AI helps cancer patients better understand CT reports

Medical reports written in technical terminology can be challenging for patients. A team from the Technical University of Munich (TUM) has investigated how artificial intelligence can make CT findings more understandable. In the study, reading time was shortened and patients rated the automatically simplified texts as more understandable and helpful.

In order to simplify the original documents, the researchers used an open source large language model that was operated on the computers of the TUM University Hospital in compliance with data protection regulations. An example: “The cardiomediastinal silhouette is in the midline. The heart chambers are usually opacified. (…) A small pericardial effusion has been detected” was simplified by the AI ​​as follows: “Heart: The report detects a small amount of fluid around your heart. This is a common finding and your doctor will determine if it requires treatment.”