Data: 5 maio
Hora: 13h
Local: Anfiteatro PA2, Pavilhão de Matemática, Campus Alameda
Orador: João Santinha (Fundação Champalimaud)
Título: “Machine learning in high-stake medical imaging tasks”
Resumo:
Medical imaging is a pillar of healthcare supporting screening, diagnosis, treatment planning and assessment, as well as follow-up. However, due to changes in acquisition systems, data shortage caused by privacy concerns, among other, developed machine learning models often lack generalizability. In this talk we will see how the invariance property of causal structures may be used to improve generalizability, covering application in tabular data obtained from images, deep learning approaches and the use of generative AI to promote such several approaches may be used for tabular data.
Nota biográfica:
João Santinha, PhD, is a researcher specializing in medical imaging, biomarkers, and artificial intelligence. He co-leads the Digital Surgery Lab. Dr. Santinha completed his PhD at Instituto de Telecomunicações – Instituto Superior Técnico, University of Lisbon. He is also an Invited Assistant Professor at the Faculty of Medicine of the University of Lisbon.
His work focuses on developing AI-driven tools to enhance surgical precision and personalize screening, diagnosis, treatments and follow-up through imaging and non-imaging data.
A Priberam integra a Comunidade IST Spin-Off®.
Os “Priberam Machine Learning Lunch Seminars” são de entrada livre, mediante inscrição prévia.
