Carlos Sáez

Carlos Sáez is an Associate Professor and Biomedical Data Science Researcher at Universitat Politècnica de València, Spain. He is affiliated with the Biomedical Data Science Lab (bdslab-upv) where he conducts research on biomedical data analysis and quality assessment.

His research interests focus on Biomedical Data Science, particularly in the areas of Electronic Health Records analysis, data quality assessment, and handling dataset shifts across multiple data sources. His work addresses challenges in biomedical data variability, with applications in public health and clinical informatics, developing metrics like Source Probabilistic Outlyingness (SPO) and Global Probabilistic Deviation (GPD) for measuring data heterogeneity.

Dr. Sáez has contributed to developing several open-source tools for biomedical data analysis, including the EHRsourceVariability package for measuring and visualizing biomedical data heterogeneity across data sources, the EHRtemporalVariability package for delineating temporal dataset shifts, and the trustworthy-ai framework for developing trustworthy health AI systems with tutorial notebooks.

• Notable Publications: His work has been published in Statistical Methods in Medical Research and the Journal of the American Medical Informatics Association
• Research Impact: His methodologies help detect anomalous data behaviors, identify similar data source groups, and assess concordance among different biomedical data repositories