Evaluating task-evoked neurovascular coupling using integrated OPM-MEG and fNIRS imaging
收藏Mendeley Data2026-04-18 收录
下载链接:
https://data.mendeley.com/datasets/h52s5bvb43
下载链接
链接失效反馈官方服务:
资源简介:
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.
本数据集旨在无创评估人类大脑在自主运动任务期间的神经血管耦合(Neurovascular Coupling, NVC)。研究采用光泵磁力计(Optically Pumped Magnetometers, OPM)搭载脑磁图(Magnetoencephalography, MEG)技术记录神经活动,并通过功能性近红外光谱技术(Functional Near-Infrared Spectroscopy, fNIRS)采集血流动力学响应。数据显示,与任务相关的感觉运动振荡功率下降后,会伴随氧合血红蛋白浓度上升、去氧血红蛋白浓度下降,且具有神经血管耦合典型的稳定时间延迟特征。
数据采集自5名健康成年受试者,他们完成了自定节奏的右手运动任务(捏球与手指对指)。实验采用定制化集成OPM-MEG与fNIRS系统进行同步记录。为每位受试者配备基于MRI影像制作的3D打印头盔,以确保OPM传感器与fNIRS光极(Optodes)在运动与体感皮层上的放置稳定且解剖位置精准。所有测量均在德国柏林德国联邦物理技术研究院(Physikalisch-Technische Bundesanstalt, PTB)的BMSR2磁屏蔽室内完成。BMSR2是七层坡莫合金磁屏蔽室(Magnetically Shielded Room, MSR),可提供超低磁场环境,非常适配OPM-MEG系统的运行。
本次测量采用了不同配置的OPM传感器。2024年3月的记录使用18个OPM传感器(16个双轴QZFM-Gen2与2个三轴QZFM-Gen3;美国QuSpin公司),其中2个Gen3传感器作为参考传感器,分别安装在12cm与16cm延长杆上。2025年6月的记录则采用美国QuSpin公司的Neuro-1系统,搭载18个三轴QZFM-Gen3传感器并以闭环模式运行;其中1个传感器作为参考传感器,安装在16cm延长杆上。通过脑磁图数据采集(Data Acquisition, DAQ)系统,额外记录了3个通道:来自fNIRS设备的同步通道(用于确保多模态数据的时间对齐)、指示任务指令起始时刻的触发通道,以及连接压力传感器以检测捏球动作的压力通道。
血流动力学信号通过自研的光纤耦合fNIRS系统采集,该系统包含15个光源与9个探测器,采样率为15Hz。光源通过2.5mm光纤束发射750nm与850nm的光线。数据通过TCP/UDP协议实时传输至定制化MATLAB界面,用于可视化与存储。光极以网格形式排布,分析重点聚焦于第一近邻与第二近邻的源-探测器对(间距分别约为1.5cm与3cm)。
本数据集包含原始OPM-MEG与fNIRS信号、事件标记以及传感器元数据。OPM-MEG数据以Neuromag FIF格式提供,fNIRS数据以SNIRF格式提供。每位受试者均附带一份去面部标识的T1加权MRI影像。所有文件均采用MNE-BIDS工具生成的脑成像数据结构(Brain Imaging Data Structure, BIDS)兼容格式进行组织。
创建时间:
2025-12-12



