BIDS Extension for Music and Audio Data in Neuroscience Research

[pawanwaldia]

Your idea

Title:
BIDS Extension for Music and Audio Data in Neuroscience Research

Introduction:
Dear BIDS community, I am Pawan Waldia, a Master’s student working under the supervision of Samuel Morgan, doctoral researcher at the University of Stuttgart and a member of Lullabyte, a multidisciplinary doctoral network which explores the effects of music on sleep using computational techniques. Within this network, there are several neuroimaging studies being conducted which use musical/auditory stimuli and recordings, but BIDS does not currently accommodate such data types, forcing researchers to use their own BIDS-like solutions. We would therefore like to propose a new BIDS extension to standardize how music and audio data can be organized, stored, and shared within neuroscience research.

Motivation:
Currently, there is no standardized structure within BIDS which accounts for music and audio stimuli or recordings. This lack of standardization limits reproducibility, hampers data sharing, and complicates automated analysis. By developing this extension, we aim to address these challenges, creating consistency across related datasets and facilitating easier analysis and collaboration.

Scope:
This extension aims to:

Develop a flexible specification for music/audio data and metadata storage within BIDS.
Engage with the neuroscience research community for ongoing refinement.
Ensure compatibility with existing BIDS standards and tools.
Provide validation mechanisms tailored specifically for music/audio data.

Preliminary Work:
Together with another BIDS extension project on the Lullabyte network, we have already conducted a small-scale survey of the research community (https://ipvs.informatik.uni-stuttgart.de/sc/limesurvey/index.php/763246?lang=en ). The goal of this was to assess researcher needs, data formats, and current limitations which would inform the development of the extension. Results from the survey demonstrated the necessity of this proposed extension and provided guidance for initial design considerations.

The key findings from our survey that support this proposal are:
Among the respondents (n=10) who indicated relevant experience ( working in neuroimaging-based music/audio research), 70% rated themselves as likely to adopt a BIDS extension for musical and auditory data (mean likelihood score: 3.9 out of 5).
Key BIDS features seen as useful by music/audio researchers include:
Facilitating data sharing and collaboration (90%).
Metadata organization and accessibility (80%).
Support for multimodal data formats, including audio (70%).
Researchers expressed a need for standardization, with 70% rating standardized structures for automated pipelines as extremely important.

Planned Deliverables:
A detailed BIDS Extension Proposal document (initially a collaborative Google Doc, then a finalized Markdown specification).
Example ‘dummy’ datasets, demonstrating practical usage of the new specification.
Updates to the bids-validator tool for compliance checking.

Use Cases:
This BIDS extension will directly benefit:

Neuroscience researchers studying music cognition, and perception.
Labs utilizing auditory stimuli in EEG, fMRI, MEG, or behavioral studies.
Researchers working with multimodal datasets involving recorded audio data, such as verbal interviews or vocal responses to stimuli.
Developers building audio feature extraction and analysis pipelines for neuroscience applications.

Community Involvement:
We welcome input and participation from researchers, data curators, and developers interested in neuroscience audio research. A collaborative Google Doc for community feedback will be available soon, alongside discussions here on GitHub.

Lead Contributors:

Name: Pawan Waldia
Affiliations: IPVS (University of Stuttgart - https://www.ipvs.uni-stuttgart.de/de/)
Email: pawan.waldia3@gmail.com

Name: Samuel Morgan (https://www.ipvs.uni-stuttgart.de/departments/sc/department/team/)
Affiliations: IPVS (University of Stuttgart - https://www.ipvs.uni-stuttgart.de/de/), Lullabyte MSCA Doctoral Network (https://lullabyte.eu/)
Email: samuel.morgan@ipvs.uni-stuttgart.de

Thank you, we look forward to the community’s valuable input and collaboration!