Banca de DEFESA: WILZA OLIVEIRA DOS SANTOS

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : WILZA OLIVEIRA DOS SANTOS
DATE: 27/06/2024
TIME: 14:00
LOCAL: Remota
TITLE:

MUSICAL HUMAN-MACHINE INTERFACES FOR EMOTION RECOGNITION IN ELECTROENCEPHALOGRAPHIC SIGNALS AS SUPPORT FOR MUSIC THERAPY

KEY WORDS:

Alzheimer's Disease, Affective Computing, Music Therapy, Dementia, Cognition, Artificial Intelligence, Electroencephalography.

PAGES: 23
BIG AREA: Engenharias
AREA: Engenharia Biomédica
SUMMARY:

The aging Brazilian population, driven by declining birth rates and rising life expectancy, has led to a surge in age-related diseases, particularly dementias like Alzheimer's disease and stroke. In this context, music therapy emerges as a promising tool to combat the effects of these conditions, demonstrating its ability to slow dementia progression through musical stimuli and musical education. Patient interaction with music promotes the stimulation of brain areas related to memory, using emotions as a means of activation. However, the effectiveness of music therapy crucially depends on the therapist's ability to accurately recognize and elicit patient emotions. To address this challenge, this research proposes the development of a brain-computer music interface (BCMI) based on deep artificial neural networks (ANNs) and evolutionary algorithms, with the aim of recognizing patient emotions from electroencephalographic (EEG) signals and voice, enabling the personalization of musical stimuli in music therapy for elderly people. This master's dissertation aimed to develop a robust model for emotion recognition in elderly people, using EEG and voice signals. The results obtained in this research demonstrate that the general objective was fully achieved. The collected database is one of the largest and most complete in the field of emotion recognition in elderly people to date, and the developed model presented high performance in emotion classification, with accuracy values above 99%. The selection of attributes using PSO contributed to the reduction of the model's complexity and the improvement of its performance. The model developed in this dissertation has the potential to be applied in different contexts, such as elderly care, human-machine interaction, and scientific research. The use of the model can assist in identifying mood changes in elderly people, evaluating the impact of therapeutic interventions, and developing more intuitive and personalized interfaces for interaction with this population. Implementing the BCMI in music therapy for elderly people has the potential to transform the treatment of patients with dementia and other age-related diseases. Through accurate recognition of patient emotions, the BCMI enables the personalization of musical stimuli, optimizing the effectiveness of music therapy and promoting a more individualized and effective therapeutic experience. In addition, the research contributes to the advancement of knowledge about the relationships between music, emotions, and the human brain in elderly people, paving the way for new applications in music therapy and other health domains focused on this population.

COMMITTEE MEMBERS:
Interno - 1807632 - WELLINGTON PINHEIRO DOS SANTOS
Externa ao Programa - 2727505 - GISELLE MACHADO MAGALHAES MORENO - UFPEExterna ao Programa - ***.618.664-** - JULIANA CARNEIRO GOMES - UFPE