Anatomical regularization methods for optical tomography

The choice of an appropriate regularization method is of crucial importance in optical tomography, since this image problem is highly ill-posed. In this work we approach the regularization problem within an iterative image iterative reconstruction scheme. The image reconstruction is considered as a minimization of an appropriately defined objective function. The objective function is composed of a least-square-error term, which compares predicted and actual detector readings, and additional regularization terms. Choosing regularization functions that include some basic anatomical information can lead to significant improvement in the image reconstruction results