Publication Details

Sentiments Analysis of fMRI using Automatically Generated Stimuli Labels under Naturalistic Paradigm

MAHRUKH Rimsha, SHAKIL Sadia and MALIK Aamir Saeed. Sentiments Analysis of fMRI using Automatically Generated Stimuli Labels under Naturalistic Paradigm. Scientific Reports, vol. 13, no. 7267, 2023, pp. 1-15. ISSN 2045-2322. Available from: https://www.nature.com/articles/s41598-023-33734-7
Czech title
Analýza sentimentů fMRI pomocí automaticky generovaných podnětů v přirozeném scénáři
Type
journal article
Language
english
Authors
Mahrukh Rimsha (IST)
Shakil Sadia, Ph.D. (DCSY FIT BUT)
Malik Aamir Saeed, Ph.D. (DCSY FIT BUT)
URL
Keywords

fMRI, natural paradigm, sentiments, automatic, machine learning

Abstract

Our emotions and sentiments are influenced by naturalistic stimuli such as the movies we watch and the songs we listen to, accompanied by changes in our brain activation. Comprehension of these brain-activation dynamics can assist in identification of any associated neurological condition such as stress and depression, leading towards making informed decision about suitable stimuli. A large number of open-access functional magnetic resonance imaging (fMRI) datasets collected under naturalistic conditions can be used for classification/prediction studies. However, these datasets do not provide emotion/sentiment labels, which limits their use in supervised learning studies. Manual labeling by subjects can generate these labels, however, this method is subjective and biased. In this study, we are proposing another approach of generating automatic labels from the naturalistic stimulus itself. We are using sentiment analyzers (VADER, TextBlob, and Flair) from natural language processing to generate labels using movie subtitles. Subtitles generated labels are used as the class labels for positive, negative, and neutral sentiments for classification of brain fMRI images. Support vector machine, random forest, decision tree, and deep neural network classifiers are used. We are getting reasonably good classification accuracy (42-84%) for imbalanced data, which is increased (55-99%) for balanced data.

Published
2023
Pages
1-15
Journal
Scientific Reports, vol. 13, no. 7267, ISSN 2045-2322
Publisher
Springer Nature Switzerland AG
DOI
UT WoS
001022553200001
EID Scopus
BibTeX
@ARTICLE{FITPUB12925,
   author = "Rimsha Mahrukh and Sadia Shakil and Saeed Aamir Malik",
   title = "Sentiments Analysis of fMRI using Automatically Generated Stimuli Labels under Naturalistic Paradigm",
   pages = "1--15",
   journal = "Scientific Reports",
   volume = 13,
   number = 7267,
   year = 2023,
   ISSN = "2045-2322",
   doi = "10.1038/s41598-023-33734-7",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12925"
}
Back to top