Semantic-hierarchical model improves classification of spoken-word evoked electrocorticography

Youngmin Na; Inyong Choi; Dong Pyo Jang; Joong Koo Kang; Jihwan Woo

doi:10.1016/j.jneumeth.2018.10.034

Back

Semantic-hierarchical model improves classification of spoken-word evoked electrocorticography

Journal article

Peer reviewed

Semantic-hierarchical model improves classification of spoken-word evoked electrocorticography

Youngmin Na, Inyong Choi, Dong Pyo Jang, Joong Koo Kang and Jihwan Woo

Journal of neuroscience methods, Vol.311, pp.253-258

01/01/2019

DOI: 10.1016/j.jneumeth.2018.10.034

PMID: 30389490

View Online

Abstract

•Semantic hierarchical model better classifies cortical responses to spoken words.•Classification accuracy reveals critical brain areas in spoken word representation.•Spectrotemporal features yield effective and accurate decoding of cortical signals. Classification of spoken word-evoked potentials is useful for both neuroscientific and clinical applications including brain-computer interfaces (BCIs). By evaluating whether adopting a biology-based structure improves a classifier’s accuracy, we can investigate the importance of such structure in human brain circuitry, and advance BCI performance. In this study, we propose a semantic-hierarchical structure for classifying spoken word-evoked cortical responses. The proposed structure decodes the semantic grouping of the words first (e.g., a body part vs. a number) and then decodes which exact word was heard. The proposed classifier structure exhibited a consistent ∼10% improvement of classification accuracy when compared with a non-hierarchical structure. Our result provides a tool for investigating the neural representation of semantic hierarchy and the acoustic properties of spoken words in human brains. Our results suggest an improved algorithm for BCIs operated by decoding heard, and possibly imagined, words.

Semantic hierarchical structure

Brain computer interface

Decoding words

Electrocorticography

Details

Title: Subtitle: Semantic-hierarchical model improves classification of spoken-word evoked electrocorticography
Creators: Youngmin Na - Department of Biomedical Engineering, University of Ulsan, Ulsan, Republic of Korea
Inyong Choi - Department of Communication Sciences and Disorders, University of Iowa, Iowa City, IA, USA
Dong Pyo Jang - Department of Biomedical Engineering, Hanyang University, Seoul, Republic of Korea
Joong Koo Kang - LTG Neuro Medical Center, Seoul, Republic of Korea
Jihwan Woo - Department of Biomedical Engineering, University of Ulsan, Ulsan, Republic of Korea
Resource Type: Journal article
Publication Details: Journal of neuroscience methods, Vol.311, pp.253-258
DOI: 10.1016/j.jneumeth.2018.10.034
PMID: 30389490
NLM abbreviation: J Neurosci Methods
ISSN: 0165-0270
eISSN: 1872-678X
Publisher: Elsevier B.V
Grant note: DOI: 10.13039/501100002568, name: University of Ulsan
Language: English
Date published: 01/01/2019
Academic Unit: Communication Sciences and Disorders; Iowa Neuroscience Institute; Neurosurgery; Otolaryngology
Record Identifier: 9984002364402771

Metrics

21 Record Views

2 Times Cited - Web of Science