The Human Connectome Project of adolescent anxiety and depression dataset

Abstract

This article describes primary data and resources available from the Boston Adolescent Neuroimaging of Depression and Anxiety (BANDA) study, a novel arm of the Human Connectome Project (HCP). Data were collected from 215 adolescents (14-17 years old), 152 of whom had current diagnoses of anxiety and/or depressive disorders at study intake. Data include cross-sectional structural (T1- and T2-weighted), functional (resting state and three tasks), and diffusion-weighted magnetic resonance images. Both unprocessed and HCP minimally-preprocessed imaging data are available within the data release packages. Adolescent and parent clinical interview data, as well as cognitive and neuropsychological data are also included within these packages. Release packages additionally provide data collected from self-report measures assessing key features of adolescent psychopathology, including: anxious and depressive symptom dimensions, behavioral inhibition/activation, exposure to stressful life events, and risk behaviors. Finally, the release packages include 6- and 12-month longitudinal data acquired from clinical measures. Data are publicly accessible through the National Institute of Mental Health Data Archive (ID: #2505).

Fig. 1

fMRI task examples. (a) Incentive Processing Task (IPT), collection alias tfMRI_GAMBLING. (b) Emotion Processing Task (EPT), collection alias tfMRI_FACEMATCHING. (c) Emotion Interference Task (EIT), collection alias tfMRI_CONFLICT. Figure images were modified from prior reports^,,.

Fig. 2

Conceptual overview and major steps of HCP’s minimal preprocessing workflows. Figure adapted from the HCP Young Adult article on minimal preprocessing workflows.

Fig. 3

Select data types, collections, and record structures. (a) Non-MRI (top) and MRI (bottom) data types. (b) MRI unprocessed superordinate directories (i.e., imagingcollection01), scan types, number of runs/directories, and some subdirectory information. (c) BANDA non-imaging data collections and associated NDA structure names. Note: K-SADS summary diagnoses provided are based upon adolescent report.

Fig. 4

imagingcollection01 structure examples adapted from BANDA Release 1.1 Reference Manual Appendix. (a) Unprocessed superordinate directories. (b) Subdirectories for diffusion-weighted images. (c) Subdirectories for T1-weighted (top) and one run of the IPT task (bottom).

Fig. 5

fmriresults01 collection structure examples adapted from BANDA Release 1.1 Reference Manual Appendix. (a) Minimally preprocessed superordinate directories. (b) Subdirectories for diffusion-weighted outputs. (c) Subdirectories for T1-weighted outputs in native space. (d) Subdirectories for EPT outputs in standard spaces. (e) Subdirectories for T1-weighted outputs in standard spaces.

Fig. 6

Example structural QC output adapted from fmriresults01 collection. Left: T1-weighted (top) and T2-weighted volumes in native space with pial (blue) and white matter (green) surfaces overlaid. Right: Inflated Conte69 cortical surface with unsmoothed myelin map (top) and midthickness native surface with curvature map (bottom). QC snapshot images were edited for space and facial features were obscured.

Fig. 7

Phenotype labels projected onto two-dimensional representations of feature sets from select study elements. Dimensions extracted using UMAP for each feature set. Table 1 details elements used for feature sets.

Fig. 8

Bayes factors for phenotype effects on MRI-related quality metrics. Colors indicate measure type, bar shade indicates qualification of the strength of evidence for null or alternative hypothesis. Anatomical signal quality (green); anatomical quality ratings (brown); diffusion signal quality (blue); frame-wise displacement during functional imaging (yellow); motivation ratings during functional imaging tasks (purple); performance on functional imaging tasks (orange).

Fig. 9

Whole-brain, sample-wide effects for select contrasts from minimally-preprocessed task-fMRI outputs. Outputs were extracted from the fmriresults01 collection. Voxels shown exceeded p < 0.001 with a cluster-extent threshold of k = 100.

Fig. 10

Whole-brain, sample-wide effects from minimally-preprocessed resting-state fMRI outputs. Outputs were extracted from the fmriresults01 collection. Voxels shown reflect the top 95% of z-connectivity scores with a single precuneus voxel as the seed.