(Sunday, 7/12/2020) [Eastern Time (GMT-4)]
Room: AAPM ePoster Library
Purpose:
This study is aimed to extract atomic numbers ( ^Z and ~Z) of tissues with triple energy CT (TECT) and to evaluate the corresponding improvement on stopping power ratio (SPR) calculation for proton therapy.
Methods:
Dual-energy CT (DECT) degenerates the subtle tissue differences while assuming ^Z= ~Z=Zeff. By adding a third energy, ^Z and ~Z, presenting the atomic number in photoelectric and coherent interaction, can be differentially derived to potentially improve the CT based tissue SPR derivation. To verify this method, a virtual CT scanner was simulated in Geant4 and 3 CT image sets using 90kVp, 120kVp and 140kVp of a water phantom including 4 standard human tissues from ICRU46 were obtained. CT images were reconstructed using ifanbeam function in MATLAB. A 3D-monotonous conversion between ^Z, ~Z and mean excitation energy (Im) was generated. Another CIRS phantom with 9 inserts were simulated to obtain K series that represent the feature of CT spectra. The accuracy of SPR calculated by TECT was compared to DECT approach (90/140kVp) and single energy CT (SECT) approach (120kVp).
Results:
The comparison of SPRs showed that TECT and DECT had better performances in SPR calculation than SECT. The SPR of lung, conventionally having a larger uncertainty, showed closer values to the theoretical SPR, with less than 1% for TECT approach, comparing to 3.26% for DECT and 4.04% for SECT approaches. The accuracy of SPRs for liver and muscle from TECT was higher than those of CTs as well, with less than 2% difference from the theoretical values. However, the SPR of breast had a different trend with the higher accuracy from SECT than DECT or TECT.
Conclusion:
The comparisons of SPRs show TECT and DECT have better performances in SPR calculation than SECT for most of tissues. However, for breast tissue, SECT performs better.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
