Thursday, 12 January 2023

Anxious brains

A study by UniTrento researchers published in Sensors

Versione stampabile

A research team from the Department of Psychology and Cognitive Science and the Centre for Medical Sciences of the University of Trento is studying brain models to predict anxiety and other emotions with the help of artificial intelligence

Trait anxiety can be predicted by observing the volume of gray and white matter in different areas of the brain. This was demonstrated by a research group from the Department of Psychology and Cognitive Science and the Center for Medical Sciences of the University of Trento. An article describing the results of their study was published days ago in the scientific journal Sensors as part of a series of works on the same subject. The study is a first step towards the development of biomarkers capable of predicting anxiety and other emotional disorders and personalized treatments.

The research work of UniTrento stands out because, for the first time, scientists created a predictive model that is capable of correctly classifying the anxiety experienced by the participants in the study. Moreover, thanks to the use of artificial intelligence, this model can predict anxiety with a certain level of accuracy even in people whose anxiety state is not known.

"By combining artificial intelligence methods (the supervised and unsupervised machine learning approaches), we try to build predictive models for the study of emotions both in normality and in pathology", Alessandro Grecucci and Teresa Baggio explained. "Our goal is to develop models that, based on morphometric features (the size and shape) of the brain, can recognize the level of dysfunctional emotions to take action promptly with more effective treatments."

"In our sample, we have observed a tendency to experience less anxiety as age increases. It is therefore important to study anxiety especially in the youngest age groups. A new study that we are conducting in collaboration with Bordeaux is in fact focused on teenagers".

"Our findings could be useful for the development of biomarkers that can predict emotional disorders in adults," they concluded.

The laboratory
The study was conducted at the Clinical and Affective Neuroscience Lab of the University of Trento, with professor Alessandro Grecucci as principal investigator, who also coordinates the international task force Artificial Intelligence Methods to Decode Emotions and Personality (AIMDEP). The laboratory of the University of Trento is focused on the study of the neural bases of emotions perception and regulation, in normality and in pathology, through precision tools such as MRI, EEG, tDCS, and artificial intelligence methods. The laboratory collaborates with other Italian universities such as Pavia, Udine, Rome and Trieste, and foreign universities such as Abhu Dhabi, Ghent, Nijmegen and Bordeaux.

About the article
The article "Anxious Brains: a Combined Data Fusion Machine Learning Approach to Predict Trait Anxiety from Morphometric Features", published in Sensors, was written by Teresa Baggio and Alessandro Grecucci with Federica Meconi and Irene Messina.

For information on the Clinical and Affective Neuroscience Lab:
The Sensors' article is available in Open Access at: