Researchers have unveiled an artificial intelligence system designed to identify fundamental human emotions based on facial expressions, capturing expressions that may have gone unnoticed by doctors.
Facial expressions reflect an individual’s emotional state and are considered a crucial element in psychotherapeutic treatment to understand someone’s feelings at a particular moment.
The researchers from the Computer Science Center at the University of Basel in Switzerland devised an artificial intelligence system capable of recognizing six basic emotions: happiness, surprise, anger, disgust, sadness, and fear, using over 30,000 facial expression images.
The system relied on the analysis of video clips related to psychotherapy sessions for 23 patients experiencing psychological disorders.
Martin Stepan, a psychologist at the university, stated that the new system was fed with over 950 hours of video recordings related to psychiatric patients, and the results of the experiment were astonishing.
The system’s analysis results aligned with the conclusions reached by three doctors.
Moreover, the system managed to detect specific facial expressions within fractions of a second that went unnoticed by doctors, such as a fleeting smile or a gesture expressing disgust.
Researchers believe that this new artificial intelligence system has successfully captured facial expressions as a highly credible method for measuring emotions, making it a valuable tool for research and therapy.
