ESTRO 2022 - Abstract Book

S1558

Abstract book

ESTRO 2022

Conclusion The observed delineation variation is clearly reflected in the variation of ADC measurements. This indicates that consistency of ADC measurements between centres may be improved by reducing delineation variation, e.g. by using delineation guidelines or computer-aided delineation.

Poster (digital): Radiomics, modelling and statistical methods

PO-1755 Machine Learning in NTCP prediction --- A superior alternative to the Lyman-Burman-Kutcher model

P. Samant 1,2 , T. Maughan 2 , F. Van Den Heuvel 3,2 , R. Canters 4 , F. Hoebers 5 , E. Hall 6 , C. Nutting 7 , D. de Ruysscher 8

1 Oxford University Hospitals NHS Foundation Trust, Radiotherapy Physics, Oxford, United Kingdom; 2 University of Oxford, Department of Oncology, Oxford, United Kingdom; 3 Zuidwest Radiotherapeutisch Instituut , Physics, Vlissingen, The Netherlands; 4 Maastricht University Medical Centre, Department of Radiation Oncology (Maastro), Maastricht, The Netherlands; 5 Maastricht University Medical Centre, Department of Radiation Oncology (Maastro), Maastricht , The Netherlands; 6 Institute of Cancer Research, Division of Clinical Studies, Sutton, United Kingdom; 7 Institute of Cancer Research, Division of Radiotherapy and Imaging, Sutton, United Kingdom; 8 Maastricht University Medical Centre, Department of Radiation Oncology (Maastro) , Maastricht , The Netherlands Purpose or Objective A popular Normal tissue Complication (NTCP) model deployed to predict radiotherapy (RT) toxicity is the Lyman-Burman Kutcher (LKB) model of tissue complication. This model consists of three parameters, n, m, and D 50 , such that