ホットスポットファントムを用いたOSEM再構成画像の基礎的検討 : iteration回数およびsubset数の評価

書誌事項

タイトル別名
  • Basic Evaluation of OSEM Algorithm by Assessing Iteration Times and Number of Subsets in a Hot Spot Phantom Study

この論文をさがす

抄録

To investigate optimal iteration times and the number of subsets on OSEM(ordered subsets expectation maximization), we estimated the values of the target-to-background ratio(T/B), signal-to-noise ratio(S/N), and NMSE(normalized mean square error). We acquired SPECT data with four scan durations, 5, 10, 15, and 20 min, using a three-headed SPECT scanner(GCA-9300 A/UI, Toshiba Co., Ltd.), and reconstructed the OSEM algorithm varying iteration and subset numbers. We determined the optimal values of T/B, S/N, and NMSE for OSEM, and compared these by filtered back projection(FBP). We observed the same convergence propotity between OSEM and FBP. We evaluated this by comparing it with the FBP method to obtain the most appropriate parameter from 36 kinds of SPECT images. In the spherical hot spot phantom study, the values of 2 and 15 were optimum for iteration times and number of subsets, respectively.

収録刊行物

被引用文献 (9)*注記

もっと見る

参考文献 (14)*注記

もっと見る

詳細情報 詳細情報について

問題の指摘

ページトップへ