Deep brain electrical neurofeedback allows Parkinson patients to control pathological oscillations and quicken movements

Abstract

Parkinsonian motor symptoms are linked to pathologically increased beta-oscillations in the basal ganglia. While pharmacological treatment and deep brain stimulation (DBS) reduce these pathological oscillations concomitantly with improving motor performance, we set out to explore neurofeedback as an endogenous modulatory method. We implemented real-time processing of pathological subthalamic beta oscillations through implanted DBS electrodes to provide deep brain electrical neurofeedback. Patients volitionally controlled ongoing beta-oscillatory activity by visual neurofeedback within minutes of training. During a single one-hour training session, the reduction of beta-oscillatory activity became gradually stronger and we observed improved motor performance. Lastly, endogenous control over deep brain activity was possible even after removing visual neurofeedback, suggesting that neurofeedback-acquired strategies were retained in the short-term. Moreover, we observed motor improvement when the learnt mental strategies were applied 2 days later without neurofeedback. Further training of deep brain neurofeedback might provide therapeutic benefits for Parkinson patients by improving symptom control using strategies optimized through neurofeedback.

Conflict of interest statement

The authors declare no competing interests.

Figures

Figure 1

Deep brain electrical neurofeedback loop. Deep brain stimulation electrodes (upper left corner) implanted into the subthalamic nucleus allowed for brain oscillation measurements. The raw, analogue signal was amplified and then digitally band-pass filtered around the individual beta-peak (± 5 Hz), rectified, averaged and used to control the position of the blue disc on the monitor (upper right corner). Patients tried to volitionally control the position of the disc, thereby closing the neurofeedback loop while controlling pathological beta-activity.

Figure 2

Learning to downregulate beta activity with DBS-neurofeedback. Beta-power estimates from all patients during downregulation—normalised to the beta-power estimate during each patients’ preceding rest block—are shown for the baseline (pre-NF), neurofeedback (NF1, NF2, NF3) and transfer rounds. We indicate the cumulative amount of time tcum that patients have spent learning downregulation through neurofeedback until the end of that respective round. The group means are represented by the horizontal red lines, the standard deviations by the vertical blue lines and the 95% confidence intervals by the red patched areas. We used the two-sided Student’s t-tests to test for significant beta-reductions compared to the baseline rest (horizontal black line at 1). Wilcoxon’s signed rank test was used to compare the dependent samples from the transfer round with their pre-NF value. *p < 0.05; ***p < 0.001.

Figure 3

Relationship between disease burden and beta-modulatory capacity. We correlated the MDS-UPDRS III OFF score with the beta-modulatory capacity of each patient (n = 8) and found a significant relationship (p = 0.0083) with a positive Pearson correlation coefficient (R = 0.84).

Figure 4

Learning bidirectional neurofeedback. Beta-power estimates from all patients during upregulation—normalised to the beta-power estimate during each patients’ downregulation condition—for the Pre-NF, NF1, NF2, NF3 and Transfer rounds show increasing beta activity upon neurofeedback aided upregulation. We indicate the cumulative amount of time tcum that patients spent learning upregulation through neurofeedback until the end of that respective round. The means are represented by the horizontal red lines, the standard deviations by the vertical blue lines and the 95% confidence intervals by the red patched areas. We used two-sided Student’s t-testing to determine significant beta-increases compared to the downregulation condition (horizontal black line at 1). *p < 0.05.

Figure 5

Behavioural output metrics during the first 12 s of pronosupination after downregulation during NF3 and short-term (2 days) mental strategy transfer. (A) The pronosupination frequency (proportional to the number of cycles during the first 12 s of pronosupination) was significantly increased after downregulation as compared to rest (p = 0.0312 and p = 0.0156 during NF3 and 2 d mental strategy transfer, respectively) while (B) the mean peak angular amplitude was only slightly and non-significantly decreased. (C) The cumulative angular displacement (i. e. degrees travelled) was generally higher after downregulation as compared to rest. The grey rectangle highlights the mutual dependence between the three variables in (A–C), i. e. frequency * mean peak angular amplitude = cumulative angular displacement. (D) The torque generated by the antagonistically acting pro- and supinator muscles was estimated through the mean absolute angular acceleration, which was significantly increased after downregulation as compared to rest during the neurofeedback experiment (p = 0.0469).