Segmentation of nuclei (Wählby algorithm)

The algorithm identifies nuclei in images of cells stained with a DNA-specific dye (e.g. DAPI).

[nuclearMask,nuclearMembranes]=wahlbynucleus(nuclearImage,gaussSD,hForMaxima,hForMinima,minSizeOfNucleus,matlabOrDipimage,weakBorderThr1,weakBorderThr2,isInteractive)
OR
wahlbynucleus(nuclearImage,gaussSD,hForMaxima,hForMinima,minSizeOfNucleus) with no output parameters.
nuclearImage - a Matlab matrix or a DipImage to be segmented
gaussSD - SD of the Gaussian filter
hForMaxima - height of the extended h-maxima transform
hForMinima - height of the extended h-minima transform
minSizeOfNucleus - size of the nucleus. After extended h-minima transform regional minima smaller than this will be removed.
matlabOrDipimage - 1: Matlab version of seeded-watershed, 2: DipImage-version of seeded=watershed
weakBorderThr1 - borders with strengths smaller than this will be removed after the 1st watershed transform
weakBorderThr2 - borders with strengths smaller than this will be removed after the 2nd watershed transform (on the distance transformed image)
isInteractive - if 1, borders can be manually removed at the end
Parameters gaussSD, hForMaxima, hForMinima and minSizeOfNucleus can be
optimized by calling the function with 5 input arguments (this is much
faster than the full segmentation).
Reasonably nice segmentation of the wahlby_testimage.tif can be achieved
with the following parameters: [nmask,nmembrane]=wahlbynucleus(a,1,5,3,100,1,25,30,1);
The program implements the algorithm described in the following paper:
Combining intensity, edge and shape information for 2D and 3D segmentation of cell nuclei in tissue sections.
Wählby C, Sintorn IM, Erlandsson F, Borgefors G, Bengtsson E.
J Microsc. 215: 67-76 (2004).
https://www.ncbi.nlm.nih.gov/pubmed/15230877
http://dx.doi.org/10.1111/j.0022-2720.2004.01338.x