Deep learning-based synthetic CT generation for paediatric brain MR-only photon and proton radiotherapy. (December 2020)

• Record Type: Journal Article
• Authors: Maspero, Matteo; Bentvelzen, Laura G.; Savenije, Mark H.F.; Guerreiro, Filipa; Seravalli, Enrica; Janssens, Geert O.; van den Berg, Cornelis A.T.; Philippens, Marielle E.P.
• Is Part Of: Radiotherapy and oncology. Volume 153(2020)
• Page Start: 197
• Page End: 204
• Publication Date: 2020-12
• Subjects: Synthetic CT -- Pediatric oncology -- Brain tumors -- Artificial intelligence -- Image-to-image translation -- Machine learning; Oncology -- Periodicals; Radiotherapy -- Periodicals; Tumors -- Periodicals; Medical Oncology -- Periodicals; Neoplasms -- radiotherapy -- Periodicals; Radiotherapy -- Periodicals; Radiothérapie -- Périodiques; Cancérologie -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9940642
• DOI: 10.1016/j.radonc.2020.09.029
• Languages: English
• ISSNs: 0167-8140

Abstract

Highlights: A deep learning network enabled photon and proton dose calculation in a pediatric brain population. The network generated accurate sCT even from a dataset heterogeneous in terms of demographic, anatomy, bone density and imaging protocols. A combination of networks trained on the three orthogonal planes outperforms networks trained on single plane. Abstract: Background and Purpose: To enable accurate magnetic resonance imaging (MRI)-based dose calculations, synthetic computed tomography (sCT) images need to be generated. We aim at assessing the feasibility of dose calculations from MRI acquired with a heterogeneous set of imaging protocol for paediatric patients affected by brain tumours. Materials and methods: Sixty paediatric patients undergoing brain radiotherapy were included. MR imaging protocols varied among patients, and data heterogeneity was maintained in train/validation/test sets. Three 2D conditional generative adversarial networks (cGANs) were trained to generate sCT from T1-weighted MRI, considering the three orthogonal planes and its combination (multi-plane sCT). For each patient, median and standard deviation ( σ ) of the three views were calculated, obtaining a combined sCT and a proxy for uncertainty map, respectively. The sCTs were evaluated against the planning CT in terms of image similarity and accuracy for photon and proton dose calculations. Results: A mean absolute error of 61 ± 14 HU (mean ± 1 σ ) was obtained in the intersection of the body contours between CT and sCT. The combined multi-plane sCTs performed better than sCTs from any single plane. Uncertainty maps highlighted that multi-plane sCTs differed at the body contours and air cavities. A dose difference of −0.1 ± 0.3% and 0.1 ± 0.4% was obtained on the D > 90% of the prescribed dose and mean γ 2 %, 2 mm pass-rate of 99.5 ± 0.8% and 99.2 ± 1.1% for photon and proton planning, respectively. Conclusion: Accurate MR-based dose calculation using a combination of three orthogonal planes for sCT generation is feasible for paediatric brain cancer patients, even when training on a heterogeneous dataset.