Room: Exhibit Hall | Forum 2
Purpose: To quantify the accuracy of stopping power ratio (SPR) calculation with a dual-layer CT system and compare range and dosimetric differences with conventional single-energy CT. We aim to reduce the proton range uncertainty with the new CT technology and provide a site-specific estimate of uncertainty for treatment planning.
Methods: A Gammex tissue-equivalent phantom was scanned with dual-layer CT (Philips IQon Spectral CT) to obtain electron density (Ï?â‚‘) and effective atomic number (Zâ‚‘(ff)) images directly. The latter were converted to mean ionization potential (I(m)) via a Zâ‚‘(ff) â€“LogI(m) relationship created using ICRU44 tissues. The SPR images were calculated with the Beth equation and imported into a treatment planning system. Comparisons were made to expected SPRs calculated based on chemical compositions and those mapped from Hounsfield units using the stoichiometric calibration curve with conventional CT. Singe-field-uniform-dose plans were designed on patient images of various anatomic sites to examine the differences in proton ranges and dose distributions.
Results: For dual-layer CT, percentage deviations of SPR from expected values were within Â±1% for all tissue-equivalent materials, except for liver (1.4%) and brain (1.5%) surrogates. The overall root-mean-square error was 0.7%. In contrast, the deviations were -3% to 9% for the conventional approach. The dual-layer CT produced consistently lower deviations across all tissue surrogates, particularly significant for lungs, brain, liver, and inner bone. For simulated treatment plans, the maximum dose difference of two approaches could reach 10 Gy(RBE) from a 2-mm range pull back in a thorax case prescribed to a target dose of 50 Gy(RBE).
Conclusion: Dual-layer CT provides more accurate SPR estimates than the stoichiometric approach with conventional CT on phantom study. Patient simulations demonstrated differences in proton ranges and calculated doses at the distal end of the targets, supporting further investigations on clinical scenarios.
Funding Support, Disclosures, and Conflict of Interest: This research was carried out under a research agreement with Philips Healthcare.