Publications by authors named "Sebastian Germer"
Article Synopsis
- The study investigates survival analysis methods for lung cancer using data from the Schleswig-Holstein Cancer Registry, comparing traditional Cox regression with newer machine learning methods such as Random Survival Forests and neural networks.
- Results indicate that the Cox Proportional Hazard model performs best when using the cancer stage classification, while the Random Survival Forests excel when considering additional tumor characteristics like size and metastasis.
- The findings highlight the importance of these models for providing insights into patient survival, aiding physicians in making better treatment decisions, and ultimately enhancing patient outcomes.
The integration of heterogeneous healthcare data sources is a necessary process to enable the secondary use valuable information in clinical research. Data integration is time-consuming for data stewards. The transformation using predefined rules for data harmonization can reduce the time-consuming and error-prone work and ease the data integration at various sites.
View Article and Find Full Text PDF