Treatment variability among low-acuity EMS patients in a university hospital ED: a retrospective registry study from Southwest Finland.

Emotion is recognized as a fundamental component of interpersonal communication [, ]. Emotion is a psychological state that reflects a core human condition and enables individuals to convey their affective disposition [, ]. Using emotions, individuals can express their mental states []. People commonly use facial expressions, hand gestures, speech, and written messages to communicate emotions []. Emotions, as a central mechanism of human communication, are important in personal, social, and professional contexts []. Therefore, emotion recognition has become a major topic in the literature []. Automatic emotion recognition and classification are mainly based on body language, facial expressions, speech, and text inputs []. However, individuals who cannot use these modalities (e.g., paralyzed individuals and patients with autism) may experience substantial difficulty in self-expression []. Recent technological advances have enabled the use of physiological signals, such as electroencephalogram (EEG) signals, for emotion classification [, ]. Traditionally, EEG has been used to support the diagnosis of neurological disorders. With the development of portable EEG devices, its use has expanded to emotion recognition []. This is because brain electrical activity changes with emotional state []. However, interpreting EEG signals remains challenging []. Manual interpretation is generally avoided because it requires specialized expertise. To address these challenges, this study includes the collection of a novel dataset designed for automatic EEG-based emotion recognition and classification. In addition, a new machine learning model was developed for emotion classification using EEG signals. Automatic emotion recognition and interpretation using EEG signals is one of the most widely studied topics in the literature [, , ]. However, the limited number of publicly available open-access datasets remains a major limitation in this research area. A summary of EEG-based automatic emotion recognition techniques is provided in Additional file 1: Table S.