Meditation aims to guide individuals into a state of deep calm and focused attention, and in recent years, it has shown promising potential in the field of medical treatment. Numerous studies have demonstrated that electroencephalogram (EEG) patterns change during meditation, suggesting the feasibility of using deep learning techniques to monitor meditation states. However, significant inter-subject differences in EEG signals poses challenges to the performance of such monitoring systems. To address this issue, this study proposed a novel model—calibrated multi-source adversarial adaptation network (CMAAN). The model first trained multiple domain-adversarial neural networks in a pairwise manner between various source-domain individuals and the target-domain individual. These networks were then integrated through a calibration process using a small amount of labeled data from the target domain to enhance performance. We evaluated the proposed model on an EEG dataset collected from 18 subjects undergoing methamphetamine rehabilitation. The model achieved a classification accuracy of 73.09%. Additionally, based on the learned model, we analyzed the key EEG frequency bands and brain regions involved in the meditation process. The proposed multi-source domain adaptation framework improves both the performance and robustness of EEG-based meditation monitoring and holds great promise for applications in biomedical informatics and clinical practice.
ObjectiveTo probe the clinical features and the characteristics of radiography and electroencephalogram (EEG) of tuberous sclerosis complex(TSC) in children with epilepsy. MethodsThe clinical data of the TSC cases with epilepsy were collected from inpatients in Jiangxi Children's Hospital from Jan. 2013 to Oct. 2015. ResultsAmong the 26 cases, 21 cases(21/26, 80.77%) involved abnormalities of the skin. Of these patients, there were 10 cases with hypomelanotic macules, 7 cases with café au lait spots and 4 cases with facial angiofibromas. There were no significant difference among the different age groups. In addition, there were 8 cases (8/26, 30.77%) with spasm seizures, of whom 3 cases had partial seizure, 10 cases (10/26, 38.46%) with complex partial seizure, 5 cases(5/26, 19.23%) with secondary generalized seizure, 2 cases(2/26, 7.69%) with tonic-clonic seizure and one case with Lennox-Gastaut syndrom(1/26, 3.85%). The average onset age of the epileptic spasms group were younger than those of the other epilepsy groups (t=2.143, P=0.042). EEG monitoring demonstrated hypsarrhythmia in 7 cases (7/26, 26.92%) in the interictal EEG, focal epileptic discharges in 11 cases (11/26, 42.31%), multifocal discharges in 5 cases, the slow background activity in 2 cases and the normal EEG in one case. Cranial imaging demonstrated subependymal nodules (SEN) in 25 cases(25/26, 96.15%) was the most common. ConclusionThe clinical manifestations and seizure types of TSC in children, especially in infants and young children, were diverse and age-dependent. It was very important to improve understanding of the clinical features and related risks of TSC at various ages, which was helpful to diagnose TSC early.
Due to the high complexity and subject variability of motor imagery electroencephalogram, its decoding is limited by the inadequate accuracy of traditional recognition models. To resolve this problem, a recognition model for motor imagery electroencephalogram based on flicker noise spectrum (FNS) and weighted filter bank common spatial pattern (wFBCSP) was proposed. First, the FNS method was used to analyze the motor imagery electroencephalogram. Using the second derivative moment as structure function, the ensued precursor time series were generated by using a sliding window strategy, so that hidden dynamic information of transition phase could be captured. Then, based on the characteristic of signal frequency band, the feature of the transition phase precursor time series and reaction phase series were extracted by wFBCSP, generating features representing relevant transition and reaction phase. To make the selected features adapt to subject variability and realize better generalization, algorithm of minimum redundancy maximum relevance was further used to select features. Finally, support vector machine as the classifier was used for the classification. In the motor imagery electroencephalogram recognition, the method proposed in this study yielded an average accuracy of 86.34%, which is higher than the comparison methods. Thus, our proposed method provides a new idea for decoding motor imagery electroencephalogram.
Brain age prediction, as a significant approach for assessing brain health and early diagnosing neurodegenerative diseases, has garnered widespread attention in recent years. Electroencephalogram (EEG), an non-invasive, convenient, and cost-effective neurophysiological signal, offers unique advantages for brain age prediction due to its high temporal resolution and strong correlation with brain functional states. Despite substantial progress in enhancing prediction accuracy and generalizability, challenges remain in data quality and model interpretability. This review comprehensively examined the advancements in EEG-based brain age prediction, detailing key aspects of data preprocessing, feature extraction, model construction, and result evaluation. It also summarized the current applications of machine learning and deep learning methods in this field, analyzed existing issues, and explored future directions to promote the widespread application of EEG-based brain age prediction in both clinical and research settings.
