Experimental validation of two dual-energy CT methods for proton therapy using heterogeneous tissue samples

Esther Bär,Arthur Lalonde,Rongxiao Zhang,Kyung-Wook Jee,Kai Yang,Gregory Sharp,Bob Liu,Gary Royle,Hugo Bouchard,Hsiao-Ming Lu

DOI: https://doi.org/10.1002/mp.12666

Abstract:Purpose: The purpose of this work is to evaluate the performance of dual-energy CT (DECT) for determining proton stopping power ratios (SPRs) in an experimental environment and to demonstrate its potential advantages over conventional single-energy CT (SECT) in clinical conditions. Methods: Water equivalent range (WER) measurements of 12 tissue-equivalent plastic materials and 12 fresh animal tissue samples are performed in a 195 MeV broad proton beam using the dose extinction method. SECT and DECT scans of the samples are performed with a dual-source CT scanner (Siemens SOMATOM Definition Flash). The methods of Schneider et al. (1996), Bourque et al. (2014), and Lalonde et al. (2017) are used to predict proton SPR on SECT and DECT images. From predicted SPR values, the WER of the proton beam through the sample is predicted for SECT and DECT using Monte Carlo simulations and compared to the measured WER. Results: For homogeneous tissue-equivalent plastic materials, results with DECT are consistent with experimental measurements and show a systematic reduction of SPR uncertainty compared to SECT, with root-mean-square errors of 1.59% versus 0.61% for SECT and DECT, respectively. Measurements with heterogeneous animal samples show a clear reduction of the bias on range predictions in the presence of bones, with -0.88% for SECT versus -0.58% and -0.14% for both DECT methods. An uncertainty budget allows isolating the effect of CT number conversion to SPR and predicts improvements by DECT over SECT consistently with theoretical predictions, with 0.34% and 0.31% for soft tissues and bones in the experimental setup compared to 0.34% and 1.14% with the theoretical method. Conclusions: The present work uses experimental measurements in a realistic clinical environment to show potential benefits of DECT for proton therapy treatment planning. Our results show clear improvements over SECT in tissue-equivalent plastic materials and animal tissues. Further work towards using Monte Carlo simulations for treatment planning with DECT data and a more detailed investigation of the uncertainties on I-value and limitations on the Bragg additivity rule could potentially further enhance the benefits of this imaging technology for proton therapy.