Fluorescence molecular tomography (FMT) is an optical technique that allows fluorescent markers to be imaged in thick biological tissues. A FMT setup, which consists of a near infrared light source and a detector collecting fluorescence light, is very simple. However, the inverse problem that consists in reconstructing the 3D marker distribution from the fluorescence signal is severely ill-posed and highly challenging.

After a brief introduction to FMT regularized reconstructions, the talk will focus on recent developments dedicated to fast FMT. In particular, two dimension reduction approaches, namely structured light illumination and data compression, will be discussed. First, it will be shown that the use of well-chosen illumination patterns can reduce both acquisition and reconstruction times. Second, it will be shown that standard image compression techniques can easily be incorporated into the inversion framework, which significantly reduces the reconstruction cost. Reconstructions obtained from both simulated and experimental data will be presented.