Noise filtration in fluorescence-enhanced optical tomography: breast phantom studies

Abstract

Noisy measurements may impact the quality of image reconstructions in fluorescence-enhanced optical tomography studies. In this contribution, different noise pre-filtration techniques were developed using fluorescence measurements of homogeneous breast phantoms. The noise pre-filtration techniques were determined based on correlation studies among various measurement parameters. We demonstrated that noise pre-filtration techniques can improve the target localization in image reconstructions when the target is deeply located (>2 cm) in imperfect uptake case (i.e. target: background fluorescence absoroption contrast ratio of 100 : 1). Moreover, filtration of measurement noise using different criteria can possibly differentiate the target from artifacts, potentially minimizing the false-positive cases in a clinical environment.