Conditional Diffusion Models for CT Image Synthesis from CBCT: A Systematic Review

Simple Summary Cone Beam Computed Tomography (CBCT) is widely used in radiotherapy because it is fast and relatively low dose, but its image quality is lower than that of conventional CT. Conditional diffusion models have recently been proposed to convert CBCT into synthetic CT with improved anatomical detail and more reliable Hounsfield Units. This systematic review summarizes current CBCT-to-CT synthesis conditional diffusion approaches, their reported quantitative performance, and their possible clinical relevance. The evidence is promising but still limited and heterogeneous, highlighting the need for standardized evaluation, stronger dose-related validation, and larger multi-institutional studies before broad clinical adoption.