RESUMEN
The intelligent recognition of epileptic electro-encephalogram (EEG) signals is a valuable tool for the epileptic seizure detection. Recent deep learning models fail to fully consider both spectral and temporal domain representations simultaneously, which may lead to omitting the nonstationary or nonlinear property in epileptic EEGs and further produce a suboptimal recognition performance consequently. In this paper, an end-to-end EEG seizure detection framework is proposed by using a novel channel-embedding spectral-temporal squeeze-and-excitation network (CE-stSENet) with a maximum mean discrepancy-based information maximizing loss. Specifically, the CE-stSENet firstly integrates both multi-level spectral and multi-scale temporal analysis simultaneously. Hierarchical multi-domain representations are then captured in a unified manner with a variant of squeeze-and-excitation block. The classification net is finally implemented for epileptic EEG recognition based on features extracted in previous subnetworks. Particularly, to address the fact that the scarcity of seizure events results in finite data distribution and the severe overfitting problem in seizure detection, the CE-stSENet is coordinated with a maximum mean discrepancy-based information maximizing loss for mitigating the overfitting problem. Competitive experimental results on three EEG datasets against the state-of-the-art methods demonstrate the effectiveness of the proposed framework in recognizing epileptic EEGs, indicating its powerful capability in the automatic seizure detection.
Asunto(s)
Epilepsia , Procesamiento de Señales Asistido por Computador , Algoritmos , Electroencefalografía , Epilepsia/diagnóstico , Humanos , Convulsiones/diagnósticoRESUMEN
Motor imagery electroencephalography (EEG) decoding is an essential part of brain-computer interfaces (BCIs) which help motor-disabled patients to communicate with the outside world by external devices. Recently, deep learning algorithms using decomposed spectrums of EEG as inputs may omit important spatial dependencies and different temporal scale information, thus generated the poor decoding performance. In this paper, we propose an end-to-end EEG decoding framework, which employs raw multi-channel EEG as inputs, to boost decoding accuracy by the channel-projection mixed-scale convolutional neural network (CP-MixedNet) aided by amplitude-perturbation data augmentation. Specifically, the first block in CP-MixedNet is designed to learn primary spatial and temporal representations from EEG signals. The mixed-scale convolutional block is then used to capture mixed-scale temporal information, which effectively reduces the number of training parameters when expanding reception fields of the network. Finally, based on the features extracted in previous blocks, the classification block is constructed to classify EEG tasks. The experiments are implemented on two public EEG datasets (BCI competition IV 2a and High gamma dataset) to validate the effectiveness of the proposed approach compared to the state-of-the-art methods. The competitive results demonstrate that our proposed method is a promising solution to improve the decoding performance of motor imagery BCIs.
Asunto(s)
Electroencefalografía/métodos , Imaginación/fisiología , Movimiento/fisiología , Redes Neurales de la Computación , Algoritmos , Interfaces Cerebro-Computador , Ritmo Gamma , Humanos , Aprendizaje Automático , Desempeño Psicomotor/fisiología , Procesamiento de Señales Asistido por ComputadorRESUMEN
A new parametric approach is proposed for nonlinear and nonstationary system identification based on a time-varying nonlinear autoregressive with exogenous input (TV-NARX) model. The TV coefficients of the TV-NARX model are expanded using multiwavelet basis functions, and the model is thus transformed into a time-invariant regression problem. An ultra-orthogonal forward regression (UOFR) algorithm aided by mutual information (MI) is designed to identify a parsimonious model structure and estimate the associated model parameters. The UOFR-MI algorithm, which uses not only the observed data themselves but also weak derivatives of the signals, is more powerful in model structure detection. The proposed approach combining the advantages of both the basis function expansion method and the UOFR-MI algorithm is proved to be capable of tracking the change of TV parameters effectively in both numerical simulations and the real EEG data.
Asunto(s)
Ondas Encefálicas/fisiología , Simulación por Computador , Redes Neurales de la Computación , Dinámicas no Lineales , Algoritmos , Electroencefalografía , Humanos , Factores de TiempoRESUMEN
In the present study, the system of nonradioactive iodine-labeled-antibodies linking inductively coupled plasma mass spectrometry for immunoassay was reported. The goat-anti-Escherichia coli and goat anti rabbit were considered as simulant antigen and antibody respectively in order to establish a new method of immunoassay by inductively coupled plasma mass spectrometry which has the advantage of high sensitivity, low detection limit and preferable linearity range. During the experiment, the N-bromosuccinimide, a mild oxidant, was used to oxidize the non-radioactive iodine (127 I) that labeled the protein. The method of nonradioactive iodine labeled protein was established and the best labeling condition was explored. The compound of I was purified by Sephadex G50 column chromatography, then the stability and activity were examined. The results showed that the labeling program was simple, reaction time was within two minutes, the labeling yield achieved 63.12% and none of I shed from the compound after 96 hours. The simulant antigen and antibody reacted on polystyrene microtiter plate and the I was detected by ICP-MS, the detection limit of the method was 0.12 mg x L(-1), relative standard deviation (n = 9) was less than 3% and the linearly dependent coefficient was 0.998 7. This system can also be used in analysis of other protein, nucleic acid and so on.