Shiro Ikeda, Hirokazu Odaka, Makoto Uemura, Tadayuki Takahashi, Shin Watanabe, Shin'ichiro Takeda
Abstract
We study the image reconstruction problem of a Compton camera which consists of semiconductor detectors. The image reconstruction is formulated as a statistical estimation problem. We employ a bin-mode estimation (BME) and extend an existing framework to a Compton camera with multiple scatterers and absorbers. Two estimation algorithms are proposed: an accelerated EM algorithm for the maximum likelihood estimation (MLE) and a modified EM algorithm for the maximum a posteriori (MAP) estimation. Numerical simulations demonstrate the potential of the proposed methods.