Preview

PDF, English Download (17MB) | Terms of use

Abstract

Radiotherapy with protons or light ions is a highly precise form of cancer treatment. In treatment planning for particle therapy, ion stopping power ratio (SPR) maps of patient tissues are used to predict particle ranges and calculate dose distributions. To more accurately calculate dose distributions and minimize irradiating healthy tissue, it is crucial to improve SPR prediction. To this end, this thesis investigated dual-layer spectral computed tomography, a dual-energy CT (DECT) technique, as an alternative to conventional single-energy CT (SECT). The SECT-based method relies on converting CT numbers to SPR, yet CT numbers acquired from photon attenuation cannot be used to accurately predict energy loss by ions, which makes the approach indirect and heuristic. The DECT-based method, however, uses measurements of relative electron density and effective atomic number to directly and patient-specifically predict SPR. SPR prediction using DECT was evaluated in tissue-equivalent materials, anthropomorphic phantoms, and non-tissue materials; clinically analyzed in a retrospective patient study; and experimentally investigated for patients with dental materials. DECT-based SPR prediction improved dose calculation accuracy in particle therapy compared to SECT with a remaining range uncertainty of about 1% in controlled experimental scenarios. DECT may thus substantially improve range prediction for highly accurate particle therapy.