Iteration algorithm to correct absorption in PET

One of the difficulties arising in PET is attenuation of emission projection data. The most difficult situation arises in the cases in which no information as to the distribution of the absorption coefficient is available. It is this case that is considered in the present paper. A new approach to the problem is proposed. Its main idea consists in that the measurement data, as soon as they all belong to one object, are bound to be self-consistent. The whole set of projection data is considered as an invariant function of the operator PR/sup -1,/ where R/sup -1/ is an algorithm solving the emission tomography problem and P is the projecting operator. An iteration algorithm that permits reconstruction of emission projection data is developed. Numerical simulations are carried out.