Fig 2 - available via license: Creative Commons Attribution-NonCommercial 4.0 International
Content may be subject to copyright.
Comparison of unmixing performance for SSASU, NLS, and PoissonNMF across ten test images taken from five samples. The relative reconstruction error (top) evaluates each method's ability to reconstruct the observed spectra image. The proportion indeterminacy (bottom) measures the non-orthogonality of the weight matrices and illustrates how well each method separates the fluorophore endmembers in the presence of autofluorescence.
Source publication
Spectral unmixing methods attempt to determine the concentrations of different fluorophores present at each pixel location in an image by analyzing a set of measured emission spectra. Unmixing algorithms have shown great promise for applications where samples contain many fluorescent labels; however, existing methods perform poorly when confronted...
Contexts in source publication
Context 1
... RRE and PI indicate the fit and quality of each unmixing solution. SSASU, NLS, and PoissonNMF were each able to ef- fectively reconstruct the test images with RREs below 0.14 (see top of Figure 2). Across all test images, NLS had the highest RRE. ...
Context 2
... having similar RRE across all test images, the qual- ity of the solutions varied dramatically between the different methods. SSASU outperformed both NLS and PoissonNMF in all but one test case at reducing proportion indeterminacy (see bottom of Figure 2). For eight of the ten test images, Poisson- NMF was least able to cleanly separate the fluorophores. ...
