Automated Interpretation of Pap Smear Tests for Cervical Cancer Screening

• Grant number: 24/20690-6
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: January 01, 2025
• End date: December 31, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Walmir Matos Caminhas
• Grantee: Leonardo Augusto Ferreira
• Host Institution: Instituto de Ciências Exatas (ICEx). Universidade Federal de Minas Gerais (UFMG). Ministério da Educação (Brasil). Belo Horizonte , SP, Brazil
• Company: Universidade Federal de Minas Gerais (UFMG). Instituto de Ciências Exatas (ICEx)
• Associated research grant: 20/09866-4 - Artificial Intelligence Innovation Center for Health (CIIA-Health), AP.PCPE

Abstract

The project "Automated Interpretation of the Pap Smear Test for Cervical Cancer Screening" aims to train a machine learning model to identify and explain cancerous cells. This initiative aligns with the primary objectives of the Center for Innovation in Artificial Intelligence in Health (CIIA-Health). This innovation center brings together various scientific fields, in addition to nine higher education institutions from the Southeast, South, and North regions of Brazil.In addition to training machine learning models, this work also incorporates XAI (Explainable Artificial Intelligence) strategies, which will be used to enhance the understanding of the model used for cell classification, highlighting the critical areas utilized in diagnosing cells as cancerous or normal. The goal of XAI is to demystify the model to facilitate its acceptance and to introduce new features that healthcare professionals can leverage to generate new insights.The project's methodology is well-structured, and the planning includes well-defined phases to yield effective results. Initially, a literature review will be conducted to cover the state of the art in image identification in healthcare. Following this stage, an image database of slides obtained from Pap smear tests will be created, adhering to the standardized classifications of INCA and the Bethesda system. In the final phase, machine learning models will be trained using this image database.