Generating Synthetic Computed Tomography for Radiotherapy: SynthRAD2023 Challenge Report

Radiation therapy requires accurate anatomical imaging for precise dose planning. This work reports the SynthRAD2023 challenge, a multi-center benchmark for synthetic CT (sCT) generation from MRI and CBCT. Using ground-truth CT data from 1080 patients, the challenge evaluated MRI-to-CT and CBCT-to-CT synthesis methods. Results indicate strong structural similarity and dose accuracy metrics among top submissions and emphasize the need for dose-centric evaluation in clinical applicability assessments of sCT generation techniques. :contentReference[oaicite:0]{index=0}

Medical Image Analysis