ESTRO 2025 - Abstract Book

S3801

Physics - Radiomics, functional and biological imaging and outcome prediction

ESTRO 2025

Conclusion: This work explores how the inherent noise/uncertainty present in the dataset affects the predictive model, highlighting the need for uncertainty-aware methods in machine learning-based outcome prediction to ensure reliable predictions and treatment decisions.

Keywords: Robust machine learning, outcome, data uncertainty

References: 1. Goodman, S. G. et al. The diagnostic and prognostic impact of the redefinition of acute myocardial infarction: lessons from the Global Registry of Acute Coronary Events (GRACE). Am Heart J 151, 654–660 (2006). 2. Mövik, L., Bäck, A. & Pettersson, N. Impact of delineation errors on the estimated organ at risk dose and of dose errors on the normal tissue complication probability model. Med Phys 50, 1879–1892 (2023). 3. Lazzarino, A. I. & Mindell, J. S. Measuring high-sensitivity cardiac troponin T blood concentration in population surveys. PLoS One 12, (2017). 4. Goodfellow, I. J., Shlens, J. & Szegedy, C. EXPLAINING AND HARNESSING ADVERSARIAL EXAMPLES.