract morphological features (centroid, range of motion, perimeter, and area) for quantifying dynamic coordination, while sample entropy of hip, knee, and ankle joint angles was calculated to quantify movement complexity. Features were categorized into four input types: fused multidimensional features, cyclogram morphological features, sample entropy features, and traditional spatiotemporal parameters. Machine learning models including Random Forest (RF), Support Vector Machine (SVM), Decision Tree (DT), and k- Nearest Neighbors (KNN) were employed for gait classification and assessment.ResultsMultidimensional feature analysis revealed a characteristic pathological compensation pattern of “decreased cyclogram features with increased sample entropy” in the KOA group, while the TKA group demonstrated postoperative improvements in both dimensions. The incorporation of multidimensional features significantly enhanced the performance of all classification models: under multidimensional feature input, RF, SVM, DT, and KNN achieved accuracies of 96.93%, 92.44%, 90.29%, and 88.98%, respectively—all significantly outperforming models using single‐dimensional features.ConclusionThe multidimensional gait feature fusion algorithm effectively overcomes the limitation of assessing either coordination or complexity in isolation, providing an interpretable quantitative tool for analyzing KOA pathological mechanisms and dynamically monitoring post-TKA rehabilitation.

Wireless Body Area Network (WBAN) is a pioneer trend in healthcare technology.A WBAN consists of small sensors that may be worn or implanted on the patient's body or around them.These sensors are responsible for sending wireless, real-time physiological signs of the patient's body (such as blood pressure, heart rate, blood sugar, temperature, breathing, etc.) to an intermediate device.An intermediate device (such as a mobile phone) collects and prepares data to send and display to the doctor.In the WBAN, since data is sent through all broadcasts, any cyber attack such as eavesdropping and data modification/ sabotage can be done.Any cyber-attack on a WBAN could jeopardize the patient's health; therefore, securing the WBAN plays a crucial role in healthcare applications.Due to the limitation of energy, memory, and processing power of sensors in WBAN, it is not possible to use traditional security methods, such as encryption and security protocols.For this reason, one of the methods to provid security in the WBAN is the use of intrusion detection systems.An intrusion detection system (IDS), as a second-line defense, is one of the security methods in computer networks.Intrusion detection is the intelligent monitoring of network or computer systems to find any security breach.These systems are divided into two groups, signature-based and anomaly-based.One of the advantages 1 55 160 of anomaly-based systems is the detection of zero-day attacks.Statistical, knowledge-based, data mining and machine learning methods can be used to implement anomaly-based intrusion detection systems.In this paper, a new IDS has been presented which is able to detect denial of service (DoS) attacks in a WBAN.The proposed IDS is a hybrid anomaly-based system using machine learning and feature engineering methods.In the proposed IDS, a genetic algorithm is used to select features of collected data, in a way that increases the performance of the IDS, and as a result the WBAN (increasing detection rate and reducing energy consumption in the node).Then, using support vector machine and k nearest neighbor algorithms, the data classification is performed to detect DoS traffic from regular data traffic.Simulation results indicate that the proposed IDS has effective performance with a 90% detection rate.

Automatic seizure detection is very essential for monitoring and rehabilitation of epilepsy patients and will open up new treatment possibilities for saving the lives of epileptic patients. In recent years, many algorithms for the automatic seizure detection have been proposed and applied, in which Support vector machines proved to be a robust machine learning algorithm. The purpose of this study is to compute relevant EEG features and apply a feature selection algorithm to select an optimum set of features for use in a classification scheme for epileptic seizure detection. Thus S VM will thereby yield a better accuracy compared to other algorithms. Effective features such as energy, relative amplitude, standard deviation, coefficient of variation, fluctuation index etc are selected and then these features are fed into the support vector machine for training and classification. This algorithm makes use of Radial Basis Function Kernels for training data and thus obtains more accurate results.

In order to accurately identify various types of ships and develop coastal defenses, a single feature extraction method based on slope entropy (SlEn) and a double feature extraction method based on SlEn combined with permutation entropy (SlEn&PE) are proposed. Firstly, SlEn is used for the feature extraction of ship-radiated noise signal (SNS) compared with permutation entropy (PE), dispersion entropy (DE), fluctuation dispersion entropy (FDE), and reverse dispersion entropy (RDE), so that the effectiveness of SlEn is verified, and SlEn has the highest recognition rate calculated by the k-Nearest Neighbor (KNN) algorithm. Secondly, SlEn is combined with PE, DE, FDE, and RDE, respectively, to extract the feature of SNS for a higher recognition rate, and SlEn&PE has the highest recognition rate after the calculation of the KNN algorithm. Lastly, the recognition rates of SlEn and SlEn&PE are compared, and the recognition rates of SlEn&PE are higher than SlEn by 4.22%. Therefore, the double feature extraction method proposed in this paper is more effective in the application of ship type recognition.

Electroencephalogram is the recording of brain electrical activity and it contains valuable information related to the different physiological states of the brain. In this study, we present a new approach to detect epileptic seizure. The new scheme was based on discrete wavelet transform and sample entropy analysis of EEG signals. Decision making is performed in two stages: feature extraction by computing the wavelet coefficients and the sample entropy and detection by using support vector machine. The analysis results depicted that during seizure activity EEG had lower sample entropy values compared to normal EEG. This suggested that epileptic EEG was more predictable or less complex than the normal EEG.

The aim of the present study was to discriminate the electroencephalogram (EEG) of 5 patients with vascular dementia (VaD), 15 patients with stroke-related mild cognitive impairment (MCI), and 15 control normal subjects during a working memory (WM) task. We used independent component analysis (ICA) and wavelet transform (WT) as a hybrid preprocessing approach for EEG artifact removal. Three different features were extracted from the cleaned EEG signals: spectral entropy (SpecEn), permutation entropy (PerEn) and Tsallis entropy (TsEn). Two classification schemes were applied - support vector machine (SVM) and k-nearest neighbors (kNN) - with fuzzy neighborhood preserving analysis with QR-decomposition (FNPAQR) as a dimensionality reduction technique. The FNPAQR dimensionality reduction technique increased the SVM classification accuracy from 82.22% to 90.37% and from 82.6% to 86.67% for kNN. These results suggest that FNPAQR consistently improves the discrimination of VaD, MCI patients and control normal subjects and it could be a useful feature selection to help the identification of patients with VaD and MCI.