Enhanced ROCKET for the Automated Detection of Epileptic Tonic-Clonic Seizures Using Accelerometer Data
-
Budapest University of Economics and Business, Budapest, Hungary
buza.krisztian@uni-bge.hu (corresponding author) -
BioIntelligence Group, Department of Mathematics-Informatics, Faculty of Technical and Human Sciences, Sapientia Hungarian University of Transylvania, Targu Mures, Romania
buza@biointelligence.hu -
Business Informatics, Baden-Wuerttemberg Cooperative State University, Mosbach, Germany
alexandros.nanopoulos@gmail.com -
Institute of Genomic Medicine and Rare Disorders, Semmelweis University, Budapest, Hungary
noemiagnesvarga@gmail.com
Abstract
The detection of epileptic tonic-clonic seizures during everyday life based on accelerometric data from wearable devices would enhance the diagnostic and the follow-up of the epileptic patients. We develop an algorithm which may contribute to recognition of tonic-clonic epileptic seizure based on accelerometer data that can be collected from mobile and wearable devices. We consider this task to be a multivariate time-series classification problem. State-of-the art solutions to this problem are based on machine learning techniques, such as Random Convolutional Kernel Transform (ROCKET). We enhance ROCKET by replacing standard convolution with dynamic convolution. Dynamic convolution was originally defined for univariate time series, therefore, we extend it to multivariate time series. We perform experiments on two publicly available real-world datasets related to tonic-clonic seizures. The experimental results show that the proposed enhancements of the ROCKET algorithm significantly reduce the average classification error. Moreover, our approach outperforms other time series classifiers, including several types of deep neural networks that are commonly used in the domain of time-series classification. An enhanced version of the ROCKET algorithm is proposed for the automated detection of epileptic tonic-clonic seizures using accelerometer data. To assist reproducibility and follow-up works, we made our implementation publicly available at https://github.com/kr7seizure .
Key words
Epilepsy, Multivariate Time Series Classification, Human Activity Recognition, Random Convolutional Kernel Transform, ROCKET, Dynamic Convolution
Digital Object Identifier (DOI)
https://doi.org/10.2298/CSIS250913020B
Publication information
Volume 23, Issue 2 (April 2026)
Year of Publication: 2026
ISSN: 2406-1018 (Online)
Publisher: ComSIS Consortium
Full text
Available in PDF
Portable Document Format
How to cite
Buza, K., Nanopoulos, A., Varga, N.Á.: Enhanced ROCKET for the Automated Detection of Epileptic Tonic-Clonic Seizures Using Accelerometer Data*. Computer Science and Information Systems, 23(2), 757–774 (2026). https://doi.org/10.2298/CSIS250913020B
Journal's Facebook page