Feret diameter (or similar) representation



Hi all: does anyone know whether there is a way to represent an object (in our case, a 3D confocal image of a cell) using a line that defines the feret diameter? In other words, if one wanted to illustrate how cells were aligned (or not) in a tissue, a single line representing each feret distance (after segmentation) would essentially be a “stick figure” that would show the relative alignments. Or perhaps another way to accomplish this same goal, illustrating alignment of the longest aspect of cells in a tissue? Thank you very much, Howard


OK, here’s my crazy idea. What about summing the vectors of each slice of your object in the stack, and then sum that result with a perpendicular vector through the stack. I think it might work, but there are plenty of better brains here.


Thank you for your response, but I am not exactly sure that I understand what you mean by “summing the vectors of each slice of your object”? Thank you, Howard


Thomas Boudier’s 3D ROI Manager returns a 3D Feret:
but I do not know if that is only its length, coordinates or the angle.

You could get the moments of inertia with BoneJ http://bonej.org/moments (not the Feret but the axis, which might be also useful to characterise directionality).


Thanks Gabriel: I’ll report back after we’ve had a chance to give those a try.