Related Experiment Video
Updated: May 27, 2026
08:05
Detection and Isolation of Cancer in Prostate Biopsies Using Stimulated Raman Histology and Artificial Intelligence
Published on: June 10, 2025
Updating risk prediction tools: a case study in prostate cancer
Donna P Ankerst1, Tim Koniarski, Yuanyuan Liang
1Department of Mathematics, Technische Universitaet Muenchen, Unit M4, Boltzmannstr 3, 85748 Garching b. Munich, Germany. ankerst@ma.tum.de
Biometrical Journal. Biometrische Zeitschrift
|November 19, 2011
Summary
Bayes rule updates cancer risk calculators using new biomarkers from external studies. This method merges data to improve prostate cancer risk prediction with new markers.
Area of Science:
- Biostatistics
- Oncology
- Medical Informatics
Background:
- Online cancer risk prediction tools aid clinical decision-making.
- Updating these tools with new biomarkers presents data integration challenges.
Purpose of the Study:
- To apply Bayes rule for updating cancer risk prediction algorithms.
- To incorporate novel biomarkers from external studies into existing risk calculators.
Main Methods:
- Utilized Bayes rule to merge data from an original study and an external case-control study.
- Updated the Prostate Cancer Prevention Trial Risk Calculator with %freePSA and [-2]proPSA markers.
- Employed state-of-the-art validation techniques using an independent dataset.
Main Results:
- Successfully updated the risk prediction tool by incorporating external biomarker data.
- Demonstrated a method for validating and evaluating updated risk prediction models.
- The updated Prostate Cancer Prevention Trial Risk Calculator incorporates new markers for improved accuracy.
Conclusions:
- Bayes rule provides a viable approach for updating cancer risk prediction tools with external data.
- This methodology enhances the utility of online risk calculators as research evolves.
- External validation is crucial for ensuring the reliability of updated prediction tools.
