Have you ever considered implementing a collapsed cone algorithm?
In ray station, the singular decompostion method is only used for optimization. However, cc algorithm is used for calculation. Have you tried to implement cc algorithm?
We have considered it, but nobody here has the time to do it. We have a prototype for sub-sampling algorithm, but it is not very performant.
Could you share some inspiration or source code about cc algorithm? I'll invite some physicists here to have a try. It should be fun.
I don't really have something useful for this, but here's two thoughts.
- The first useful thing one could implement is explicit TERMA calculation, since it is required and may also be useful for other purposes
- It does probably not make sense to start without having good polyenergetic point-spread dose kernels to build the collapsed cones
I can generate the MC-based polyenergetic point-spread dose kernels (or polyenergetic energy deposition kernels) for this.