Recommended Programs & Utilities🔗

Download Platforms🔗

HBCD study data can be downloaded via the NBDC Data Access Platform or DEAP- see Access & Download Data.

Tabulated Data🔗

NBDC Sandbox🔗

The NBDC Sandbox provides a secure, cloud-based analysis environment designed to support data analysis workflows, particularly for neuroimaging and large-scale tabular datasets, without having to download and manage data locally. See the NBDC Sandbox Tutorial Series to get started.

NBDCtools🔗

NBDCtools, available in both R and Python, is a package for creating custom, analysis-ready datasets by simply specifying the variable or table names you need. NBDCtools automatically retrieves the specified columns from locally downloaded HBCD tabulated data and assembles them into a single in-memory data frame, minimizing storage and memory use. This provides a flexible alternative to building datasets through the NBDC Data Access Platform or DEAP, eliminating the need to manually parse files or resolve formatting issues. In addition to dataset assembly, NBDCtools includes functions for working with shadow matrices as well as applying transformations and filters.

Brain Imaging Data🔗

ITK-Snap & FSLeyes🔗

For interactive visualization of MRI images, we recommend using either ITK-Snap (see Andy's Brain Blog for a primer) or FSLeyes, part of the FSL software suite. Both are free, open-source image viewers for 3D and 4D medical images.

Connectome Workbench🔗

Connectome Workbench is a free, open-source software package that includes critically useful tools such as:

• wb_view: For interactive visualization of surface-based and connectivity data, creating overlays of structural/functional data, etc.
• wb_command: A set of command-line tools for processing and analysis of neuroimaging data (see usage)

NMIND🔗

NMIND is a collaborative initiative dedicated to improving transparency, reproducibility, and efficiency in neuroimaging research. NMIND principles, standards, and tools were used to develop the HBCD Processing & Derivative Data Standards. Explore a growing collection of tools tested and improved through the NMIND process at Evaluated Tools.

ReproSchema🔗

HBCD-ReproSchema

Consistent data collection is essential to reproducible longitudinal research, particularly for large, multi-site studies like HBCD. ReproSchema (https://www.repronim.org/reproschema) provides both a schema and software platform for structuring, versioning, and managing research questionnaires over time, including integration with survey and data collection platforms like REDCap.

Schema🔗

As a schema, ReproSchema defines a standardized structure for questionnaires, including consistent formatting across time points, with explicitly defined questions, response options, and skip logic linked metadata. Questionnaires are organized following a hierarchical model with three main components:

Protocol (all study instruments) → Activity (single instrument) → Item (single question within instrument)

Platform🔗

As a platform, ReproSchema provides tools for automated storage, versioning in GitHub, tracking changes (e.g. wording updates, added/removed/re-ordered items, updated skip logic, etc.), and comparing questionnaires over time. It also provides a utility for browser-based survey deployment, with survey responses stored in JSON-LD format to link each answer to its protocol, activity, and item in the ReproSchema library. See Chen et al. 2025 Figure 1 for details.

Tracking Changes: Example🔗

ReproSchema maintains transparent, detailed version histories so researchers can understand how questionnaires evolve between releases. For example, the following shows how a sleep question and its response options may evolve across releases. ReproSchema versioning captures these changes to allow researchers to adjust longitudinal analyses as needed.

Release	Item Question	Response Options	Implications for Analysis
1.0	How many hours do you sleep on a typical night?	Free-text numeric entry	-
2.0	On average, how many hours of sleep do you get per night?	Dropdown menu options (e.g. Less than 5, 5-6, etc.)	1.0 → 2.0: Categorical response options reduce variability, but lose detail
3.0	On average, how many hours of sleep do you get in a 24-hour period, including naps?	Same as Release 2.0	2.0 → 3.0: Adding naps changes the construct, affecting cross-release comparability