EUVIMEDEUVIMED entdecken
Crossref

A Class-Imbalance Aware and Explainable Spatio-Temporal Graph Attention Network for Neonatal Seizure Detection

Khadijeh Raeisi, Mohammad Khazaei, Gabriella Tamburro, Pierpaolo Croce, Silvia Comani, Filippo Zappasodi

International Journal of Neural Systems · 2023-09 · 33 (09)

DOI: 10.1142/s0129065723500466 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Neural Systems, online erschienen am 28.07.2023. Verfasst von Khadijeh Raeisi und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neuroscience, Imaging and Clinical Sciences, Universita Gabriele d'Annunzio, Chieti 66100, Italy. Hauptthemen laut MEDLINE: Electroencephalography, Epilepsy. Die Arbeit wurde bisher 53-mal zitiert (Crossref) und nennt 54 Literaturangaben. Gefördert von: European Union, INtegrating Functional Assessment measures for Neonatal Safeguard.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Seizures are the most prevalent clinical indication of neurological disorders in neonates. In this study, a class-imbalance aware and explainable deep learning approach based on Convolutional Neural Networks (CNNs) and Graph Attention Networks (GATs) is proposed for the accurate automated detection of neonatal seizures. The proposed model integrates the temporal information of EEG signals with the spatial information on the EEG channels through the graph representation of the multi-channel EEG segments. One-dimensional CNNs are used to automatically develop a feature set that accurately represents the differences between seizure and nonseizure epochs in the time domain. By employing GAT, the attention mechanism is utilized to emphasize the critical channel pairs and information flow among brain regions. GAT coefficients were then used to empirically visualize the important regions during the seizure and nonseizure epochs, which can provide valuable insight into the location of seizures in the neonatal brain. Additionally, to tackle the severe class imbalance in the neonatal seizure dataset using under-sampling and focal loss techniques are used. Overall, the final Spatio-Temporal Graph Attention Network (ST-GAT) outperformed previous benchmarked methods with a mean AUC of 96.6% and Kappa of 0.88, demonstrating its high accuracy and potential for clinical applications.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

ElectroencephalographyEpilepsyInfant, NewbornHumansSeizuresNeural Networks, Computer

Stichwörter der Autor:innen

Convolutional Neural Networks (CNN)Electroencephalography (EEG)Graph Attention Networks (GAT)attention mechanismfocal lossneonatal seizure detection

Autor:innen und Institutionen

  • Department of Neuroscience, Imaging and Clinical Sciences, Universita Gabriele d'Annunzio, Chieti 66100, Italy.
  • Department of Neuroscience, Imaging and Clinical Sciences-Behavioral Imaging and Neural Dynamics Center, Universita Gabriele d'Annunzio, Chieti 66100, Italy.
  • Department of Neuroscience, Imaging and Clinical Sciences-Behavioral, Imaging and Neural Dynamics Center-Institute for, Advanced Biomedical Technologies, Universita Gabriele d'Annunzio, Chieti 66100, Italy.

Förderung

  • European Union
  • INtegrating Functional Assessment measures for Neonatal Safeguard

Zitieren

APA

Raeisi, K., Khazaei, M., Tamburro, G., Croce, P., Comani, S., & Zappasodi, F. (2023). A Class-Imbalance Aware and Explainable Spatio-Temporal Graph Attention Network for Neonatal Seizure Detection. International Journal of Neural Systems, 33(09). https://doi.org/10.1142/s0129065723500466

Vancouver

Raeisi K, Khazaei M, Tamburro G, Croce P, Comani S, Zappasodi F. A Class-Imbalance Aware and Explainable Spatio-Temporal Graph Attention Network for Neonatal Seizure Detection. International Journal of Neural Systems. 2023;33(09). doi:10.1142/s0129065723500466

BibTeX
@article{Raeisi2023,
  author = {Raeisi, Khadijeh and Khazaei, Mohammad and Tamburro, Gabriella and Croce, Pierpaolo and Comani, Silvia and Zappasodi, Filippo},
  title = {A Class-Imbalance Aware and Explainable Spatio-Temporal Graph Attention Network for Neonatal Seizure Detection},
  journal = {International Journal of Neural Systems},
  year = {2023},
  volume = {33},
  number = {09},
  doi = {10.1142/s0129065723500466}
}
Kostenlos starten →