Cellular Neural Networks (Cnn) With Linear Weight Functions for a Prediction of Epileptic Seizures

Tetzlaff, Ronald; Kunz, R.; Niederhofer, C.

doi:10.1142/s0129065703001790

Cited by 6 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, CNN has been further developed in the medical field including seizure, brain coma, imagination, etc. [18]- [20]. However, the application of deep learning is seldom studied on anesthesia.…”

Section: Introductionmentioning

confidence: 99%

Spectrum Analysis of EEG Signals Using CNN to Model Patient’s Consciousness Level Based on Anesthesiologists’ Experience

et al. 2019

View full text Add to dashboard Cite

One of the most challenging predictive data analysis efforts is an accurate prediction of depth of anesthesia (DOA) indicators which has attracted growing attention since it provides patients a safe surgical environment in case of secondary damage caused by intraoperative awareness or brain injury. However, many researchers put heavily handcraft feature extraction or carefully tailored feature engineering to each patient to achieve very high sensitivity and low false prediction rate for a particular dataset. This limits the benefit of the proposed approaches if a different dataset is used. Recently, representations learned using the deep convolutional neural network (CNN) for object recognition are becoming a widely used model of the processing hierarchy in the human visual system. The correspondence between models and brain signals that holds the acquired activity at high temporal resolution has been explored less exhaustively. In this paper, deep learning CNN with a range of different architectures is designed for identifying related activities from raw electroencephalography (EEG). Specifically, an improved short-time Fourier transform is used to stand for the time-frequency information after extracting the spectral images of the original EEG as input to CNN. Then CNN models are designed and trained to predict the DOA levels from EEG spectrum without handcrafted features, which presents an intuitive mapping process with high efficiency and reliability. As a result, the best trained CNN model achieved an accuracy of 93.50%, interpreted as CNN's deep learning to approximate the DOA by senior anesthesiologists, which highlights the potential of deep CNN combined with advanced visualization techniques for EEG-based brain mapping.INDEX TERMS Depth of anesthesia, convolutional neural network, electroencephalography, short-time Fourier transform.The associate editor coordinating the review of this manuscript and approving it for publication was Shaojun Wang. anesthesia, which brings clinical safety hazards to patients during surgery [3]. Thus, scientists have been looking for parameters that characterize the DOA from medical signals, so that anesthetic drugs can be used more accurately for achieving anesthesia. However, from which the study of electroencephalogram (EEG) parameters is the most effective [4]-[6], it is still non-standard and no any best solution so far.Recently, EEG-based DOA assessment method has been rapidly developed. With the reason that general anesthesia makes the brain's conscious activity disappear mainly

show abstract

Section: Introductionmentioning

confidence: 99%

Spectrum Analysis of EEG Signals Using CNN to Model Patient’s Consciousness Level Based on Anesthesiologists’ Experience

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The scalp EEG shows a drastic increase in amplitude and shows sharp wave, spike wave or spike (or sharp) slow wave complex during ictal state, and thus is considered as a preferred tool to detect the seizures (Iasemidis et al 2003;Benbadis and Allen Hauser 2000). However, the accurate diagnosis of epilepsy that is based on the visual inspection of continuous temporal EEG waveform usually requires several days of EEG monitoring, and is tedious, time-consuming and prone to human error (Tetzlaff et al 2003;Medvedev et al 2011;Martis et al 2013;Donos et al 2015;Hortal et al 2016;Yuan et al 2018). Therefore, we need to propose a new, more discriminative feature for the diagnosis of epilepsy.…”

Section: Introductionmentioning

confidence: 99%

Variation of functional brain connectivity in epileptic seizures: an EEG analysis with cross-frequency phase synchronization

Zhu

Cai

et al. 2019

Cogn Neurodyn

View full text Add to dashboard Cite

Frequency coupling in nervous system is believed to be associated with normal and impaired brain functions. However, most of the existing experiments have been concentrated on the coupling strength within frequency bands, while the coupling strength between different bands is ignored. In this work, we apply phase synchronization index (PSI) to investigate the cross-frequency coupling (CFC) of Electroencephalogram (EEG) signals. The PSI matrixes for the multichannel EEG signals are calculated from interictal to ictal period in each sliding time window. The results show that CFC changes obviously once seizure occurs between the different bands, and such alteration is earlier than the appearance of clinical symptoms in seizure. Considering the similar role of the within-frequency coupling (WFC), we further reconstruct multi-layered brain networks, including CFC networks and WFC networks. The graph metrics are applied to investigate the variation of network structure of the epileptic brain. Significant decreases/increases of the local/global efficiency are found in d-b, d-a, h-a and d-h bands from the CFC network, while WFC network shows a significant decline in the local efficiency in h and a bands. These findings suggest that CFC may provide a new perspective to observe the alteration of brain structure when seizure occurs, and the investigation of functional connectivity across the full frequency spectrum can give us a deeper understanding of epileptic brains.

show abstract

“…Previous studies [5], [6], [8]- [11], [13], [15], [16] have shown that interesting results are obtained by the application of algorithms based on Cellular Nonlinear Networks [2]- [4] and Volterra-Systems [14]. Especially, distinct changes of the relative mean square error prior to epileptic seizures could be observed in a EEG-signal [17] prediction.…”

Section: Introductionmentioning

confidence: 99%

Dynamics of EEG-signals in epilepsy: Spatio temporal analysis by Cellular Nonlinear Networks

Niederhofer

Gollas

Tetzlaff

2007

2007 18th European Conference on Circuit Theory and Design

View full text Add to dashboard Cite

Meanwhile, numerous publications address the feature extraction problem in epilepsy. Up to now a precursor detection based on changes of EEG-signal features could not be performed with a sufficient sensitivity and specifity for an automated seizure warning system. Different approaches including procedures using stochastic models, as well as algorithms based on Cellular Nonlinear Networks (CNN) and Volterra-Systems have been discussed throughout previous publications. Therin interesting findings have been discussed involving e.g. signal prediction algorithms and the calculation of synchronisation measures. In this contribution new results obtained in a spatio temporal linear prediction of segmented electrode signals using long-term SEEG and ECoG recordings of patients in epilepsy will be discussed in detail.

show abstract

Cellular Neural Networks (Cnn) With Linear Weight Functions for a Prediction of Epileptic Seizures

Cited by 6 publications

References 2 publications

Spectrum Analysis of EEG Signals Using CNN to Model Patient’s Consciousness Level Based on Anesthesiologists’ Experience

Spectrum Analysis of EEG Signals Using CNN to Model Patient’s Consciousness Level Based on Anesthesiologists’ Experience

Variation of functional brain connectivity in epileptic seizures: an EEG analysis with cross-frequency phase synchronization

Dynamics of EEG-signals in epilepsy: Spatio temporal analysis by Cellular Nonlinear Networks

Contact Info

Product

Resources

About