A test-retest fMRI dataset for motor, language and spatial attention functions

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Medical Imaging

Background:

• Functional magnetic resonance imaging (fMRI) is widely used to study human cognition.
• Between-session variability in fMRI tasks is understudied but critical for clinical use.
• A test-retest dataset was created to assess fMRI task reliability for pre-surgical planning.

Purpose of the Study:

• To validate fMRI tasks for reliable single-subject results in pre-surgical planning.
• To investigate the reliability of different task-based fMRI protocols.
• To identify confounding factors affecting fMRI data reliability.

Main Methods:

• Acquired test-retest fMRI data from ten healthy participants (50s) scanned twice, 2-3 days apart.
• Utilized five task-related fMRI protocols: finger, foot, lip movement, overt/covert verb generation, overt word repetition, and landmark tasks.
• Acquired diffusion tensor MRI (DTI) and high-resolution 3D T1-weighted structural scans.

Main Results:

• Motor and language fMRI tasks demonstrated reliable subject-level results.
• The landmark task showed lower reliability despite expected group-level activations.
• Task-specific differences and task-by-motion interactions were primary factors influencing reliability.

Conclusions:

• The dataset enables investigation into the reliability of various fMRI tasks.
• It facilitates the study of analysis, de-noising, and data integration methods for fMRI, DTI, and structural MRI.
• Findings highlight the importance of task selection and motion control for reliable fMRI in clinical settings.