Related Experiment Video
Updated: Sep 15, 2025
06:37
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
Published on: December 15, 2023
4.1K
Decoding basic emotional states through integration of an fNIRS-based brain-computer interface with supervised
Ayşenur Eser1, Sinem Burcu Erdoğan1
1Faculty of Engineering and Natural Sciences, Department of Biomedical Engineering, Acıbadem Mehmet Ali Aydınlar University, İstanbul, Türkiye.
Plos One
|July 14, 2025
Summary
Brain-computer interfaces using functional near-infrared spectroscopy (fNIRS) accurately detect emotions. This fNIRS-BCI system shows high performance for real-time emotion identification in clinical and daily applications.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Affective Computing
Background:
- Brain-computer interfaces (BCIs) offer personalized user experiences.
- Automated emotion detection is crucial for mental health and adaptive systems.
- Functional near-infrared spectroscopy (fNIRS) is a non-invasive neuroimaging technique.
Purpose of the Study:
- To assess the feasibility of an fNIRS-based BCI for identifying basic emotional states (positive, negative, neutral).
- To evaluate the efficacy of fNIRS signals in predicting emotional valence from standardized stimuli.
- To determine the classification performance of machine learning algorithms using fNIRS data.
Main Methods:
- Collected fNIRS data from 20 healthy participants viewing International Affective Picture System (IAPS) images.
- Recorded hemodynamic responses in prefrontal cortical (PFC) regions using a 22-channel system.
- Extracted 20 time-domain features from HbO signals and applied kNN, Ensemble (Subspace kNN), and SVM algorithms for classification.
Main Results:
- Three-class classification (positive, neutral, negative) achieved >90% accuracy, sensitivity, specificity, F-1 score, and precision.
- Two-class classification accuracy exceeded 93% for all performance metrics.
- High classification performance indicates fNIRS-BCI's potential for objective emotion detection.
Conclusions:
- fNIRS-based BCIs demonstrate high accuracy in identifying fundamental emotional states.
- The system shows promise for real-time emotion detection in clinical and daily life applications.
- fNIRS-BCIs offer practicality and low computational complexity for personalized user experiences.
