I'll try to describe the method as I understand it (this comment will be edited over time whenever I have a chance to read through the paper).
Calculate the smallest sphere that encompasses some number (k) of the nearest coordinates from different studies, for each coordinate. So for coordinate i, you would have a volume, dVi, describing that coordinate's sphere.
It seems like k (the number of independent studies to require for each coordinate) is a key parameter.
The minimum possible k value is 4. The number must be kept small because "the density estimate is only valid for small volumes to meet anatomical constraints."
Analysis of Brain Coordinates (ABC) is a new CBMA algorithm proposed in Tench et al. (2022).
Replication data are available at https://rdmc.nottingham.ac.uk/handle/internal/9171 and https://rdmc.nottingham.ac.uk/handle/internal/9121.
I'll try to describe the method as I understand it (this comment will be edited over time whenever I have a chance to read through the paper).