Unraveling reproducible dynamic states of individual brain functional parcellation

Abstract

Data-driven parcellations are widely used for exploring the functional organization of the brain, and also for reducing the high dimensionality of fMRI data. Despite the flurry of methods proposed in the literature, functional brain parcellations are not highly reproducible at the level of individual subjects, even with very long acquisitions. Some brain areas are also more difficult to parcellate than others, with association heteromodal cortices being the most challenging. An important limitation of classical parcellations is that they are static, that is, they neglect dynamic reconfigurations of brain networks. In this paper, we proposed a new method to identify dynamic states of parcellations, which we hypothesized would improve reproducibility over static parcellation approaches. For a series of seed voxels in the brain, we applied a cluster analysis to regroup short (3 min) time windows into “states” with highly similar seed parcels. We split individual time series of the Midnight scan club sample into two independent sets of 2.5 hr (test and retest). We found that average within-state parcellations, called stability maps, were highly reproducible (over 0.9 test-retest spatial correlation in many instances) and subject specific (fingerprinting accuracy over 70% on average) between test and retest. Consistent with our hypothesis, seeds in heteromodal cortices (posterior and anterior cingulate) showed a richer repertoire of states than unimodal (visual) cortex. Taken together, our results indicate that static functional parcellations are incorrectly averaging well-defined and distinct dynamic states of brain parcellations. This work calls to revisit previous methods based on static parcellations, which includes the majority of published network analyses of fMRI data. Our method may, thus, impact how researchers model the rich interactions between brain networks in health and disease.

Author Summary: Functional brain parcellation has been a very active topic of investigation for the past two decades, yet there is no evidence to date of reproducible results at the individual level—see for example Figure 2 in Gordon, Laumann, Gilmore, et al. (2017), with a Dice coefficient plateauing around 0.7 using 40 min or more of data. In this paper, we show that highly reproducible brain parcels can be observed using short (3 min) time windows. Different modes—or states—of reproducible parcellations can be observed in a single brain region, and these modes have only little overlap with each other. We carefully quantified these individual dynamic states of parcellation using the Midnight Brain Scan dataset, featuring 5 hr of functional MRI per subject. Our results indicate that static functional parcellation are incorrectly averaging well-defined and distinct dynamic states. This brings important caution for any work based on static atlases, which is the dominant approach currently in so-called network analysis of fMRI data.