Evaluating task-evoked neurovascular coupling using integrated OPM-MEG and fNIRS imaging
Description
This dataset was acquired to non-invasively evaluate neurovascular coupling (NVC) in the human brain during voluntary motor tasks. Neural activity was measured with magnetoencephalography (MEG) using optically pumped magnetometers (OPM), and hemodynamic responses were measured using functional near-infrared spectroscopy (fNIRS). The data demonstrate that task-related decreases in sensorimotor oscillatory power are followed by increases in oxy-hemoglobin and decreases in deoxy-hemoglobin, with a consistent temporal delay characteristic of NVC. Data were collected from five healthy adult participants performing self-paced right-hand motor tasks (ball squeezing and finger opposition). Simultaneous recordings were obtained using a custom-built, integrated OPM-MEG and fNIRS system. For each subject, an MRI-derived 3D-printed helmet was used to ensure stable and anatomically accurate placement of OPM sensors and fNIRS optodes over the motor and somatosensory cortices. All measurements were conducted inside the BMSR2 at the Physikalisch-Technische Bundesanstalt (PTB) in Berlin, Germany. The BMSR2 is a seven-layer mu-metal magnetically shielded room (MSR) that provides an ultra-low magnetic field environment, making it ideal for the operation of OPM-MEG systems. Different configurations of OPM sensors were available for the measurement sessions. In March 2024, recordings used 18 OPM sensors (16 dual-axis QZFM-Gen2 and two triple-axis QZFM-Gen3; QuSpin, USA), with two Gen3 sensors serving as reference sensors and mounted on 12 cm and 16 cm extenders. In June 2025, recordings were performed using the Neuro-1 system (QuSpin, USA) with 18 triple-axis QZFM-Gen3 sensors operated in closed-loop mode; one sensor served as a reference and was mounted on a 16 cm extender. With the MEG data acquisition (DAQ) system, three additional channels were recorded: a synchronization channel from the fNIRS device to ensure temporal alignment between modalities; a trigger channel indicating the onset of the task command; and a pressure channel connected to a pressure sensor measuring ball squeezes. Hemodynamic signals were recorded using an in-house built fibre-coupled fNIRS system with 15 sources and 9 detectors operating at 15 Hz. Sources emitted light at 750 nm and 850 nm through 2.5 mm fiber bundles. Data were streamed in real time via TCP/UDP to a custom MATLAB interface for visualization and storage. The optodes were arranged in a grid, and analysis focused on the first- and second-nearest neighbor source–detector pairs (approximately 1.5 cm and 3 cm, respectively). The dataset includes raw OPM-MEG and fNIRS signals, event markers, and sensor metadata. OPM-MEG data are provided in Neuromag FIF format and fNIRS data in SNIRF format. For each participant, a defaced T1-weighted MRI is provided. All files are organized in a BIDS-compatible structure created with MNE-BIDS.
Files
Steps to reproduce
Detailed steps to reproduce the measurements and preprocessing pipeline are provided in Tappayuthpijarn, P., Wojtkiewicz, S., Sawosz, P., Jazbinšek, V., Liebert, A., Sander, T., Marhl, U., 2025. Evaluating task-evoked neurovascular coupling using integrated OPM-MEG and fNIRS imaging. NeuroImage 121644. https://doi.org/10.1016/j.neuroimage.2025.121644
Institutions
- Physikalisch-Technische Bundesanstalt
- Polska Akademia Nauk Instytut Biocybernetyki i Inzynierii Biomedycznej
- Instituta za matematiko fiziko in mehaniko