7T functioanl MRI reveals frequency-dependent responses during vibrotactile stimulation at somatosensory cortex
Boyi Qu1,2, Xiao Yu1,3, Kaiyue Wang1,2, Mengjie Xin1, and Hsin-Yi Lai2,4

1Interdisciplinary Institute of Neuroscience and Technology, Qiushi Academy for Advanced Studies, Zhejiang University, Hangzhou, China, 2College of Biomedical Engineering & Instrument Science´╝îZhejiang University, Hangzhou, China, 3Department of Neurobiology, Zhejiang University School of Medicine, Hangzhou, China, 4The Second Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China


To investigate whether the brain can directly encode different vibrotactile frequencies, we made a MR-compatible lab-designed vibrotactilestimulator to presented pulse stimuli on middle finger (with a frequency difference in the same duration) andusing BOLD fMRI to explore the effectsofvibrotactilefrequencyin somatosensory cortex with 7T MR research system. The results showedthat the lab-designed MR compatible vibrotactile stimulator could elicit BOLD responsesin the somatosensory cortex. Also, the BOLD response was frequency-dependent with the peak at 6Hz.


Primary somatosensory cortex(S1)is sensitive to the tactile frequency1.Mapping functional location of the S1 is necessary for understanding of the properties of tactile perception in human2-4. Generally, the ultra-high field functional magnetic resonance imaging (fMRI) could provide the reliable cortical responses of digits in the S1 because of the higher signal-to-noise ratioand spatial resolution in blood-oxygen-level dependent (BOLD) signals. The previous study demonstrated BOLD fMRIisa valid tool for mapping the somatosensory systemand the study mentioned thatit is unlear aboutwhat the effect of vibrotactile frequency responded to the cortical function in S1 and how did the BOLD signals present the effects of vary frequency in vibrotactile.The goal of this study is to make MR-compatible lab-designed vibrotactilestimulator and combined with ultra-high field7T research MRI for investigating the vibrotactile frequencyin human S1.


Ten right-handed healthy volunteers (9 females, 1 male, Age: 25.43 ±3.55 years)were recruitedin this study. The lab-designed MR-compatiblevibrotactile stimulator is consist of a piezoelectric device connected to a 2-mm diameter round plastic probewhich implement vibrotactile stimulation to the tip of middle finger, as shown in Figure1. Stimulation frequencies of 1, 6, 12, 24, and 48 Hz were used with a pulse width of 15 ms. The stimulation parameters were sequenced in a pseudo-random manner. Two to three repeated trials were performed to improve measurement accuracy and optimize SNR. The stimulation paradigm was 60 s initial rest, 30 s stimulation, followed by 60 s rest and an additional 5 min minimum resting interval between trials.All scans were performed on a 7T research system (Siemens, Erlangen, Germany). Anatomical images were acquired by a prototype MP2RAGE sequence (TR = 5000 ms, TE= 2.27 ms, BW= 300Hz, voxel size: 0.7×0.7×0.7 mm3) and BOLD fMRI were acquired by a prototype multi-band EPIsequence (TR = 2000 ms, TE = 20.6 ms, voxel size: 1.5×1.5×1.5 mm3, BW= 1732 Hz).The BOLD fMRI signals, acquired during tactile stimulation,were analyzed by astandard processing using an open source software AFNI.Activation was defined by voxels that exhibited significantly correlated BOLD signal changes (p < 0.001, FDR corrected). fMRI activation maps (with statistical t-values, typically thresholded at p=0.0001, FDR corrected) were spatially interpolated and then superimposed on the corresponding high-resolution T1 anatomical images.

Results and Discussion

All frequencies induced the BOLD activated during the stimulation period and the responses were located along the posterior of the central sulcus (Figure2),primary somatosensory cortex (p<0.001).The Grand-averaged BOLD time courses to 5vibrotactile frequencies, as show in Figure 3 (A-E). The BOLD responses exhibited a tuning curve shape that rapidly descended from 1Hz to 6Hz and ascended gradually from 12 Hz to 48 Hz(Figure 3(F)). The BOLD responses of 1 Hz stimulationwas significantly higher than those of other frequencies, and the recovery time waslonger thanthose ofother frequencies. The resultsmay be thelow frequency stimulation consumed less oxygen, so the time of BOLD signals took longer time to reach baseline, as compare with higher frequency stimulus.


This study demonstratedBOLD response was frequency-dependent and the low-frequency stimuli may affect the recover time of BOLD curve.This MR-compatible lab-designed vibrotactilestimulatorcould be a useful tool to evaluate brain functional recoverin patients withstrokein the future.


No acknowledgement found.


1. YG Chung, J Kim, SW Han, HS Kim SP Kim et al. Frequency-dependent patterns of somatosensory cortical responses to vibrotactile stimulation in humans: A fMRI study[J]// Brain research, 2013:47-57.

2.YH Li, Ying Lee, Wolfgang Grodd, Christoph Brarun. Comparing tactile pattern and vibrotactile frequency discrimination: a human fMRI study[J]//Journal of neurophysiology, 2010.

3.Renate Schweizer, Dirk Voit, and Jens Frahm. Finger representations in human primary somatosensory cortex as revealed by high-resolution functional MRI of tactile stimulation[J]// Neruoimage, 2008:28-35

4.Christian Kalberlah, ArnoVillringer, Burkhard Pleger. Dynamic causal modeling suggests serial processing of tactile vibratory stimuli in the human somatosensory cortex—An fMRI study[J]// Neruoimage, 2013:164-171


Figure 1.Participants placed their left middle finger on the MR compatible tactile stimulator.

Figure 2. Results of functional mapping. The frequency-dependent cortical activation patterns of a representative subject in response to vibration 6Hz.

Figure 3. Results of BOLD signals. With yellow bars indicating the period where stimulation was applied at the fMRI

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)