Presenter: Xinyuan Chen
BME 4th year Ph.D. student
PI Affiliation: Dr. Tianyu Zhao lab
Accuracy in proton stopping power ratio (SPR) measurement is critical to proton radiation therapy. Current standard practice to determine SPR involves calibrating SPR from X-ray computed tomography (xCT) Hounsfield Units (HUs) which tends to bring in extra sources of errors such as calibration error, xCT error, and patient positioning error. These errors lead to range uncertainty margins of ±3.5% and thereby compromise treatment effectiveness. Proton computed tomography (pCT) that utilizes proton directly for imaging can improve SPR accuracy by eliminating additional errors inherent in xCT. Previous study by our group has proposed a novel design of pCT system detected by multiple-layer ionization chambers, which has compact configuration for easy integration to proton treatment room. While employing a brass collimator during the image process confined the proton spot size to 1mm in diameter and thus improved SPR accuracy and spatial resolution, the interaction between brass collimator and proton source may induce additional neutron dose to the patient and degrade proton source efficiency. In this presentation, a model-based pCT reconstruction method for our pCT design that reconstructs SPR distribution from measurements acquired with uncollimated proton beam, as wide as 4mm in diameter, will be introduced. To achieve spatial resolution that is smaller than the proton spot size, the wide proton beam was divided into a set of narrow beamlets separated by 1mm. Preliminary reconstruction results from simulated measurements using TOPAS Monte Carlo simulation toolkit will be presented.