Trivial completion using the “other connectome”

In the case in which one of the two connectomes is missing (e.g., just SC available but not FC) one may think to use the available connectome (in this example, SC) as a “good guess” for the missing one (in this example, FC). We refer to this trivial procedure as a completion using the other connectome. If the match quality between surrogate connectomes obtained via more complex procedures and the target empirical connectome to reconstruct happened to be comparable with the one that one can get via the trivial completion, then it would not be worth using more sophisticated methods. We assessed then, for comparison with other strategies, the performance of such trivial completion approach on the SC_emp + FC_emp subset of the ADNI dataset and on the whole healthy aging dataset. In order for a completion approach to be considered viable, it is necessary that it outperforms significantly this trivial completion via the “other type” connectome, which can be quantified by a relative improvement coefficient: