Using time-dependent neural networks for EEG classification.

E Haselsteiner; G Pfurtscheller

doi:10.1109/86.895948

Home/Archive/PubMed

PubMed · 11204036

Using time-dependent neural networks for EEG classification.

E Haselsteiner·G Pfurtscheller

View original source ↗Export metadata ↓

Abstract

This paper compares two different topologies of neural networks. They are used to classify single trial electroencephalograph (EEG) data from a brain-computer interface (BCI). A short introduction to time series classification is given, and the used classifiers are described. Standard multilayer perceptrons (MLPs) are used as a standard method for classification. They are compared to finite impulse response (FIR) MLPs, which use FIR filters instead of static weights to allow temporal processing inside the classifier. A theoretical comparison of the two architectures is presented. The results of a BCI experiment with three different subjects are given and discussed. These results demonstrate the higher performance of the FIR MLP compared with the standard MLP.

Explore related subjects

Cerebral Cortex Electroencephalography Image Processing, Computer-Assisted Neural Networks, Computer User-Computer Interface Comparative Study Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

E Haselsteiner, G Pfurtscheller. 2000. Using time-dependent neural networks for EEG classification.. https://doi.org/10.1109/86.895948

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2026

Histone H3K9 methyltransferases regulate cortical growth by coordinating heterochromatin formation and neural progenitor dynamics.

DNA packaging into heterochromatin is a fundamental mechanism of transcriptional silencing, yet its role in regulating neural progenitor behavior during brain development remains poorly understood. Trimethylation of histone H3 lysine 9 (H3K9me3), catalyzed by the methyltransferases SETDB1, SUV39H1, and SUV39H2, is a defining feature of heterochromatin, but functional redundancy among these enzymes has obscured their developmental roles. Here, we generated a cortex-specific triple knockout mouse model lacking Setdb1, Suv39h1, and Suv39h2 to directly interrogate H3K9me3 function during corticogenesis. Combined loss of H3K9 methyltransferases caused genome-wide depletion of H3K9me3, disruption of neural progenitor cell-cycle progression, and impaired cortical neurogenesis, resulting in microcephaly. H3K9 methyltransferases preserve neural progenitor identity and function by silencing clustered protocadherins, meiosis-associated genes, and a cell-cycle restraint program through H3K9me3 deposition. Loss of H3K9me3 promoted local chromatin opening and increased transcription factor occupancy, enabling transposable elements to acquire cryptic enhancer activity and modulate proximal gene expression. Together, these findings establish H3K9me3 heterochromatin as an active regulator of neural progenitor dynamics and lineage fidelity, revealing a central epigenetic mechanism that restricts aberrant transcriptional programs to ensure cortical growth.

Cerebral Cortex↗

PubMed2003

Retrospective diagnosis of congenital cytomegalovirus infection and cortical maldevelopment.

Congenital cytomegalovirus (CMV) infection can cause malformations of cortical development (MCD). It is difficult to establish CMV as a cause of MCD several months postpartum. This can now be done by detection of CMV DNA in dried blood spots (DBS test) on Guthrie cards. The authors used DBS tests to assess 10 patients with MCD of unknown cause. Four of the 10 patients were positive for CMV.

Cerebral Cortex↗

PubMed2003

Inability of Lyapunov exponents to predict epileptic seizures.

It has been claimed that Lyapunov exponents computed from electroencephalogram or electrocorticogram (ECoG) time series are useful for early prediction of epileptic seizures. We show, by utilizing a paradigmatic chaotic system, that there are two major obstacles that can fundamentally hinder the predictive power of Lyapunov exponents computed from time series: finite-time statistical fluctuations and noise. A case study with an ECoG signal recorded from a patient with epilepsy is presented.

Cerebral Cortex↗