Effective EEG Feature Selection for Interpretable MDD (Major Depressive Disorder) Classification

Simple item page
dc.contributor.authorMrázek, Vojtěchcs
dc.contributor.authorJawed, Soyibacs
dc.contributor.authorArif, Muhammadcs
dc.contributor.authorMalik, Aamir Saeedcs
dc.date.issued2023-04-14cs
dc.description.abstractIn this paper, we propose an interpretable electroencephalogram (EEG)-based solution for the diagnostics of major depressive disorder (MDD). The acquisition of EEG experimental data involved 32 MDD patients and 29 healthy controls. A feature matrix is constructed involving frequency decomposition of EEG data based on power spectrum density (PSD) using the Welch method. Those PSD features were selected, which were statistically significant. To improve interpretability, the best features are first selected from feature space via the non-dominated sorting genetic (NSGA-II) evolutionary algorithm. The best features are utilized for support vector machine (SVM), and k-nearest neighbors (k-NN) classifiers, and the results are then correlated with features to improve the interpretability. The results show that the features (gamma bands) extracted from the left temporal brain regions can distinguish MDD patients from control significantly. The proposed best solution by NSGA-II gives an average sensitivity of 93.3%, specificity of 93.4% and accuracy of 93.5%. The complete framework is published as open-source at https://github.com/ehw-fit/eeg-mdd.en
dc.description.abstractIn this paper, we propose an interpretable electroencephalogram (EEG)-based solution for the diagnostics of major depressive disorder (MDD). The acquisition of EEG experimental data involved 32 MDD patients and 29 healthy controls. A feature matrix is constructed involving frequency decomposition of EEG data based on power spectrum density (PSD) using the Welch method. Those PSD features were selected, which were statistically significant. To improve interpretability, the best features are first selected from feature space via the non-dominated sorting genetic (NSGA-II) evolutionary algorithm. The best features are utilized for support vector machine (SVM), and k-nearest neighbors (k-NN) classifiers, and the results are then correlated with features to improve the interpretability. The results show that the features (gamma bands) extracted from the left temporal brain regions can distinguish MDD patients from control significantly. The proposed best solution by NSGA-II gives an average sensitivity of 93.3%, specificity of 93.4% and accuracy of 93.5%. The complete framework is published as open-source at https://github.com/ehw-fit/eeg-mdd.en
dc.formattextcs
dc.format.extent1427-1435cs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationGECCO 2023 - Proceedings of the 2023 Genetic and Evolutionary Computation Conference. 2023, p. 1427-1435.en
dc.identifier.doi10.1145/3583131.3590398cs
dc.identifier.isbn979-8-4007-0119-1cs
dc.identifier.orcid0000-0002-9399-9313cs
dc.identifier.orcid0000-0003-1085-3157cs
dc.identifier.other185129cs
dc.identifier.researcheridAAF-8828-2019cs
dc.identifier.researcheridC-6904-2009cs
dc.identifier.scopus56559922700cs
dc.identifier.scopus12800348400cs
dc.identifier.urihttp://hdl.handle.net/11012/244320
dc.language.isoencs
dc.publisherAssociation for Computing Machinerycs
dc.relation.ispartofGECCO 2023 - Proceedings of the 2023 Genetic and Evolutionary Computation Conferencecs
dc.relation.urihttps://dl.acm.org/doi/10.1145/3583131.3590398cs
dc.rightsCreative Commons Attribution 4.0 Internationalcs
dc.rights.accessopenAccesscs
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/cs
dc.subjectelectroencephalogram (EEG)en
dc.subjectfeature extractionen
dc.subjectmajor depressive disorderen
dc.subjectelectroencephalogram (EEG)
dc.subjectfeature extraction
dc.subjectmajor depressive disorder
dc.titleEffective EEG Feature Selection for Interpretable MDD (Major Depressive Disorder) Classificationen
dc.title.alternativeEffective EEG Feature Selection for Interpretable MDD (Major Depressive Disorder) Classificationen
dc.type.driverconferenceObjecten
dc.type.statusPeer-revieweden
dc.type.versionpublishedVersionen
sync.item.dbidVAV-185129en
sync.item.dbtypeVAVen
sync.item.insts2025.10.14 14:13:24en
sync.item.modts2025.10.14 10:28:37en
thesis.grantorVysoké učení technické v Brně. Fakulta informačních technologií. Ústav počítačových systémůcs
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
3583131.3590398.pdf
Size:
908.58 KB
Format:
Adobe Portable Document Format
Description:
file 3583131.3590398.pdf
Download
Collections
Ústav počítačových systémů