Back to datasets
Dataset assetOpen Source CommunityNeuroimagingFunctional Magnetic Resonance Imaging
BOLD5000
The BOLD5000 dataset contains image data of brains, objects, and landscapes for functional task acquisition, including MRI and fMRI data from multiple participants. The dataset records detailed scene data, anatomical data, and preprocessing steps for each functional scan.
Source
github
Created
Sep 12, 2018
Updated
Mar 24, 2022
Signals
205 views
Availability
Linked source ready
Overview
Dataset description and usage context
Dataset Overview
Dataset Name
BOLD5000: Brains, Objects, Landscapes Dataset
Participant Information
- Four participants: CSI1, CSI2, CSI3, & CSI4
- Functional task acquisition sessions: total of 15 sessions (#1-15)
- Anatomical data acquisition session: #16
Data Content
- Functional session data:
- Three sets of field maps (EPI opposite phase encoding; spin‑echo opposite phase encoding pairs, including partial and non‑partial Fourier)
- 9 or 10 functional scans covering 5,000 scene data (5000scenes)
- 1 or 0 functional localizer scans for defining scene‑selective regions (localizer)
- Each event.json file lists each stimulus, stimulus onset time, and participant response (participants performed a simple affective task)
- Anatomical data:
- T1‑weighted MPRAGE scans
- T2‑weighted SPACE scans
- Diffusion spectrum imaging
Data Processing
- Preprocessing: All functional data were preprocessed using fMRIprep, including motion correction, susceptibility distortion correction, and alignment to anatomical data.
- Derived data directory contents:
- fMRIprep: preprocessed data and reports
- Freesurfer: reconstruction results after fMRIprep preprocessing
- MRIQC: image quality metrics (IQM) obtained using MRIQC
- spm: directory containing masks used to define regions of interest (ROI) for each participant
Special Notes
- All MRI and fMRI data are provided after Siemens pre‑scan normalization filter processing.
- CSI4 participated in only 10 MRI sessions; sessions 1‑9 were functional acquisition sessions, and session 10 was an anatomical data acquisition session.
Need downstream help?
Pair the dataset with AI analysis and content workflows.
Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.