Datensatz

Zusammenfassung

Purpose/Introduction: Time-series of DWI (diffusion-weighted imaging) have shown large potential to image focal spontaneous mechanical activities in musculature (SMAM) at rest [1]. A high correlation between SMAMs and electrical activities measured by surface electromyography (sEMG) of the musculature was already shown in retrospective measurements [2, 3]. However, the sensitivity of DW sequences varies and depends on the diffusion preparation scheme [4, 5]. In order to optimize the imaging process and overcome sensitivity limitations of DW sequences, the feasibility of prospective measurements by triggering the DWI on sEMG activities was investigated. Subjects and Methods: Concurrent sEMG and DWI were acquired of two healthy volunteers (age: 51.5 ± 3.5, BMI: 26.6 ± 1.7 kg/m2) at the position of maximum diameter of the right calf on a 3 T MR scanner (MAGNETOM Prisma fit) with an MR-compatible system (BrainAmp ExG MR). The system architecture is depicted in Fig. 1. DWI: diffusion-weighted Stejskal-Tanner spin-echo EPI [6], matrix: 64 9 64, FoV: 192 9 192 mm2, TE = 36 ms, b = 100 s/mm2, repetitions: 100, BW = 2005 Hz/px, physiological trigger input: external. sEMG: bi-polar channels: 1, fsampling: 5 kHz, resolution: 0.5 lV, inter-electrode-distance: 2 cm, current-limiting resistor: 20/20 kX. Signal analysis: Software was implemented in C based on the remote data access functionality (BrainVision Recorder). Highfrequency noise, power supply distortions and physiological signals were filtered by a 20th order real-time filter (IIR Cascaded Second-Order Sections Direct-Form II) with fPass = 20–500 Hz and fNotch = 50 Hz. To suppress event detection during MR gradient switching a dead-time of tD = 200 ms was chosen as minimum interevent time. sEMG activities were detected in real-time by applying a threshold at 30 lV. Evaluation: All DWI were slice-wise normalized and registered [7, 8]. Average delay between measured sEMG activities and onset of DWI was determined within EEGLAB [9]. Results: An average systemic delay of 98 ± 13 ms between time points of sEMG activities and onsets of DWI was measured. 82.4 of all sEMG-triggered DWI have shown intensity changes in focal muscular regions. In Fig. 2, an sEMG signal with two small spontaneous electrical activities and successive triggered DWI is given. Exemplary DWIs without and with trigger event are illustrated in Fig. 3. Discussion/Conclusion: The feasibility of sEMG-triggered DWI for prospective imaging of unconscious spontaneous focal muscle contractions (SMAMs) was shown. This technique potentially enables an optimized strategy to image SMAMs in resting musculature as well as the assessment of the contraction time of this specific motion.