Multimodal Domain Adaptation and Deep Learning for Medical Image Analysis

• Grant number: 24/22853-0
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: January 01, 2026
• End date: December 31, 2028
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: João Paulo Papa
• Grantee: João Renato Ribeiro Manesco
• Host Institution: Faculdade de Ciências (FC). Universidade Estadual Paulista (UNESP). Campus de Bauru. Bauru , SP, Brazil
• Company: Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC)
• Associated research grant: 23/14427-8 - Data Science for Smart Industry (CDII), AP.PCPE

Abstract

Advancements in medical imaging technologies have revolutionized clinical diagnostics, providing modalities such as X-rays, computed tomography (CT), and magnetic resonance imaging (MRI), each offering unique perspectives on human anatomy. However, the diversity of sources, acquisition protocols, and patient populations results in misaligned data and discrepancies in imaging patterns, posing significant challenges for deep learning models. These models, although promising, often fail to generalize across data from different origins due to domain shifts, limiting their applicability in real-world clinical settings. To address these challenges, domain adaptation emerges as an effective approach, enabling knowledge transfer across datasets. Simultaneously, multimodal analysis, which integrates complementary information from distinct modalities such as X-rays and CT, holds the potential to enrich model representations. This project proposes to combine these approaches by using multimodal domain adaptation techniques to enhance the performance of deep learning models. The research aims to correlate observable structures in CT scans with artifacts in X-ray images, improving tasks such as medical device detection and disease classification while fostering robustness and generalization in clinical scenarios. (AU)