Selective heating in industrial microwave drying could be more efficiently addressed by intelligent control of distributed microwave sources. As a result, increasing system efficiency and reducing thermal runaway while processing low loss dielectric samples. However, applying such a precise microwave control requires non-invasive in-situ measurement of the unknown distribution of moisture inside the material. In this work, the feasibility of integrating a microwave tomography (MWT) with the drying system is demonstrated. The studied imaging modality is applied to estimate the moisture content distribution in a polymer foam. To solve the estimation problem in a fast way, a neural network based approach is proposed in this work. Promising estimation results are shown using synthetic measurement data.