Mean EFC grade agreement stays higher than raw agreement for reference change ranges less than 0.0057072359681184 with constant MR-ART v1.0.2 grades and motion values.

In the Movement-Related ARTefacts (MR-ART) v1.0.2 dataset, we calculated a sufficient range of changes to participants’ standard entropy focus criterion (EFC) references for mean expert-grade agreement to remain higher than raw agreement, while grades and motion EFC values stay fixed. One reference change applies to the same participant in both head motion conditions, which have equal weight in the mean. Across the full sample of 148 participants, the sufficient condition is a change range below 0.0057072359681184. The corresponding limits are 0.0057388291952627 for 140 complete triplets and 0.006443288169841 for 119 participants with a standard scan graded 1. The limits are strict: equality is not included. The method gives a lower bound and can be conservative because condition-specific worst-case changes can conflict. The result applies to the recorded values and does not establish measurement accuracy, uncertainty for new participants, or actual standard-scan repeatability. MR-ART has one standard scan per participant, so repeat changes could not be measured. Artificial change tests are constructed checks, not repeat-scan observations.

MR-ART · MRI · entropy focus criterion · image quality · grade agreement · reference change · sensitivity bound · repeatability