Automatic Epileptic Seizure Onset-Offset Detection Based On CNN in Scalp EEG

Boonyakitanont, Poomipat; Lek-uthai, Apiwat; Songsiri, Jitkomut

doi:10.1109/icassp40776.2020.9053143

Cited by 14 publications

(18 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The counting-based approach in [BLuS20] first considered epoch-based prediction outcomes as inputs, and the inputs were either 0 or 1. In the neighborhood of the epoch i having the width 2w + 1 epochs, if there was at least one predicted positive, a seizure candidate c i was counted as one.…”

Section: Formalization Of Counting-based Methodsmentioning

confidence: 99%

“…We now formalize the above description in [BLuS20] in an analytic form as follows. The seizure candidate can be expressed by applying a filter of 1's with length 2w + 1 followed by setting one as a threshold.…”

Section: Formalization Of Counting-based Methodsmentioning

confidence: 99%

“…The total score of the candidates in the group determines the possibility that the candidate group is a seizure. We first describe a counting-based method [BLuS20] in a general form of convolution. This form is then generalized to be the ScoreNet by allowing its parameters to be optimally tuned using an optimization algorithm such as a conjugate gradient method.…”

Section: Scorenetmentioning

confidence: 99%

“…The copyright holder for this preprint this version posted December 22, 2020. ; https://doi.org/10.1101/2020.12.21.423728 doi: bioRxiv preprint short periods. Recently, a CNN model was designed to spatially and temporally capture ictal patterns in EEG epochs [BLuS20]. An additional post-processing method designed from clinical characteristics of seizures was added to recorrect false positives and false negatives in the prior classification.…”

Section: Introductionmentioning

confidence: 99%

“…Several existing models including logistic regression, SVM, and decision tree can be applied in the first stage. In the onset and offset detection, we establish a novel post-processing model called ScoreNet extending the work in [BLuS20]. The ScoreNet is a neural network-based model that automatically determines a group of seizure candidates from inputs, and then assigns a score to each individual candidate.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

ScoreNet: A Neural network-based post-processing model for identifying epileptic seizure onset and offset in EEGs

Boonyakitanont

Lek-uthai

Songsiri

2020

Preprint

Self Cite

View full text Add to dashboard Cite

This article aims to design an automatic detection algorithm of epileptic seizure onsets and offsets in scalp EEGs. A proposed scheme consists of two sequential steps: the detection of seizure episodes, and the determination of seizure onsets and offsets in long EEG recordings. We introduce a neural network-based model called ScoreNet as a post-processing technique to determine the seizure onsets and offsets in EEGs. A cost function called a log-dice loss that has an analogous meaning to F1 is proposed to handle an imbalanced data problem. In combination with several classifiers including random forest, CNN, and logistic regression, the ScoreNet is then verified on the CHB-MIT Scalp EEG database. As a result, in seizure detection, the ScoreNet can significantly improve F1 to 70.15% and can considerably reduce false positive rate per hour to 0.05 on average. In addition, we propose detection delay metric, an effective latency index as a summation of the exponential of delays, that includes undetected events into account. The index can provide a better insight into onset and offset detection than conventional time-based metrics.

show abstract

Section: Formalization Of Counting-based Methodsmentioning

confidence: 99%

Section: Formalization Of Counting-based Methodsmentioning

confidence: 99%

Section: Scorenetmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

ScoreNet: A Neural network-based post-processing model for identifying epileptic seizure onset and offset in EEGs

Boonyakitanont

Lek-uthai

Songsiri

2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Epileptic Seizure Prediction, Detection, and Localization

2021

EEG Signal Processing and Machine Learning

View full text Add to dashboard Cite

Automatic Detection of Epileptic Seizures with Recurrent and Convolutional Neural Networks

Carrión

López-Chilet

Martínez-Bernia

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Computer-aided diagnosis based on intelligent systems is an effective strategy to improve the efficiency of healthcare systems while reducing their costs. In this work, the epilepsy detection task is approached in two different ways, recurrent and convolutional neural networks, within a patient-specific scheme. Additionally, a detector function and its effects on seizure detection performance are presented. Our results suggest that it is possible to detect seizures from scalp EEGs with acceptable results for some patients, and that the DeepHealth framework is a proper deep learning software for medical research.

show abstract

Automatic Epileptic Seizure Onset-Offset Detection Based On CNN in Scalp EEG

Cited by 14 publications

References 13 publications

ScoreNet: A Neural network-based post-processing model for identifying epileptic seizure onset and offset in EEGs

ScoreNet: A Neural network-based post-processing model for identifying epileptic seizure onset and offset in EEGs

Epileptic Seizure Prediction, Detection, and Localization

Automatic Detection of Epileptic Seizures with Recurrent and Convolutional Neural Networks

Contact Info

Product

Resources

About