Heidelberg 2022 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

ST: Fachverband Strahlen- und Medizinphysik

ST 2: Poster Session

ST 2.4: Poster

Montag, 21. März 2022, 16:15–17:30, P

Design and validation of a Digital Twin for tumor stage prediction of prostate cancer patients — •Anna-Katharina Nitschke, Carlos Andres Brandl, and Matthias Weidemüller — Physikalisches Institut, Ruprecht-Karls Universität Heidelberg, Im Neuenheimer Feld 226, 69120 Heidelberg, Germany

Digital Twins (DT) are virtual representations of physical assets. In healthcare, DT can help to personalize diagnosis and treatment for complex diseases like prostate cancer.

To be able to generate a DT which is supporting the complete clinical journey of a cancer patient, a conceptual development for combining several individual algorithms is needed. This approach also ensures best possible practices for patient data protection during collaborative work. As interpretability and uncertainty quantification are crucial in medical applications, on hast to find a solution that meets these requirements as well as clinical guidelines. Through using ensemble methods, we develop a concept of a DT, which is combining machine learning algorithms and expert systems with the medical doctor's opinion to improve decision-making. We exemplify the approach with tumor stage prediction (TNM) before treatment planning.

This work is part of the project CLINIC 5.1 and is supported by the BMWi.