| Home > Publications database > Large language models as a communication and organizational infrastructure in urology: evidence, limitations, and clinical responsibility. [„Large language models“ als Kommunikations- und Organisationsinfrastruktur in der Urologie: Evidenz, Grenzen und klinische Verantwortung]. |
| Journal Article (Review Article) | DKFZ-2026-00805 |
; ; ; ; ;
2026
Springer Medizin
New York]
Abstract: Large language models (LLMs) mark a structural turning point in digital medicine, as they are now capable not only of processing medical information but of conveying it in ways that are context-sensitive, linguistically coherent, and tailored to the needs of specific audiences. In urology, a specialty defined by high communication demands, explanation-intensive disease entities, and complex organizational interfaces, the central question is no longer one of technical feasibility but of the fundamental repositioning of clinical communication. This article analyzes the role of LLMs as a communication and organizational infrastructure within urologic care, grounded in current empirical evidence. Systematic comparisons demonstrate that artificial intelligence (AI)-generated responses to patient questions are perceived in most studied contexts as equivalent or superior to physician responses, particularly with respect to clarity, empathy, and overall satisfaction, while revealing a pronounced divergence between patient-centered and professional evaluative standards. Urology-specific data on the automated generation of layperson summaries further show that LLMs can achieve significantly improved readability and formal quality without compromising factual accuracy. At the same time, studies on AI-assisted documentation and ambient scribe systems underscore that gains in efficiency and reductions in documentation burden remain inseparable from the need for physician oversight and institutional governance. These findings are not interpreted as a substitution of medical expertise but as the emergence of a new mediating layer between clinical knowledge, organizational requirements, and patient perception. The integration of generative AI, thus, becomes a professional and institutional challenge that extends beyond the deployment of individual tools and shapes the future communication architecture of urology.
Keyword(s): Ambient scribe system ; Digital health communication ; Electronic health records ; Generative artificial intelligence ; eHealth
; BIOSIS Previews ; Biological Abstracts ; Clarivate Analytics Master Journal List ; Current Contents - Clinical Medicine ; DEAL Springer ; DEAL Springer ; Essential Science Indicators ; IF < 5 ; JCR ; SCOPUS ; Science Citation Index Expanded ; Web of Science Core Collection
|
The record appears in these collections: |
guest ::
