Comparison of deep learning with three other methods to generate pseudo-CT for MRI-only radiotherapy

A. Largent (1, 2) , A. Barateau (1, 2) , J. Nunes (1, 2) , C. Lafond (1, 2) , P. B. Greer (3, 4) , J. A. Dowling (5) , Hervé Saint-Jalmes (1, 2) , Oscar Acosta (1) , R. de Crevoisier (1, 2)

1 LTSI - Laboratoire Traitement du Signal et de l'Image
2 CRLCC - CRLCC Eugène Marquis
3 Calvary Mater Newcastle Hospital
4 UoN - University of Newcastle [Australia]
5 CSIRO - Commonwealth Scientific and Industrial Research Organisation [Canberra]

Deep learning methods (DLM) have recently been developed to generate pseudo-CT (pCT) from MRI for radiotherapy dose calculation. The main advantage of these methods is the speed of pCT generation. The objective of this study was to compare a DLM to a patchbased method (PBM), an atlas-based method (ABM) and a bulk density method (BDM) for prostate MRI-only radiotherapy.

Domains: Life Sciences [q-bio]; Bioengineering; Imaging; Life Sciences [q-bio]; Bioengineering; Nuclear medicine

A. Largent, A. Barateau, J. Nunes, C. Lafond, P. B. Greer, et al.. Comparison of deep learning with three other methods to generate pseudo-CT for MRI-only radiotherapy. Radiotherapy & Oncology, 2019, 133 (Suppl 1), pp.S555-S556. ⟨10.1016/S0167-8140(19)31427-6⟩. ⟨hal-02177165⟩