Causal insights from clinical information in radiology: Enhancing future multimodal AI development

Open Webpage

Michael Jantscher, Felix Gunzer, Gernot Reishofer, Roman Kern

Published: 01 Jan 2025, Last Modified: 28 Jul 2025Comput. Methods Programs Biomed. 2025EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Understanding radiology data generation is essential for multimodal AI in this field.•Ignoring causal data generation risks spurious correlations and biased AI models.•We stress the need for open-source imaging data with further clinical information.