You have already solved the problem. You can get the histogram with any resolution and then sum values of the histogram to get total number of voxels in a certain intensity range.
Alternatively, you can find a complete implementation of this feature as a module with a nice GUI in LungCTAnalyzer extension: it computes volume of each region that is specified by voxel intensity range. It shows nice real-time preview, generates, PDF report, etc. To get started, you can use the module as is, just ignoring that takes “left lung” and “right lung” segments as inputs. You can put any other structure in those segments and you can specify any intensity range. If it is close enough to what you want to achieve, you can easily customize it to be more convenient and intuitive for your data sets - it is just a Python scripted module that you can modify with a text editor.