How to understand/convert the output of ksdensity and mvksdensity

Hi all,

I am trying to use Matlab's kernel smoothed density estimate functions like mvksdensity to create a smoothed 'histogram' of some data.

I really like the output and it works well for my data, the problem is I'm not sure how to handle the output for further analyses.

How can I convert the output of mvksdensity from probability into units of 'stuff'? i.e. if I compute the pdf of a series of 3D points, how can I get a map that represents the estimated count of points in every voxel, rather than the probability?

Naively I am tempted to divide the pdf by its sum then multiply by the number of 3D points, but this seems overly simplistic.

In another thread, someone mentioned multiplying by the size of the interval used first: https://uk.mathworks.com/matlabcentral/answers/157422-the-magnitude-of-y-axis-in-a-ksdensity-plot

But I'm not sure how this would apply to multivariate data from mvksdensity...

Thanks for any help,

Rod.