ESTRO 2024 - Abstract Book

S3616

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2024

Figure-2 – Left: values of ipsilateral lung PC1 for each Consortium within the internal cross-validation cohort for Consortium A and the external cross-validation data set for Consortium B. ‘‘Inside” were values less or equal to the 90th percentile of the corresponding training set for all patients. Right: CONSORTIUM A models' prediction performance on CONSORTIUM B external cross-validation data test on the ipsilateral lung predicted DVHs: optimal prediction, suboptimal prediction, improved prediction and failed prediction.

Conclusion:

In this study comparable ipsilateral lung transferability (poor transferability only for institute 6 for both Consortia and institute 10 for Consortium B only) and good dose/statistical prediction was demonstrated, supporting the possibility of using RWB-TF models for the prediction of RWB-TF outside the national consortium where were generated. As the final next step, assessment of heart, contralateral breast and lung dose/statistical prediction will complete this first national inter-consortium evaluation of KB DVH prediction evaluation.

Keywords: KB model, international transferability, breast RT

References:

[1] Tudda A et al., Knowledge-based multi-institution plan prediction of whole breast irradiation with tangential fields. Radiother Oncol. 2022 Oct;175:10-16. doi: 10.1016/j.radonc.2022.07.012. Epub 2022 Jul 19. PMID: 35868603.

The current study was performed within the MIKAPOCo project, supported by AIRC Grant IG23150, 248/2021