EKG‐based detection of deep brain stimulation in fMRI studies. Issue 4 (2nd August 2017)

• Record Type: Journal Article
• Authors: Fiveland, Eric; Madhavan, Radhika; Prusik, Julia; Linton, Renee; Dimarzio, Marisa; Ashe, Jeffrey; Pilitsis, Julie; Hancu, Ileana
• Is Part Of: Magnetic resonance in medicine. Volume 79:Issue 4(2018)
• Page Start: 2432
• Page End: 2439
• Publication Date: 2017-08-02
• Subjects: deep brain stimulation -- DBS -- MRI -- fMRI -- ECG -- EKG; Nuclear magnetic resonance -- Periodicals; Electron paramagnetic resonance -- Periodicals; 616.07548
• DOI: 10.1002/mrm.26868
• Languages: English
• ISSNs: 0740-3194

Abstract

Abstract : Purpose: To assess the impact of synchronization errors between the assumed functional MRI paradigm timing and the deep brain stimulation (DBS) on/off cycling using a custom electrocardiogram‐based triggering system Methods: A detector for measuring and predicting the on/off state of cycling deep brain stimulation was developed and tested in six patients in office visits. Three‐electrode electrocardiogram measurements, amplified by a commercial bio‐amplifier, were used as input for a custom electronics box (e‐box). The e‐box transformed the deep brain stimulation waveforms into transistor‐transistor logic pulses, recorded their timing, and propagated it in time. The e‐box was used to trigger task‐based deep brain stimulation functional MRI scans in 5 additional subjects; the impact of timing accuracy on t‐test values was investigated in a simulation study using the functional MRI data. Results: Following locking to each patient's individual waveform, the e‐box was shown to predict stimulation onset with an average absolute error of 112 ± 148 ms, 30 min after disconnecting from the patients. The subsecond accuracy of the e‐box in predicting timing onset is more than adequate for our slow varying, 30‐/30‐s on/off stimulation paradigm. Conversely, the experimental deep brain stimulation onset prediction accuracy in the absence of the e‐box, which could be off by as much as 4 to 6 s, could significantly decrease activation strength. Conclusions: Using this detector, stimulation can be accurately synchronized to functional MRI acquisitions, without adding any additional hardware in the MRI environment. Magn Reson Med 79:2432–2439, 2018. © 2017 International Society for Magnetic Resonance in Medicine.