Health Data Set Bias Examination in the European Health Data Space
Evangelia Anna Markatou (HL7 Europe, TU Delft - Electrical Engineering, Mathematics and Computer Science)
Catherine Chronaki (HL7 Europe)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Diversity in health datasets is a necessary, quantifiable property of inclusivity across key domains (e.g., demographic, socioeconomic, health, and environmental) that directly shapes how well research generalizes and how fair its impacts are. In its absence, interventions risk encoding bias and exacerbating health disparities. In this work, we outline how dataset diversity can be measured and how these measures can be surfaced in the EHDS via metadata. We argue that simple annotations are insufficient: structured interaction with the data owner is required to assess the utility of health datasets for specific research purposes.
help_outline