Brain-phenotype models fail for individuals who defy sample stereotypes.

Individual differences in brain functional organization track a range of traits, symptoms and behaviours1-12. So far, work modelling linear brain-phenotype relationships has assumed that a single such relationship generalizes across all individuals, but models do not work equally well in all participants13,14. A better understanding of in whom models fail and why is crucial to revealing robust, useful and unbiased brain-phenotype relationships. To this end, here we related brain activity to phenotype using predictive models-trained and tested on independent data to ensure generalizability15-and examined model failure. We applied this data-driven approach to a range of neurocognitive measures in a new, clinically and demographically heterogeneous dataset, with the results replicated in two independent, publicly available datasets16,17. Across all three datasets, we find that models reflect not unitary cognitive constructs, but rather neurocognitive scores intertwined with sociodemographic and clinical covariates; that is, models reflect stereotypical profiles, and fail when applied to individuals who defy them. Model failure is reliable, phenotype specific and generalizable across datasets. Together, these results highlight the pitfalls of a one-size-fits-all modelling approach and the effect of biased phenotypic measures18-20 on the interpretation and utility of resulting brain-phenotype models. We present a framework to address these issues so that such models may reveal the neural circuits that underlie specific phenotypes and ultimately identify individualized neural targets for clinical intervention.

Predicting the future of neuroimaging predictive models in mental health.

Predictive modeling using neuroimaging data has the potential to improve our understanding of the neurobiology underlying psychiatric disorders and putatively information interventions. Accordingly, there is a plethora of literature reviewing published studies, the mathematics underlying machine learning, and the best practices for using these approaches. As our knowledge of mental health and machine learning continue to evolve, we instead aim to look forward and "predict" topics that we believe will be important in current and future studies. Some of the most discussed topics in machine learning, such as bias and fairness, the handling of dirty data, and interpretable models, may be less familiar to the broader community using neuroimaging-based predictive modeling in psychiatry. In a similar vein, transdiagnostic research and targeting brain-based features for psychiatric intervention are modern topics in psychiatry that predictive models are well-suited to tackle. In this work, we target an audience who is a researcher familiar with the fundamental procedures of machine learning and who wishes to increase their knowledge of ongoing topics in the field. We aim to accelerate the utility and applications of neuroimaging-based predictive models for psychiatric research by highlighting and considering these topics. Furthermore, though not a focus, these ideas generalize to neuroimaging-based predictive modeling in other clinical neurosciences and predictive modeling with different data types (e.g., digital health data).

(Un)common space in infant neuroimaging studies: A systematic review of infant templates.

In neuroimaging, spatial normalization is an important step that maps an individual's brain onto a template brain permitting downstream statistical analyses. Yet, in infant neuroimaging, there remain several technical challenges that have prevented the establishment of a standardized template for spatial normalization. Thus, many different approaches are used in the literature. To quantify the popularity and variability of these approaches in infant neuroimaging studies, we performed a systematic review of infant magnetic resonance imaging (MRI) studies from 2000 to 2020. Here, we present results from 834 studies meeting inclusion criteria. Studies were classified into (a) processing data in single subject space, (b) using an off the shelf, or "off the shelf," template, (c) creating a study specific template, or (d) using a hybrid of these methods. We found that across the studies in the systematic review, single subject space was the most used (no common space). This was the most used common space for diffusion-weighted imaging and structural MRI studies while functional MRI studies preferred off the shelf atlases. We found a pattern such that more recently published studies are more commonly using off the shelf atlases. When considering special populations, preterm studies most used single subject space while, when no special populations were being analyzed, an off the shelf template was most common. The most used off the shelf templates were the UNC Infant Atlases (24%). Using a systematic review of infant neuroimaging studies, we highlight a lack of an established "standard" template brain in these studies.