DEVELOPMENTS WITH MAXIMUM-LIKELIHOOD X-RAY COMPUTED-TOMOGRAPHY - INITIAL TESTING WITH REAL DATA

We investigate the potential and present limitations of a maximum-like lihood (ML) approach to x-my computed tomography that utilizes Poisson modeling and an iterative gradient-based algorithm. This model and al gorithm incorporate the finite width of the x-ray beam, and they were extended from an approach originally proposed by Lange et al. [IEEE Tr ans. Med. Imaging MI-6,106-114(1987)]. Low-count data, obtained from a n industrial computed-tomography scanner, are used to reconstruct an i mage of a concrete cube with metal reinforcing bars. We utilize both M L and filtered backprojection to reconstruct a cross section of the in ternal structure of the cube. In this initial evaluation with low-coun t data the image reconstructed by ML show several potential advantages over those reconstructed by filtered backprojection. The advantages s hown are the following: (1) there are significantly reduced noise and streak artifacts in the ML image; (2) some of the known structural det ail is more apparent in the ML image; (3) there is a closer quantitati ve fit, based on log-likelihood and residual calculations, between the ML image and the observed data; (4) the ML approach shows the potenti al to achieve finer spatial resolution than filtered backprojection. W e observe two present, yet addressable, limitations of the ML approach . First, the ML image currently has a peripheral smoothing artifact th at seems to disappear gradually with increasing iteration numbers. Thi s smoothing is possibly caused by the slow rate of convergence of the algorithm and may be addressed by future acceleration strategies. Seco nd, the finer spatial resolution achieved with the ML approach current ly occurs at the expense of noise and edge artifacts. This limitation may be addressed by a number of extended ML and maximum a posteriori a pproaches that are currently under investigation in other modalities o f imaging to address similar noise and edge artifacts.