Project

Research Project Title : A Pilot analysIs of behavioRal and opeRational data for detEcting Socio-emotional PrEcursors of mild CogniTIVE impairments (MCI) and demEntia (IRRESPECTIVE)

Principal Investigator: Prof. Anna Esposito

Università degli Studi della Campania “Luigi Vanvitelli”

Brief description

IRRESPECTIVE aims at developing a multi-modal, multi-episode and multi-tasks approach for dementia’s automatic, personalized detection and assistance, by exploiting data (video-recordings of targeted tasks), which describe the ability of individuals to exhibit and process the above-listed social signals.

The Project will define adequate tasks/scenarios to assess initial manifestations of peculiar changes in social cognition and behaviors in people in the prodromal or early stages of dementia. The features describing
these behavioral changes will be compared with classical dementia cognitive tests to assess their robustness in describing prodromal and early stages of dementia.

Automatic supervised/unsupervised artificial intelligent methods will be developed to detect such features from naturalistic data and devise early risk prediction, prevention and intervention protocols helping human experts to detect and discriminate the prodromes of these neurodegenerative diseases suggesting personalized care, thus resulting in better-quality individual healthcare services and healthcare outcomes

Objectives

• To construct a set of video recordings of social exchanges in the healthy population, in order to have a reference sample for further investigation in individuals with cognitive decline.
• Investigate the impaired aspects of social cognition (social perception, empathy, Theory of Mind, emotional competence and regulation, social behavior) in pathological populations, resulting from a comparison with the healthy reference population in order to identify significant differences in terms of socio-emotional behaviors.
• Integrate the knowledge acquired with the instrumental evidence (results of cognitive tests and the standard diagnostic procedure) of the aforementioned patient populations in order to improve the accuracy of the diagnosis.
• Identify artificial intelligence and signal processing methods – supervised and unsupervised – that best capture such changes and that range from the analysis of video recordings and the operational signals collected to the automation of their detection and classification.