Investigating the impact of CNN depth on neonatal seizure detection performance

Loading...
Thumbnail Image
Files
5925.pdf(422.5 KB)
Accepted Version
Date
2018-10-29
Authors
O'Shea, Alison
Lightbody, Gordon
Boylan, Geraldine B.
Temko, Andriy
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Research Projects
Organizational Units
Journal Issue
Abstract
This study presents a novel, deep, fully convolutional architecture which is optimized for the task of EEG-based neonatal seizure detection. Architectures of different depths were designed and tested; varying network depth impacts convolutional receptive fields and the corresponding learned feature complexity. Two deep convolutional networks are compared with a shallow SVMbased neonatal seizure detector, which relies on the extraction of hand-crafted features. On a large clinical dataset, of over 800 hours of multichannel unedited EEG, containing 1389 seizure events, the deep 11-layer architecture significantly outperforms the shallower architectures, improving the AUC90 from 82.6% to 86.8%. Combining the end-to-end deep architecture with the feature-based shallow SVM further improves the AUC90 to 87.6%. The fusion of classifiers of different depths gives greatly improved performance and reduced variability, making the combined classifier more clinically reliable.
Description
Keywords
Pediatrics , Support vector machines , Feature extraction , Convolution , Electroencephalography , Filter banks , Task analysis
Citation
O'Shea, A., Lightbody, G., Boylan, G. and Temko, A. (2018) 'Investigating the impact of CNN depth on neonatal seizure detection performance', 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Honolulu, HI, USA, 18-21 July, pp. 5862-5865. doi: 10.1109/EMBC.2018.8513617
Copyright
© 2018, IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.