ESTRO 2023 - Abstract Book

S156

Saturday 13 May

ESTRO 2023

segmentation (using the autocontouring software Limbus AI) in the reconstructed vs the baseline HR images was assessed using mean distance-to-agreement (mDTA). As Limbus AI is not validated for paediatric data, we carefully inspected the segmentations. Three datasets were available: 1) 10 new ABCD images (dataset 1); 2) 18 images from the Children’s Brain Tumour Network (CBTN) study (acquired HR and simulated LR images, age 2–20years, dataset 2) and 3) 6 “real-world” follow-up images of a paediatric head and neck cancer patient (acquired HR and acquired LR, 14-19years, dataset 3).

Results The proposed CNN outperformed simple interpolation. PSNR for images CNN were on average(sd) 26.1(2.1) for dataset 1 and 24.4(2.6) for dataset 2, while for all images interp were 20.5(1.9) and 21.4(2.8), respectively.

Similarly, structure segmentation was more precise (closer to that of baseline images) in images CNN compared to images interp (Figures 2a and 2b).

Conclusion This work demonstrates that deep learning methods can successfully reconstruct 3D HR images from 2D LR ones, potentially advancing research for paediatric radiotherapy effects. Our model outperforms standard interpolation, both in perceptual