Evaluation of feature descriptors for representation of multimodal medical data

• Grant number: 25/17296-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: October 01, 2025
• End date: September 30, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Agma Juci Machado Traina
• Grantee: Matheus Rodrigues Ferreira
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Associated research grant: 24/13328-9 - Intelligent management of multimodal health data for decision-making in big data scenarios: IHealth-MD, AP.TEM

Abstract

This undergraduate research consists of a study to evaluate existing methods for extracting features from complex data to be instantiated for the medical domain. Specifically, descriptors for images, time series, and texts will be evaluated. To this end, open-source approaches reported in the literature will be used for analysis and evaluation, which will be performed on open data. The objectives include the use of standardized databases, the instantiation and evaluation of existing algorithms on them, and the identification (or ranking) of the evaluated approaches. The different information modalities will be evaluated separately, but the expected final result is a broad quantitative study, consisting of a systematic experimental summarization, also involving the fine-tuning of existing approaches, where applicable. (AU)