I have not experience in this particular application of MC simulations but the usual way to treat heterogeneities in MC simulations is by dividing the geometry into regions with similar  properties. The transport within each region is made  like in a homogeneous medium. In other words, you would have to do a segmentation of the tissues in question into regions with similar transport properties.

Thank You Dr Mario, I understood what you say but the idea is any tool or experiment can do to allow to measure this kind of heterogeneity inside tissues and there after can make simulations

In Monte Carlo you usually need to know the transport properties of the media in question prior to do any simulation. I do not now how to obtain these optical properties in tissues, maybe through optical tomography.

There are methods that can be used to measure tissue optical properties, e.g. inverse adding doubling with integration spheres, oblique angle diffuse reflectance measurement, fibre-based optical reflectance spectroscopy, differential pathlength spectroscopy etc