Loading...
Please wait, while we are loading the content...
Similar Documents
Functional Link Network with Genetic Algorithm for Evoked Potentials
| Content Provider | Semantic Scholar |
|---|---|
| Author | Lin, Bor-Shyh Lin, Bor-Shing Chien, Jen-Chien Chong, Fok-Ching |
| Copyright Year | 2005 |
| Abstract | Evoked potentials (EP) are electrical responses of the brain caused by external events, such as sounds, lights or electrical stimuli. The changes of latencies and amplitudes of evoked potentials from stimuli provide important information for various neurological disorders. However, evoked potentials are usually heavily buried in ongoing electroencephalogram (EEG). Even their signal to noise ratio (SNR) is less than -5 dB to -20 dB. [1]. In general, ensemble averaging method is used to extract evoked potentials from measured electrical activities of the brain. It is well-known that the ensemble averaging method requires a lot of trials to extract evoked potentials and is easy to smooth out any information under the variation of evoked potentials. Recently, many approaches were investigated to extract evoked potentials. One of these is adaptive filtering techniques [2]-[10]. This technique is simple to design and is different from Wiener filter and Kalman filter on that it need not priori statistic knowledge of interesting signals and noises [2]. Adaptive filtering technique for extracting evoked potentials was first implemented by Thakor in 1987 [3]. For adaptive filtering techniques, providing a suitable reference signal is important. However, in practice, it is difficult for extracting evoked potentials because the measurement of correlated EEG noises is hard. Therefore, several structures with different sources of reference signals were proposed [4]-[8]. Vaz and Thakor proposed the use of a finite number of sine and cosine waves as reference signals to estimate models of evoked potentials in time domain [4]. Laguna et al used a unit impulse sequence synchronized with the beginning of each recurrence as reference signals [5]. Madhavan employed adaptive line enhancement (ALE) which uses the delay version of desired signals as reference signals, to extract evoked potentials [6]. Since the structure of adaptive line enhancement is effective for signal enhancement ABSTRACT |
| File Format | PDF HTM / HTML |
| Language | English |
| Access Restriction | Open |
| Subject Keyword | Acoustic Evoked Brain Stem Potentials Adaptive filter Dirac delta function Electroencephalography Phase Synchronization Ensemble averaging (machine learning) Evoked Potentials Expectation propagation Experiment FLNA wt Allele Genetic algorithm Kalman filter Less Than Light Short Interspersed Nucleotide Elements Signal-to-noise ratio Simulation Statistic (data) Wiener filter nervous system disorder |
| Content Type | Text |
| Resource Type | Article |
