GEM-CRAP: a fusion architecture for focal seizure detection

Abstract Background Identification of seizures is essential for the treatment of epilepsy. Current machine-learning and deep-learning models often perform well on public datasets when classifying generalized seizures with prominent features. However, their performance was less effective in detecting...

Full description

Saved in:
Bibliographic Details
Main Authors: Jianwei Shi, Yuanyuan Zhang, Ziang Song, Hang Xu, Yanfeng Yang, Lei Jin, Hengxin Dong, Zhaoying Li, Penghu Wei, Yongzhi Shan, Guoguang Zhao
Format: Article
Language:English
Published: BMC 2025-04-01
Series:Journal of Translational Medicine
Online Access:https://doi.org/10.1186/s12967-025-06414-5
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849734616721653760
author Jianwei Shi
Yuanyuan Zhang
Ziang Song
Hang Xu
Yanfeng Yang
Lei Jin
Hengxin Dong
Zhaoying Li
Penghu Wei
Yongzhi Shan
Guoguang Zhao
author_facet Jianwei Shi
Yuanyuan Zhang
Ziang Song
Hang Xu
Yanfeng Yang
Lei Jin
Hengxin Dong
Zhaoying Li
Penghu Wei
Yongzhi Shan
Guoguang Zhao
author_sort Jianwei Shi
collection DOAJ
description Abstract Background Identification of seizures is essential for the treatment of epilepsy. Current machine-learning and deep-learning models often perform well on public datasets when classifying generalized seizures with prominent features. However, their performance was less effective in detecting brief, localized seizures. These seizure-like patterns can be masked by fixed brain rhythms. Methods Our study proposes a supervised multilayer hybrid model called GEM-CRAP (gradient-enhanced modulation with CNN-RES, attention-like, and pre-policy networks), with three parallel feature extraction channels: a CNN-RES module, an amplitude-aware channel with attention-like mechanisms, and an LSTM-based pre-policy layer integrated into the recurrent neural network. The model was trained on the Xuanwu Hospital and HUP iEEG dataset, including intracranial, cortical, and stereotactic EEG data from 83 patients, covering over 8500 labeled electrode channels for hybrid classification (wakefulness and sleep). A post-SVM network was used for secondary training on channels with classification accuracy below 80%. We introduced an average channel deviation rate metric to assess seizure detection accuracy. Results For public datasets, the model achieved over 97% accuracy for intracranial and cortical EEG sequences in patients, and over 95% for mixed sequences, with deviations below 5%. In the Xuanwu Hospital dataset, it maintained over 94% accuracy for wakefulness seizures and around 90% during sleep. SVM secondary training improved average channel accuracy by over 10%. Additionally, a strong positive correlation was found between channel accuracy distribution and the temporal distribution of seizure states. Conclusions GEM-CRAP enhances focal epilepsy detection through adaptive adjustments and attention mechanisms, achieving higher precision and robustness in complex signal environments. Beyond improving seizure interval detection, it excels in identifying and analyzing specific epileptic waveforms, such as high-frequency oscillations. This advancement may pave the way for more precise epilepsy diagnostics and provide a suitable artificial intelligence algorithm for closed-loop neurostimulation.
format Article
id doaj-art-d2249fe42bf34de18fda33f77bc2b6c6
institution DOAJ
issn 1479-5876
language English
publishDate 2025-04-01
publisher BMC
record_format Article
series Journal of Translational Medicine
spelling doaj-art-d2249fe42bf34de18fda33f77bc2b6c62025-08-20T03:07:44ZengBMCJournal of Translational Medicine1479-58762025-04-0123112110.1186/s12967-025-06414-5GEM-CRAP: a fusion architecture for focal seizure detectionJianwei Shi0Yuanyuan Zhang1Ziang Song2Hang Xu3Yanfeng Yang4Lei Jin5Hengxin Dong6Zhaoying Li7Penghu Wei8Yongzhi Shan9Guoguang Zhao10Department of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityDepartment of Neurosurgery, Xuanwu Hospital, Capital Medical UniversityAbstract Background Identification of seizures is essential for the treatment of epilepsy. Current machine-learning and deep-learning models often perform well on public datasets when classifying generalized seizures with prominent features. However, their performance was less effective in detecting brief, localized seizures. These seizure-like patterns can be masked by fixed brain rhythms. Methods Our study proposes a supervised multilayer hybrid model called GEM-CRAP (gradient-enhanced modulation with CNN-RES, attention-like, and pre-policy networks), with three parallel feature extraction channels: a CNN-RES module, an amplitude-aware channel with attention-like mechanisms, and an LSTM-based pre-policy layer integrated into the recurrent neural network. The model was trained on the Xuanwu Hospital and HUP iEEG dataset, including intracranial, cortical, and stereotactic EEG data from 83 patients, covering over 8500 labeled electrode channels for hybrid classification (wakefulness and sleep). A post-SVM network was used for secondary training on channels with classification accuracy below 80%. We introduced an average channel deviation rate metric to assess seizure detection accuracy. Results For public datasets, the model achieved over 97% accuracy for intracranial and cortical EEG sequences in patients, and over 95% for mixed sequences, with deviations below 5%. In the Xuanwu Hospital dataset, it maintained over 94% accuracy for wakefulness seizures and around 90% during sleep. SVM secondary training improved average channel accuracy by over 10%. Additionally, a strong positive correlation was found between channel accuracy distribution and the temporal distribution of seizure states. Conclusions GEM-CRAP enhances focal epilepsy detection through adaptive adjustments and attention mechanisms, achieving higher precision and robustness in complex signal environments. Beyond improving seizure interval detection, it excels in identifying and analyzing specific epileptic waveforms, such as high-frequency oscillations. This advancement may pave the way for more precise epilepsy diagnostics and provide a suitable artificial intelligence algorithm for closed-loop neurostimulation.https://doi.org/10.1186/s12967-025-06414-5
spellingShingle Jianwei Shi
Yuanyuan Zhang
Ziang Song
Hang Xu
Yanfeng Yang
Lei Jin
Hengxin Dong
Zhaoying Li
Penghu Wei
Yongzhi Shan
Guoguang Zhao
GEM-CRAP: a fusion architecture for focal seizure detection
Journal of Translational Medicine
title GEM-CRAP: a fusion architecture for focal seizure detection
title_full GEM-CRAP: a fusion architecture for focal seizure detection
title_fullStr GEM-CRAP: a fusion architecture for focal seizure detection
title_full_unstemmed GEM-CRAP: a fusion architecture for focal seizure detection
title_short GEM-CRAP: a fusion architecture for focal seizure detection
title_sort gem crap a fusion architecture for focal seizure detection
url https://doi.org/10.1186/s12967-025-06414-5
work_keys_str_mv AT jianweishi gemcrapafusionarchitectureforfocalseizuredetection
AT yuanyuanzhang gemcrapafusionarchitectureforfocalseizuredetection
AT ziangsong gemcrapafusionarchitectureforfocalseizuredetection
AT hangxu gemcrapafusionarchitectureforfocalseizuredetection
AT yanfengyang gemcrapafusionarchitectureforfocalseizuredetection
AT leijin gemcrapafusionarchitectureforfocalseizuredetection
AT hengxindong gemcrapafusionarchitectureforfocalseizuredetection
AT zhaoyingli gemcrapafusionarchitectureforfocalseizuredetection
AT penghuwei gemcrapafusionarchitectureforfocalseizuredetection
AT yongzhishan gemcrapafusionarchitectureforfocalseizuredetection
AT guoguangzhao gemcrapafusionarchitectureforfocalseizuredetection