Application of fuzzy neural networks for parameter estimation and decision making in disease diagnosis

• Grant number: 24/03621-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2024
• End date: June 30, 2025
• Field of knowledge: Physical Sciences and Mathematics - Mathematics - Applied Mathematics
• Principal Investigator: Vinícius Francisco Wasques
• Grantee: Matheus Zaia Monteiro
• Host Institution: Centro Nacional de Pesquisa em Energia e Materiais (CNPEM). Ministério da Ciência, Tecnologia e Inovação (Brasil). Campinas , SP, Brazil

Abstract

The area of deep learning has been revolutionizing the large area of artificial intelligence (AI), through artificial neural network (ANN) models. Despite obtaining good results, ANNs lack interpretability, due to the large number of parameters in the models. In this context, there is a search for more interpretable ANNs, some of which incorporate the principles of fuzzy logic into the structure of neural networks. Fuzzy logic has the ability to deal with uncertainty data, since objects may have different degrees of association with sets, differing from classical logic. In particular, the Adaptive Neuro Fuzzy Inference Systems (ANFIS) models use fuzzy inference methods in ANN architectures to generate models with greater interpretability. This project focuses on studying ANFIS models, using different fuzzy inference methods to explore the interpretability of such systems. Finally, ANFIS systems will be applied to estimate parameters in disease modeling and to assist in medical diagnostic decision-making.