CORRIGENDUM Exact iterative reconstruction for the interior problem

The statement after equation (20) in our paper was not accurate. The statement was that h(z) defined by (20) is an analytic function. In fact, this statement is true only when the attenuation map μ(x, y) is smooth (i.e. differentiable) within the ROI, because the attenuation map is embedded in the integrand of equation (20). If the attenuator μ(x, y) has sharp edges within the ROI, the ROI reconstruction is not uniquely determined and may be corrupted by the attenuator edges that are within the ROI.