Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer

Qinyu Chen; Congyi Sun; Chang Gao; Shih Chii Liu

doi:10.1109/ISCAS58744.2024.10558341

Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer

Qinyu Chen^*, Congyi Sun, Chang Gao^*, Shih Chii Liu

^*Corresponding author for this work

Electronics

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

7 Downloads (Pure)

Abstract

Epilepsy is a common disease of the nervous system. Timely prediction of seizures and intervention treatment can significantly reduce the accidental injury of patients and protect the life and health of patients. This paper presents a tiny neuromorphic Spiking Convolutional Transformer, named Spiking Conformer, to detect and predict epileptic seizure segments from scalped long-term electroencephalogram (EEG) recordings. We report evaluation results from the Spiking Conformer model using the Boston Children's Hospital-MIT (CHB-MIT) EEG dataset. By leveraging spike-based addition operations, the Spiking Conformer significantly reduces the classification computational cost compared to the non-spiking model. Additionally, we introduce an approximate spiking neuron layer to further reduce spike-triggered neuron updates by nearly 38% without sacrificing accuracy. Using raw EEG data as input, the proposed Spiking Conformer achieved an average sensitivity rate of 94.9% and a specificity rate of 99.3% for the seizure detection task, and 96.8%, 89.5% for the seizure prediction task, and needs >10x fewer operations compared to the non-spiking equivalent model.

Original language	English
Title of host publication	ISCAS 2024 - IEEE International Symposium on Circuits and Systems
Publisher	IEEE
Number of pages	5
ISBN (Electronic)	9798350330991
DOIs	https://doi.org/10.1109/ISCAS58744.2024.10558341
Publication status	Published - 2024
Event	2024 IEEE International Symposium on Circuits and Systems, ISCAS 2024 - Singapore, Singapore Duration: 19 May 2024 → 22 May 2024

Publication series

Name	Proceedings - IEEE International Symposium on Circuits and Systems
ISSN (Print)	0271-4310

Conference

Conference	2024 IEEE International Symposium on Circuits and Systems, ISCAS 2024
Country/Territory	Singapore
City	Singapore
Period	19/05/24 → 22/05/24

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

EEG data
epilepsy seizure detection
epilepsy seizure prediction
spiking neural networks
Transformer

Access to Document

10.1109/ISCAS58744.2024.10558341

Epilepsy_Seizure_Detection_and_Prediction_using_an_Approximate_Spiking_Convolutional_TransformerFinal published version, 1.41 MB

Cite this

@inproceedings{6b3ace0c33454302b05c612fe628857c,

title = "Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer",

abstract = "Epilepsy is a common disease of the nervous system. Timely prediction of seizures and intervention treatment can significantly reduce the accidental injury of patients and protect the life and health of patients. This paper presents a tiny neuromorphic Spiking Convolutional Transformer, named Spiking Conformer, to detect and predict epileptic seizure segments from scalped long-term electroencephalogram (EEG) recordings. We report evaluation results from the Spiking Conformer model using the Boston Children's Hospital-MIT (CHB-MIT) EEG dataset. By leveraging spike-based addition operations, the Spiking Conformer significantly reduces the classification computational cost compared to the non-spiking model. Additionally, we introduce an approximate spiking neuron layer to further reduce spike-triggered neuron updates by nearly 38% without sacrificing accuracy. Using raw EEG data as input, the proposed Spiking Conformer achieved an average sensitivity rate of 94.9% and a specificity rate of 99.3% for the seizure detection task, and 96.8%, 89.5% for the seizure prediction task, and needs >10x fewer operations compared to the non-spiking equivalent model.",

keywords = "EEG data, epilepsy seizure detection, epilepsy seizure prediction, spiking neural networks, Transformer",

author = "Qinyu Chen and Congyi Sun and Chang Gao and Liu, {Shih Chii}",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public. ; 2024 IEEE International Symposium on Circuits and Systems, ISCAS 2024 ; Conference date: 19-05-2024 Through 22-05-2024",

