TY - GEN
T1 - EEG-Based Classification of Olfactory Response to Pleasant Stimuli
AU - Abbasi, Nida Itrat
AU - Bose, Rohit
AU - Bezerianos, Anastasios
AU - Thakor, Nitish V.
AU - Dragomir, Andrei
N1 - Funding Information:
This work was supported by BMRC SPF grant number APG2013/085 from Procter and Gamble, Singapore and ASTAR, Singapore.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/7
Y1 - 2019/7
N2 - Olfactory perception involves complex processing distributed along several cortical and sub-cortical regions in the brain. Although several studies have shown that the power spectra of the electroencephalography (EEG) contain information that can be used to differentiate between pleasant and unpleasant stimuli, there are still no studies which investigate whether EEG can be used to differentiate between the neural responses to olfactory stimuli of different levels of pleasantness. For this purpose, in the present study, local brain information within established frequency bands (θ, α and γ) has been used to devise discriminative features in a classification approach. A comparative study of four widely used classifiers is presented and SVM gives the best performance (accuracy = 75.71%). The results reveal that is it possible to objectively discriminate using EEG spectral features between fine levels of perceived pleasantness using the SVM-based classifier within a cross-validation procedure.
AB - Olfactory perception involves complex processing distributed along several cortical and sub-cortical regions in the brain. Although several studies have shown that the power spectra of the electroencephalography (EEG) contain information that can be used to differentiate between pleasant and unpleasant stimuli, there are still no studies which investigate whether EEG can be used to differentiate between the neural responses to olfactory stimuli of different levels of pleasantness. For this purpose, in the present study, local brain information within established frequency bands (θ, α and γ) has been used to devise discriminative features in a classification approach. A comparative study of four widely used classifiers is presented and SVM gives the best performance (accuracy = 75.71%). The results reveal that is it possible to objectively discriminate using EEG spectral features between fine levels of perceived pleasantness using the SVM-based classifier within a cross-validation procedure.
UR - http://www.scopus.com/inward/record.url?scp=85077897461&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85077897461&partnerID=8YFLogxK
U2 - 10.1109/EMBC.2019.8857673
DO - 10.1109/EMBC.2019.8857673
M3 - Conference contribution
C2 - 31947020
AN - SCOPUS:85077897461
T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
SP - 5160
EP - 5163
BT - 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019
Y2 - 23 July 2019 through 27 July 2019
ER -