Improvement of spatial resolution of PIXE-CT using ML-EM algorithm in TIARA

TIARAにおけるML-EMアルゴリズムを用いたPIXE-CTの空間分解能の向上

佐藤 隆博; 江夏 昌志; 加田 渉; 横山 彰人; 大久保 猛; 山崎 明義; 石井 保行; 神谷 富裕

Sato, Takahiro; Koka, Masashi; Kada, Wataru; Yokoyama, Akihito; Okubo, Takeru; Yamazaki, Akiyoshi; Ishii, Yasuyuki; Kamiya, Tomihiro

In order to get better spatial resolution of PIXE-CT, ML-EM (maximum likelihood expectation maximization) iterative algorithm) was used instead of FBP. The ML-EM algorithm had been developed for emission computed tomography such as SPECT (single photon emission computed tomography). It is relatively easy to apply the ML-EM to micro-PIXE because both SPECT and micro-PIXE originated from photon emission phenomena. Moreover, unlike FBP, the ML-EM algorithm has highly resistance against the errors of projection images and is suitable for rare event analyses such as micro-PIXE. As a result, the estimated FWHM is 4.4 0.6 m, which is the best performance of PIXE-CT in our study.