year = "2024",

doi = "10.1109/ISCAS58744.2024.10558341",

language = "English",

series = "Proceedings - IEEE International Symposium on Circuits and Systems",

publisher = "IEEE",

booktitle = "ISCAS 2024 - IEEE International Symposium on Circuits and Systems",

address = "United States",

}

Chen, Q, Sun, C, Gao, C & Liu, SC 2024, Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer. in ISCAS 2024 - IEEE International Symposium on Circuits and Systems. Proceedings - IEEE International Symposium on Circuits and Systems, IEEE, 2024 IEEE International Symposium on Circuits and Systems, ISCAS 2024, Singapore, Singapore, 19/05/24. https://doi.org/10.1109/ISCAS58744.2024.10558341

Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer. / Chen, Qinyu; Sun, Congyi; Gao, Chang et al.
ISCAS 2024 - IEEE International Symposium on Circuits and Systems. IEEE, 2024. (Proceedings - IEEE International Symposium on Circuits and Systems).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer

AU - Chen, Qinyu

AU - Sun, Congyi

AU - Gao, Chang

AU - Liu, Shih Chii

N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2024

Y1 - 2024

N2 - Epilepsy is a common disease of the nervous system. Timely prediction of seizures and intervention treatment can significantly reduce the accidental injury of patients and protect the life and health of patients. This paper presents a tiny neuromorphic Spiking Convolutional Transformer, named Spiking Conformer, to detect and predict epileptic seizure segments from scalped long-term electroencephalogram (EEG) recordings. We report evaluation results from the Spiking Conformer model using the Boston Children's Hospital-MIT (CHB-MIT) EEG dataset. By leveraging spike-based addition operations, the Spiking Conformer significantly reduces the classification computational cost compared to the non-spiking model. Additionally, we introduce an approximate spiking neuron layer to further reduce spike-triggered neuron updates by nearly 38% without sacrificing accuracy. Using raw EEG data as input, the proposed Spiking Conformer achieved an average sensitivity rate of 94.9% and a specificity rate of 99.3% for the seizure detection task, and 96.8%, 89.5% for the seizure prediction task, and needs >10x fewer operations compared to the non-spiking equivalent model.

AB - Epilepsy is a common disease of the nervous system. Timely prediction of seizures and intervention treatment can significantly reduce the accidental injury of patients and protect the life and health of patients. This paper presents a tiny neuromorphic Spiking Convolutional Transformer, named Spiking Conformer, to detect and predict epileptic seizure segments from scalped long-term electroencephalogram (EEG) recordings. We report evaluation results from the Spiking Conformer model using the Boston Children's Hospital-MIT (CHB-MIT) EEG dataset. By leveraging spike-based addition operations, the Spiking Conformer significantly reduces the classification computational cost compared to the non-spiking model. Additionally, we introduce an approximate spiking neuron layer to further reduce spike-triggered neuron updates by nearly 38% without sacrificing accuracy. Using raw EEG data as input, the proposed Spiking Conformer achieved an average sensitivity rate of 94.9% and a specificity rate of 99.3% for the seizure detection task, and 96.8%, 89.5% for the seizure prediction task, and needs >10x fewer operations compared to the non-spiking equivalent model.

KW - EEG data

KW - epilepsy seizure detection

KW - epilepsy seizure prediction

KW - spiking neural networks

KW - Transformer

UR - http://www.scopus.com/inward/record.url?scp=85198551786&partnerID=8YFLogxK

U2 - 10.1109/ISCAS58744.2024.10558341

DO - 10.1109/ISCAS58744.2024.10558341

M3 - Conference contribution

AN - SCOPUS:85198551786

T3 - Proceedings - IEEE International Symposium on Circuits and Systems

BT - ISCAS 2024 - IEEE International Symposium on Circuits and Systems

PB - IEEE

T2 - 2024 IEEE International Symposium on Circuits and Systems, ISCAS 2024

Y2 - 19 May 2024 through 22 May 2024

ER -

Epilepsy Seizure Detection and Prediction using an Approximate Spiking Convolutional Transformer

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this