Stability of MRI Radiomics Features of Hippocampus: An Integrated Analysis of Test-Retest and Inter-Observer Variability

Abstract

Radiomics is a quantitative method to extract medical features that can extend to details that are invisible to human eyes. Studies have suggested that radiomics of the hippocampus is a potential imaging biomarker for several diseases, including Alzheimer's disease, mild cognitive impairment, and multiple sclerosis. This study investigates the test-retest reproducibility of radiomics features of the hippocampus. For this purpose, the intra-class correlation coefficient (ICC) and overall concordance correlation coefficient (OCCC) of the radiomics features of the hippocampus were well studied in four datasets consisting of 15, 19, 23, and 3 subjects who were scanned three, two, four, and forty times, respectively. The results highlight that the hippocampal texture features are the most repeatable, and that texture features are a possible neuroimaging biomarker for clinical studies.