Fluorescence molecular tomography (FMT) is a promising research tool that enables accurate localization and quantification of fluorophore distribution noninvasively in intact small animal model systems. Typically, FMT data is collected using a number of optical sensors combined with sources, placed on the periphery of the imaged domain where near infrared (NIR) light is projected into the domain using a single source, while light is collected at all the other sensors simultaneously. The source of the light is cycled through the encircled optical sensors around the periphery of the domain facilitating collection of “boundary data,” which is used for image reconstruction. By definition, image reconstruction from this boundary data is an ill posed inverse mathematical problem requiring model-based solutions. Many reconstruction algorithms have been developed to facilitate improved image reconstruction from collected boundary data; however, improved image reconstruction is still possible. In the current study by Yi et al., a reconstruction frame based on three-way decisions (TWD) to enable image reconstruction for FMT is introduced. TWD was first proposed in 2009 and has been used for other applications, but its first use for FMT is presented herein. Model and small animal imaging studies demonstrate that TWD-based reconstruction improves overall image reconstruction and is a promising framework for FMT image reconstruction.
You must log in
to add comments.