Detalhes bibliográficos
| Ano de defesa: |
2022 |
| Autor(a) principal: |
MENDES, Jean Pablo Marques
 |
| Orientador(a): |
TELES, Ariel Soares
|
| Banca de defesa: |
TELES, Ariel Soares
,
SILVA, Francisco José da Silva e
,
COUTINHO, Luciano Reis
,
RODRIGUES, Joel José Puga Coelho
|
| Tipo de documento: |
Dissertação
|
| Tipo de acesso: |
Acesso aberto |
| Idioma: |
por |
| Instituição de defesa: |
Universidade Federal do Maranhão
|
| Programa de Pós-Graduação: |
PROGRAMA DE PÓS-GRADUAÇÃO EM CIÊNCIA DA COMPUTAÇÃO/CCET
|
| Departamento: |
DEPARTAMENTO DE INFORMÁTICA/CCET
|
| País: |
Brasil
|
| Palavras-chave em Português: |
|
| Palavras-chave em Inglês: |
|
| Área do conhecimento CNPq: |
|
| Link de acesso: |
https://tedebc.ufma.br/jspui/handle/tede/4312
|
Resumo: |
Mental health disorders have a high prevalence in the world population. With the coron- avirus (COVID-19) pandemic, there was a requirement for social distancing, aggravating problems related to mental health and well-being. The proliferation of smartphones presents opportunities for data collection to study human behavior and health. They have been used in mental health studies, as they have several embbeded sensors that can capture measurements in people’s daily lives. Traditionally, mental disorders are diagnosed by mental health professionals (e.g., psychiatrists, psychologists) on the basis of symptoms identified from patient interviews and self-reported experiences. However, patients often resort to events that occurred days, weeks or months ago, which can compromise the diag- nosis, due to memory and desirability biases. To mitigate these biases, digital phenotyping appears, collecting data passively (without direct user interaction) using mobile devices. This work aims to present a framework to facilitate the development of digital phenotyping mobile applications, called OpenDP. The proposed solution is extensible and reusable, as it allows the inclusion of modules for processing raw context data, and can be applied to different mental disorders and to the monitoring of human well-being. In addition, the framework was developed using the middleware M-Hub/CDDL to collect data from sensors (physical and virtual) and distribute them among the internal components of the framework and with the broker external. Case studies were conducted to demonstrate that the proposed solution was able to compose digital phenotypes from the inference of high-level information generated by the data processing modules. It was also demonstrated the ability of the mobile solution to add data processing modules (plugins). By aiming at developing a mobile solution for mobile devices that does not cause a great impact on energy consumption, an experimental evaluation was carried out analyzing the impact on energy consumption of the smartphone user. The results were satisfactory, showing through the experimental evaluation that the battery consumption was small. |
