Characterizing oscillatory mechanisms of motor control: towards therapeutic closed-loop brain stimulation Sara J. Hussain, Ph.D. Human Cortical Physiology and Neurorehabilitation Section National Institutes of Neurological Disorders and Stroke National Institutes of Health School of Kinesiology University of Minnesota December 2019
Stroke is the leading cause of long-term disability in the United States. Although spontaneous motor recovery often occurs after stroke, this recovery is often incomplete. At the chronic stage, more than 40% of stroke survivors still have hemiparesis. Cramer et al. 1997
Transcranial Magnetic Stimulation (TMS)
“open-loop” brain stimulation ignores brain state “closed-loop” brain stimulation accounts for brain state Which brain state should we target? López-Alonso et al. 2016
The ideal brain state should: 1) be non-invasively detectable 2) be temporally well-defined 3) reflect functional state of the human motor system
Neural oscillations reflect rhythmic electrical activity generated by the central nervous system. 10 Hz high Described by: power 1) Frequency 10 Hz 2) Power low 3) Phase power 1 sec Cole and Voytek 2017
high Rest power preparation Movement low Oscillations recorded over power sensorimotor regions are most dominant between 8-12 Hz (mu). Movement execution low power high Rest power
Rec ecen ent work 1. Establish sensorimotor oscillatory contributions to corticospinal motor output. 2. Determine causal contributions of sensorimotor oscillatory brain states to skill learning.
Do sensorimotor oscillatory phase and power interact to determine human corticospinal output? 90° 90° Haegens et al. 2011; Hussain et al. 2019
peak trough trough peak p<0.001 p<0.001 power, p=0.513 phase, p=0.950 interaction, p=0.002 Hussain et al. 2019
Conclusions #1 Corticospinal motor output is determined by mu power during troughs. • Mu peaks reflect a neutral output state unaffected by power. • Phase-power interactions may be a general mechanism of brain function across multiple domains. • Given differences between mu peak and trough phases, these phases may differentially contribute to motor behavior.
Do motor cortical contributions to skill learning differ between mu peak and trough phases? mu mu mu Hussain et al., in prep
270° 270° 270° Hussain et al., in prep
Hussain et al., in prep
Closed-loop TMS during mu troughs improved consolidation by 54% (Cohen’s d=0.60, p<0.01). Closed-loop TMS during mu peaks improved consolidation by only 4% (Cohen’s d=0.06, NS). Hussain et al., in prep
Conclusions #2 Delivering closed-loop TMS to human motor cortex during mu trough but not mu peak phases facilitated • consolidation of a newly learned motor skill. First demonstration of phase-dependent enhancement of human skill memory. • Motor cortical activity causally contributes to skill consolidation during oscillatory trough phases only. • Neural mechanisms underlying skill consolidation fluctuate coherently with sensorimotor oscillations.
Overall conclusions Mechanisms of human corticospinal motor output and skill learning fluctuate coherently with sensorimotor • oscillations. Rather than being sustained over time, neural mechanisms of motor control are cyclic . • Delivering brain stimulation interventions during optimal phases may improve descending motor output and • skill learning after stroke.
Fu Future wo work 1. Characterizing cyclic mechanisms of skill learning in the healthy and lesioned brain 2. Synchronizing corticospinal oscillations to enhance human motor function
re-consolidation re-consolidation Day 1 Day 2 Day 3 consolidation memory memory memory acquisition re-activation re-activation This process can repeat indefinitely Cyclic mechanisms of skill acquisition and reconsolidation in healthy adults • Replicate and rescue post-stroke reconsolidation deficits using closed-loop TMS • Research Project Grant (R01, PA-18-345); A0 Sept 2021
In-phase = effective communication Out-of-phase = ineffective communication A A B B 5 ms 5 ms 5 ms 12 ms 12 ms 12 ms Entrain spinal motor neurons to oscillate synchronously and in-phase with sensorimotor oscillations. Enhance effects of existing corticospinal spike timing-dependent plasticity protocols: high versus low synchronization states • Closed-loop peripheral nerve stimulation to entrain spinal motor neurons to oscillate synchronously with cortical oscillations • NINDS Exploratory Neuroscience Research Grant (R21, PA-18-358); A0 Sept 2022 Womelsdorf et al. 2007
Re Research Program Mission Statement Seek understanding of the brain oscillatory mechanisms supporting normal and abnormal movement. Develop novel brain stimulation interventions that selectively target these mechanisms. Translate these interventions to restore motor function in patients with residual motor deficits due to neurological damage.
Acknowledgements Human Cortical Physiology and Neurorehabilitation Section (NINDS at NIH) Leonardo G Cohen, MD Ethan Buch, PhD Margaret Hayward, CNRP Marta Gozzi, PhD Marlene Bönstrup, MD Farah Fourcand, MD Gabriel Cruciani Ryan Thompson Katie Vollmer Jessica Stimely Brain Networks and Plasticity Lab (University of Tübingen, Germany) Ulf Ziemann, MD Christoph Zrenner, MD Funding NINDS Intramural Competitive Fellowship program
Recommend
More recommend
