I found that the size of features can be obtained by means of the contrast and dissimilarity metric on the Gray Level Co-occurrence Matrix (GLCM), see for instance the review (Hallur & Gavade, 2025).
Such plot should be shown to the user when they need to setup the subvolume size: click a button to show the graph, explain that they should choose a value approx twice of when they think the curve saturates.
I found that the size of features can be obtained by means of the contrast and dissimilarity metric on the Gray Level Co-occurrence Matrix (GLCM), see for instance the review (Hallur & Gavade, 2025).
Such plot should be shown to the user when they need to setup the subvolume size: click a button to show the graph, explain that they should choose a value approx twice of when they think the curve saturates.