Fig. 2: Low consistency of subtyping solutions between and within data modalities.
From: Neuroimaging-based variability in subtyping biomarkers for psychiatric heterogeneity
A Subtype solutions across modalities. The Sankey diagram depicts different subtype assignments across clustering solutions for the three data modalities. B Between modalities consistency of the identified subtypes. The adjusted rand index (ARI) and adjusted mutual information (AMI) values were estimated between clustering solutions derived using different data modalities. C–E Consistency of the subtypes using two different feature types from the (C) structural, (D) task-based, or (E) resting-state data. For each data modality, the ARI and AMI values were estimated between clustering solutions derived using two different clustering features. Bounds of the box represent the 1st (25%) and 3rd (75%) quartiles, the central line represents the median, the whiskers represent the values within 1.5 times of the interquartile range, the flier points represent outliers falling beyond the whiskers.
