Electrophysiological brain imaging based on simulation-driven deep learning in the context of epilepsy.

Identifying the location, the spatial extent and the electrical activity of distributed brain sources in the context of epilepsy through ElectroEncephaloGraphy (EEG) recordings is a challenging task because of the highly ill-posed nature of the underlying Electrophysiological Source Imaging (ESI) problem. To guarantee a unique solution, most existing ESI methods pay more attention to solve this inverse problem by imposing physiological constraints. This paper proposes an efficient ESI approach based on simulation-driven deep learning. Epileptic High-resolution 256-channels scalp EEG (Hr-EEG) signals are simulated in a realistic manner to train the proposed patient-specific model. More particularly, a computational neural mass model developed in our team is used to generate the temporal dynamics of the activity of each dipole while the forward problem is solved using a patient-specific three-shell realistic head model and the boundary element method. A Temporal Convolutional Network (TCN) is considered in the proposed model to capture local spatial patterns. To enable the model to observe the EEG signals from different scale levels, the multi-scale strategy is leveraged to capture the overall features and fine-grain features by adjusting the convolutional kernel size. Then, the Long Short-Term Memory (LSTM) is used to extract temporal dependencies among the computed spatial features. The performance of the proposed method is evaluated through three different scenarios of realistic synthetic interictal Hr-EEG data as well as on real interictal Hr-EEG data acquired in three patients with drug-resistant partial epilepsy, during their presurgical evaluation. A performance comparison study is also conducted with two other deep learning-based methods and four classical ESI techniques. The proposed model achieved a Dipole Localization Error (DLE) of 1.39 and Normalized Hamming Distance (NHD) of 0.28 in the case of one patch with SNR of 10 dB. In the case of two uncorrelated patches with an SNR of 10 dB, obtained DLE and NHD were respectively 1.50 and 0.28. Even in the more challenging scenario of two correlated patches with an SNR of 10 dB, the proposed approach still achieved a DLE of 3.74 and an NHD of 0.43. The results obtained on simulated data demonstrate that the proposed method outperforms the existing methods for different signal-to-noise and source configurations. The good behavior of the proposed method is also confirmed on real interictal EEG data. The robustness with respect to noise makes it a promising and alternative tool to localize epileptic brain areas and to reconstruct their electrical activities from EEG signals.

Research on effective connectivity of intracerebral electroencephalogram based on Wiener-Granger Causality Index modified by generalized Akaike's Information Criterion / 生物医学工程学杂志

The objective is to deal with brain effective connectivity among epilepsy electroencephalogram (EEG) signals recorded by use of depth electrodes in the cerebral cortex of patients suffering from refractory epilepsy during their epileptic seizures. The Wiener-Granger Causality Index (WGCI) is a well-known effective measure that can be useful to detect causal relations of interdependence in these kinds of EEG signals. It is based on the linear autoregressive model, and the issue of the estimation of the model parameters plays an important role in the calculation accuracy and robustness of WGCI to do research on brain effective connectivity. Focusing on this issue, a modified Akaike's information criterion algorithm is introduced in the computation of the WGCI to estimate the orders involved in the underlying models and in order to advance the performance of WGCI to detect brain effective connectivity. Experimental results support the interesting performance of the proposed algorithm to characterize the information flow both in a linear stochastic system and a physiology-based model.

Scalp localization of human auditory cortical activity modified by GSM electromagnetic fields.

PURPOSE: This study attempted to determine whether there is a localized effect of GSM (Global System for Mobile communications) microwaves by studying the Auditory Evoked Potentials (AEP) recorded at the scalp of nine healthy subjects and six epileptic patients. MATERIALS AND METHODS: We determined the influence of GSM RadioFrequency (RF) on parameters characterizing the AEP in time or/and frequency domains. A parameter selection method using SVM (Support Vector Machines)-based criteria allowed us to estimate those most altered by the radiofrequencies. The topography of the parameter modifications was computed to determine the localization of the radiofrequency influence. A statistical test was conducted for selected scalp areas, in order to determine whether there were significant localized alterations due to the RF. RESULTS: The epileptic patients showed a lengthening of the scalp component N100 (100 ms latency) in the frontal area contralateral to the radiation, which may be due to an afferent tract alteration. For the healthy subjects, an amplitude increase of the P200 wave (200 ms latency) was identified in the frontal area. CONCLUSIONS: The present study suggests that radiofrequency fields emitted by mobile phones modify the AEP. Nevertheless, no direct link between these findings and RF-induced damages in brain function was established.

Evaluation of two computational models of amplitude modulation coding in the inferior colliculus.

Two computational models replicating amplitude-modulation encoding in the inferior colliculus (IC) are presented and compared. Neurons in this nucleus are modeled as point neurons using Mc Gregor equations, and receive depolarizing currents from action potentials delivered by stellate cells (chopper units) in the cochlear nucleus (CN). Stellate cells are modeled using modified Hodgkin-Huxley equations and receive inputs from a peripheral auditory model. The CN models of the two proposed architectures are characterized by an important dispersion of cellular characteristics, and therefore by various cellular best modulation frequencies (BMFs) ranging from 60 to 300 Hz. In contrast with the previous model proposed by [M.J. Hewitt, R. Meddis, A computer model of amplitude-modulation sensitivity of single units in the inferior colliculus, J. Acoust. Soc. Am. 95 (1994) 2145], each IC cell model receives convergent input from stellate cells with various BMFs. This approach assumes therefore minimal constraints on the model architecture and cell characteristics. The two models differ in terms of the neuronal structure of the IC, composed of 1 or 2 layers of point neurons acting as coincidence detectors. Each model is evaluated using two metrics: mean firing rate and modulation gain. Rate and temporal modulation transfer functions (r-MTFs and t-MTFs, respectively) are simulated and compared with physiological data. Simulations reveal that (i) an important dispersion of BMFs in the CN cells providing input to IC cells yields plausible IC cells responses to AM stimuli, (ii) the 2-layer IC structure yields the best approximation of IC responses measured in vivo.