I'm Dilermando Queiroz,
a Research Engineer and PhD candidate building AI systems that ensure fairness in medical imaging foundation models
matter?
Medical AI systems trained on biased datasets can yield disparate diagnostic accuracy across demographic subgroups, disproportionately affecting underrepresented populations. My research develops foundation models with fairness-aware training strategies to achieve equitable performance across patient demographics.
→ See my perspective paperExperiences
FairMI Project
Developing fairness frameworks for medical imaging foundation models (FairMI project, FAPESP-SNSF). Using self-supervised learning and distributed training on HPC. Published at MICCAI 2024 and ECCV 2024. Brazil-Switzerland collaboration.
→ See my publications
IBM Research
Contributed to IBM-NASA Prithvi geospatial foundation model. Co-authored 2 publications (JAMES, AIES 2022) and filed 2 patents on parallel computing and AI-driven forecasting.
→ See my publications
Mathematics and Chemistry
Developed Python library for Geometric Algebra applications in computational chemistry. Work with molecular distance geometry problems using Nuclear Magnetic Resonance data.
→ See my libraryLatest Papers
Fair Foundation Models for Medical Image Analysis: Challenges and Perspectives
Dilermando Queiroz, Anderson Carlos, André Anjos and Lilian Berton
→ Read paperDoes Data-Efficient Generalization Exacerbate Bias in Foundation Models?
Dilermando Queiroz, Anderson Carlos, Maíra Fatoretto, Luis Filipe Nakayama, André Anjos and Lilian Berton
→ Read paperUsing Backbone Foundation Model for Evaluating Fairness in Chest Radiography Without Demographic Data
Dilermando Queiroz, André Anjos and Lilian Berton
→ Read paperDirect Sampling for Spatially Variable Extreme Event Generation in Resampling-Based Stochastic Weather Generators
Jorge Guevara, Maria Garcia, Priscilla Avegliano, Debora Lima, Dilermando Queiroz, Maysa Macedo, Leonardo Tizzei, Daniela Szwarcman, Bianca Zadrozny, Campbell Watson and Anne Jones
→ Read paperPractical Skills Demand Forecasting via Representation Learning of Temporal Dynamics
Maysa Macedo, Wyatt Clarke, Eli Lucherini, Tyler Baldwin, Dilermando Queiroz, Rogerio Abreu de Paula, and Subhro Das
→ Read paper