Banca de QUALIFICAÇÃO: RAYONARA MEDEIROS DE AZEVEDO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : RAYONARA MEDEIROS DE AZEVEDO
DATE: 18/12/2025
TIME: 09:00
LOCAL: GOOGLE MEET
TITLE:

Prediction of Cervical Cytopathological Changes Using Artificial Intelligence in Primary Health Care

KEY WORDS:

Artificial intelligence; Cervical neoplasms; Machine learning.

PAGES: 11
BIG AREA: Ciências da Saúde
AREA: Enfermagem
SUMMARY:

Cervical cancer remains a significant public health challenge in Brazil, particularly due to the
low effectiveness of traditional cytopathological screening and regional inequalities. The
incorporation of more sensitive methods, such as HPV testing and emerging technologies like
Artificial Intelligence, emerges as a promising strategy to improve early detection. In this
context, machine learning models can optimize care in Primary Health Care (PHC), enhancing
diagnostic accuracy and reducing unnecessary referrals. Therefore, this study aims to develop
a machine learning algorithm for cervical cancer prediction based on cytopathological,
sociodemographic, and clinical data from patients followed in PHC. This is a cross-sectional,
multicenter study with a quantitative approach. Data collection will involve structured
interviews, cytopathological examination, and obtaining standardized cervical samples from
1,182 women receiving care at PHC units and university teaching clinics in Rio Grande do
Norte, from June to December 2025. Following collection, data will be organized, coded, and
standardized, generating different machine learning models that will be trained and compared
through cross-validation. Performance evaluation will be conducted using metrics such as
accuracy, sensitivity, and specificity. The final model will be selected considering the balance
between performance and generalization capacity, complemented by interpretability techniques
to enable clinical understanding of identified patterns. This study was approved by the Research
Ethics Committee of UFRN, under opinion number 7.296.331. As contributions, the research
may reduce inequalities in access to early cervical cancer diagnosis, optimize care flow, support
faster clinical decisions, and inform public policies aligned with the global goal of cervical
cancer elimination. Additionally, it offers a methodological model applicable to other low- and
middle-income contexts, representing a relevant technological advancement for nursing and
healthcare in Brazil.

 

COMMITTEE MEMBERS:
Presidente - 1054749 - MARIA ISABEL DA CONCEICAO DIAS FERNANDES
Interna - 3149773 - DANDARA NAYARA AZEVEDO DANTAS
Externa ao Programa - 3995493 - MARCELLY SANTOS COSSI - UFRNExterno ao Programa - ***.192.524-** - MARQUIONY MARQUES DOS SANTOS - UERN