Artificial Intelligence for Improved Infectious Diseases Outcomes in Kidney Transplant Recipients (AIIDKIT)

• Grant number: 24/04761-0
• Support Opportunities: Regular Research Grants
• Start date: February 01, 2025
• End date: January 31, 2029
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: José Fernando Rodrigues Júnior
• Grantee: José Fernando Rodrigues Júnior
• Principal researcher abroad: Douglas Henrique Teodoro
• Institution abroad: Université de Genève, Switzerland
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Associated researchers: Adriano Kamimura Suzuki ; Christian van Delden ; Luís Gustavo Modelli de Andrade
• Associated scholarship(s): 25/02134-1 - Systematization of an Artificial Intelligence system for Improved Infectious Diseases Outcomes in Kidney Transplant Recipients, BP.PD

Abstract

The collaborative research project on Artificial Intelligence for Improved Infectious Diseases Outcomes in Kidney Transplant Recipients (AIIDKIT) represents an initiative that leverages extensive kidney transplant datasets from Brazil and Switzerland to propel predictive modeling capabilities in the realm of transplantation medicine. By capitalizing on the wealth of data originating from diverse patient populations and healthcare systems, the project endeavors to innovate on the prediction of infection risks in kidney transplant recipients through the application of advanced computational techniques based on deep learning. This approach not only aims to enhance the accuracy of prognostic assessments but also to provide personalized insights into individual patient trajectories by integrating longitudinal transplant cohort data and supplementary contextual information. The collaboration of a multidisciplinary team comprised of researchers from the University of São Paulo (USP), Paulista State University (UNESP), University of Geneva, and University Hospitals of Geneva underscores a commitment to advancing AI-driven healthcare. Through data analysis, model refinement, and the exploration of novel data representations, the project seeks to establish a predictive framework that can adapt to the intricacies of kidney transplantation dynamics. By modeling the transplant trajectory data as input to large language models and as graph structures to investigate graph attention networks, the project aims to provide personalized insights into patient outcomes and infection susceptibility. The goal is to guide decision-making regarding prophylaxis treatments and immunosuppression therapy. By fostering the exchange of expertise and resources between the Brazilian and Swiss research institutions, AIIDKIT aims to not only address the immediate challenges in infectious diseases post-kidney transplantation but also to lay the foundation for AI-based innovations in the field of transplantation medicine. (AU)