Banca de DEFESA: FELIPE RICARDO DOS SANTOS FERNANDES

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
STUDENT : FELIPE RICARDO DOS SANTOS FERNANDES
DATE: 31/08/2023
TIME: 09:00
LOCAL: Laboratório de Inovação Tecnológica em Saúde (LAIS)/UFRN
TITLE:

DIGITAL HEALTH SOLUTION FOR ALTERNATIVE COMMUNICATION FOR PEOPLE WITH AMYOTROPHIC LATERAL SCLEROSIS

 

KEY WORDS:

Computer Vision; Artificial intelligence; Machine Learning; Image Processing; Motor Neurone Disease; Neurodegenerative Diseases; Public health

PAGES: 58
BIG AREA: Engenharias
AREA: Engenharia Biomédica
SUMMARY:

Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disease that irreversibly impairs an individual's motor system and functional abilities, even causing the progressive loss of communication skills and autonomy. Technological resources based on Augmentative and Alternative Communication, Computer Vision, and Machine Learning are essential for developing digital health solutions to enable the communicative process and autonomy that, consequently, promote improvements in the quality of life and survival of people with ALS. Focused on a Human-Computer Interaction (HCI) approach based on images of the eyes from a simple camera not mounted on the body, this work presents an assistive technology resource for Augmentative and Alternative Communication for people with ALS. The approach proposed in this work consists of an algorithmic model capable of recognizing the state of the eye (open or closed) in real-time and interoperating with Autonomus, a digital health solution designed by the Laboratory of Technological Innovation in Health at the Federal University of Rio Grande do Norte (LAIS/UFRN) for the communication of people with ALS. The model consists of four methodological processes: (i) image acquisition; (ii) Face detection; (iii) eye detection; and (iv) classifying the state of the eye, which is the foremost step for Human-Computer Interaction. An algorithmic study with a control group was conducted to evaluate the model's overall performance and the Convolutional Neural Network (CNN) classification ability. The results related to the proposed model for classifying the state of the eye in real-time are promising and reach significant values of accuracy and f1-score above 92%. The results also point to the viability of developing low-cost assistive technology resources that guarantee universal access, health promotion, well-being, and reduced inequalities, which go beyond improvements in the communicative process of people with ALS. Therefore, the object of study of this work is also to enable and promotes the exercise of rights, citizenship, fundamental freedoms, and health care for people with ALS.

COMMITTEE MEMBERS:
Presidente - 2488270 - RICARDO ALEXSANDRO DE MEDEIROS VALENTIM
Interno - 2524053 - ANTONIO LUIZ PEREIRA DE SIQUEIRA CAMPOS
Externo ao Programa - 1510735 - DANILO ALVES PINTO NAGEM - UFRNExterno ao Programa - 3966965 - ERNANO ARRAIS JUNIOR - UFRNExterna ao Programa - 2562782 - KARILANY DANTAS COUTINHO - UFRNExterno à Instituição - ANTONIO HIGOR FREIRE DE MORAIS - IFRN
Externo à Instituição - CRISTINE MARTINS GOMES DE GUSMÃO - UFPE