Fear emotion is a typical negative emotion that is commonly present in daily life and significantly influences human behavior. A deeper understanding of the mechanisms underlying negative emotions contributes to the improvement of diagnosing and treating disorders related to negative emotions. However, the neural mechanisms of the brain when faced with fearful emotional stimuli remain unclear. To this end, this study further combined electroencephalogram (EEG) source analysis and cortical brain network construction based on early posterior negativity (EPN) analysis to explore the differences in brain information processing mechanisms under fearful and neutral emotional picture stimuli from a spatiotemporal perspective. The results revealed that neutral emotional stimuli could elicit higher EPN amplitudes compared to fearful stimuli. Further source analysis of EEG data containing EPN components revealed significant differences in brain cortical activation areas between fearful and neutral emotional stimuli. Subsequently, more functional connections were observed in the brain network in the alpha frequency band for fearful emotions compared to neutral emotions. By quantifying brain network properties, we found that the average node degree and average clustering coefficient under fearful emotional stimuli were significantly larger compared to neutral emotions. These results indicate that combining EPN analysis with EEG source component and brain network analysis helps to explore brain functional modulation in the processing of fearful emotions with higher spatiotemporal resolution, providing a new perspective on the neural mechanisms of negative emotions.
Electroencephalogram (EEG) is characterized by high temporal resolution, and various EEG analysis methods have developed rapidly in recent years. The EEG microstate analysis method can be used to study the changes of the brain in the millisecond scale, and can also present the distribution of EEG signals in the topological level, thus reflecting the discontinuous and nonlinear characteristics of the whole brain. After more than 30 years of enrichment and improvement, EEG microstate analysis has penetrated into many research fields related to brain science. In this paper, the basic principles of EEG microstate analysis methods are summarized, and the changes of characteristic parameters of microstates, the relationship between microstates and brain functional networks as well as the main advances in the application of microstate feature extraction and classification in brain diseases and brain cognition are systematically described, hoping to provide some references for researchers in this field.
Individuals with motor dysfunction caused by damage to the central nervous system are unable to transmit voluntary movement commands to their muscles, resulting in a reduced ability to control their limbs. However, traditional rehabilitation methods have problems such as long treatment cycles and high labor costs. Functional electrical stimulation (FES) based on brain-computer interface (BCI) connects the patient’s intentions with muscle contraction, and helps to promote the reconstruction of nerve function by recognizing nerve signals and stimulating the moving muscle group with electrical impulses to produce muscle convulsions or limb movements. It is an effective treatment for sequelae of neurological diseases such as stroke and spinal cord injury. This article reviewed the current research status of BCI-based FES from three aspects: BCI paradigms, FES parameters and rehabilitation efficacy, and looked forward to the future development trend of this technology, in order to improve the understanding of BCI-based FES.
ObjectiveTo analyze and summarize the clinical and video EEG (VEEG) characteristics of tuberous sclerosis (TSC) with epilepsy.MethodsClinical data of 30 children with TSC who met the revised diagnostic criteria of TSC in 2012 from Jan. 2016 to May 2019 in Zhengzhou Children’s Hospital were collected, including 29 children with epileptic seizures. The characteristics of skin lesions, imaging, seizures and long-term VEEG were analyzed retrospectively.ResultsThe mean age was (2.88 ± 2.64), 12 males and 18 females, 1 case of lumbar acid as the first symptom, 29 cases with epilepsy as the first symptom, the incidence of epilepsy is high, and the onset age is less than 1 year old; TSC can cause different degrees of cognitive impact; depigmentation or milk coffee spots are the most common skin changes in young children; TSC with infantile spasm has a high incidence; children younger than 10 years old may have lesions of other organs except nervous system lesions. However, the incidence of other organ lesions was relatively low. Most of TSC children with epilepsy were accompanied by abnormal EEG discharge.ConclusionThe clinical characteristics of TSC with epileptic seizures are various, and early diagnosis is of great significance.
Emotion classification and recognition is a crucial area in emotional computing. Physiological signals, such as electroencephalogram (EEG), provide an accurate reflection of emotions and are difficult to disguise. However, emotion recognition still faces challenges in single-modal signal feature extraction and multi-modal signal integration. This study collected EEG, electromyogram (EMG), and electrodermal activity (EDA) signals from participants under three emotional states: happiness, sadness, and fear. A feature-weighted fusion method was applied for integrating the signals, and both support vector machine (SVM) and extreme learning machine (ELM) were used for classification. The results showed that the classification accuracy was highest when the fusion weights were set to EEG 0.7, EMG 0.15, and EDA 0.15, achieving accuracy rates of 80.19% and 82.48% for SVM and ELM, respectively. These rates represented an improvement of 5.81% and 2.95% compared to using EEG alone. This study offers methodological support for emotion classification and recognition using multi-modal physiological signals.
In clinical, manually scoring by technician is the major method for sleep arousal detection. This method is time-consuming and subjective. This study aimed to achieve an end-to-end sleep-arousal events detection by constructing a convolutional neural network based on multi-scale convolutional layers and self-attention mechanism, and using 1 min single-channel electroencephalogram (EEG) signals as its input. Compared with the performance of the baseline model, the results of the proposed method showed that the mean area under the precision-recall curve and area under the receiver operating characteristic were both improved by 7%. Furthermore, we also compared the effects of single modality and multi-modality on the performance of the proposed model. The results revealed the power of single-channel EEG signals in automatic sleep arousal detection. However, the simple combination of multi-modality signals may be counterproductive to the improvement of model performance. Finally, we also explored the scalability of the proposed model and transferred the model into the automated sleep staging task in the same dataset. The average accuracy of 73% also suggested the power of the proposed method in task transferring. This study provides a potential solution for the development of portable sleep monitoring and paves a way for the automatic sleep data analysis using the transfer learning method.