New research from the Sohn Lab, recently featured by RSNA News, found that large language model-generated, patient-friendly summaries of lung cancer screening CT reports improved comprehension, reduced anxiety, and increased participants’ willingness to wait for a scheduled follow-up appointment. The benefits were greatest among individuals reporting lower English and health literacy, suggesting that carefully evaluated AI tools may help make radiology communication more accessible and equitable. The study, led by Juan Serna, MD and published in Radiology Advances in February 2026, is available here.