Outlier detection in multimodal MRI identifies rare individual phenotypes among 20,000 brains

Outliers in neuroimaging represent spurious data or the data of unusual phenotypes that deserve special attention such as clinical follow-up. Outliers have usually been detected in a supervised or semi-supervised manner for labeled neuroimaging cohorts. There has been much less work using unsupervised outlier detection on large unlabeled cohorts like the UK Biobank brain imaging dataset. Given its large sample size, rare imaging phenotypes within this unique cohort are of interest, as they are often clinically relevant and could be informative for discovering new processes. Here we developed a two-level outlier detection and screening methodology to characterize individual outliers of multiple different brain imaging phenotypes from 20,000 UK Biobank subjects. In primary screening, every subject was parameterized with an outlier score per imaging phenotype to quantitate the degree of outlierness. This approach enabled the assessment of test-retest reliability via outlier scores, which ranged from excellent reliability for ventricular volume, white matter lesion volume, and fractional anisotropy, to good reliability for mean diffusivity and cortical thickness. Resting-state functional connectivity was eliminated for individual-level outlier screening due to its low test-retest reliability of outlier scores. In secondary screening, the extreme outliers (1110 subjects) were examined individually, and those arising from data collection/processing errors were eliminated. A subgroup (108 subjects) of the remaining non-artifactual outliers were radiologically reviewed, and radiological findings were identified in 98%. This study establishes an unsupervised framework for investigating rare individual imaging phenotypes within a large neuroimaging cohort.