Power Analysis for High Dimensional Neuroimaging Studies

Abstract

This article considers the problem of sample size determination for power analysis of neuroimaging studies involving multiple comparisons. High dimensional neuroimaging data arise naturally in the context of developing biomarkers for patients with neurological disorders. The standard methodology for power analysis cannot be used or extended in such cases because multiple comparisons involving high dimensional data are more complex and require more parametric information, rather than simply providing the effect size that often plays the vital role in a traditional power analysis problem. This article introduces concepts such as varying effect size and discrimination of null and non-null distributions, while exploring the complexity of relevant data. It systematically develops a sample size determination approach that has a simple interpretation and satisfies some desirable intuitive properties (eg, larger effect size requires a smaller sample, larger sample provides better power). Results are illustrated with a real data set. [ Psychiatr Ann . 2023;53(3):137–144.]