| Abstract |
Fluorescence Molecular Tomography (FMT) is a well established method of acquiring three dimensional fluorescence images. Based on the principles of the diffuse optical tomography (DOT) it extracts tomographic images from non contact measurements, when investigated sample carries at least one fluorescence target. In the case of two or more fluorophores the spectrum that will be recorded by the detection channel is a linear combination of the individual components.
In this study we present the different unmixing strategies that can be followed in order to separate the fluorescence signal of two overlapping fluorophores. Many of the most useful proteins such as GFP or DsRed have strongly overlapping emission spectra and hence it is not easy to separate them only with the use of filters. However, with the use of tissue-like phantoms with overlapping spectra, we would try to separate them in two different ways and two different modes. Generally, the unmixing processing is a linear algorithm that determine the contribution of each one of the fluorophore that exist simultaneously in the under examination sample, to the total signal. Therefore, in this way we can isolate the signal that we want to examine from the other signals that exist in the same detection area.
In order to do the unmixing process we have used two different dyes CFSE and Atto590, since we already know that their spectral is overlapping. We excited them at two wavelengths (514.5nm and 488nm), and continuously, using a spectrograph and the FMT setup we tried to separate the mixed signal that we recorded. The first way was by applying the unmixing algorithm in the already reconstruction data and the other way was by applying firstly the unmixing algorithm and them reconstructed the raw data. In both cases we used two different kinds of algorithms referring to the spectral strengths that are used. In this way we found that the unmixing in the reconstructed data is the most accurately way of unmixing.
|
Πολυφασματικός διαχωρισμός τρισδιάστατων εικόνων φθορισμού
