Künstliche Intelligenz in der Entscheidungsunterstützung und medikamentösen Tumortherapie

Due to the continuous increase in medical information generation, ever more factors must be simultaneously taken into account during the decision-making process. To compensate for the associated higher burden on physicians, computer-assisted solutions are increasingly being developed and evaluated.

Combination of intelligent applications with an evidence-based approach requires specific system architectures. These are presented and explained in more detail in this article.

By structuring medical information and bundling it into disease-specific knowledge bases, versatile tools for clinical decision support can be generated. An important methodological approach is the simulation of treatment outcomes based on retrospective data. Artificial intelligence (AI) methods play a fundamental role in making these processes scalable and efficient.

The combination of AI-supported solutions for information extraction, formalization, and integration allows for the development of platforms that can significantly support information processing in everyday clinical practice. The integration of additional data sources in patient monitoring can sustainably improve the existing status quo by detecting risks more quickly.

To establish effective end-to-end solutions in evidence-based clinical decision support, holistic concepts must be developed and implemented. This includes rigorous test management, but also a rapid response to current developments and findings in the respective field of application.