Purpose: To assess relative stopping power (RSP) accuracy achieved with the proton CT (pCT) phase II prototype scanner and compare it to the one achievable with a state-of-the-art dual-energy X-ray CT (DECT) scanner. Furthermore, to investigate limitations of the pCT prototype scanner which affect the RSP accuracy.
Methods: Two cylindrical phantoms, containing in total 13 cylindrical inserts, of different tissue equivalent plastic materials were scanned with the pCT prototype scanner at the Northwestern Medicine Chicago Proton Center and with the SOMATOM Definition FORCE scanner (Siemens Healthineers, Forchheim, Germany) at the University Hospital of LMU Munich. The reference RSP of the inserts was provided by variable water column measurements performed at the Heidelberg Ion Therapy center. RSP accuracy was defined as the percentage error between the reconstructed mean RSP and the reference RSP. An overall RSP accuracy for each modality was calculated as the mean absolute percentage error (MAPE) for all inserts. To further investigate potential limitations of the pCT scanner performance affecting the RSP accuracy, a realistic Geant4 based simulation platform was used.
Results: The obtained RSP MAPE was 0.87% both for pCT and DECT. RSP errors of more than 1% were observed for both imaging modalities. High RSP errors in pCT were related to image ring artifacts, caused by protons stopping near the interface between two adjacent energy detector segments.
Conclusion: We report on the results of the first experimental comparison between pCT and DECT for RSP accuracy, based on the same phantoms. The two modalities reached at this stage equivalent performance in terms of RSP determination. Room for significant improvement of energy-detector-related limitations has been identified for the pCT prototype scanner.
Funding Support, Disclosures, and Conflict of Interest: Research supported by the Bavaria California Technology Center (BaCaTeC)
Not Applicable / None Entered.