0185

Emotion Self-Regulation Training with Simultaneous Real-Time fMRI and EEG Neurofeedback in Major Depression
Vadim Zotev1, Ahmad Mayeli1,2, Masaya Misaki1, and Jerzy Bodurka1,3

1Laureate Institute for Brain Research, Tulsa, OK, United States, 2Electrical and Computer Engineering, University of Oklahoma, Tulsa, OK, United States, 3Stephenson School of Biomedical Engineering, University of Oklahoma, Norman, OK, United States

Synopsis

We report a study of emotion self-regulation training in patients with major depressive disorder (MDD) using simultaneous real-time fMRI and EEG neurofeedback (rtfMRI-EEG-nf). Emotion-relevant target nf measures included fMRI activities of the left amygdala and left rACC, and frontal EEG asymmetries in the alpha and high-beta bands. MDD patients successfully learned to upregulate all four measures simultaneously using rtfMRI-EEG-nf during a happy emotion induction task. EEG-fMRI data analyses provided new insights into mechanisms of rtfMRI-EEG-nf. These findings may lead to development of more efficient neurotherapies for MDD.

Introduction

Simultaneous multimodal real-time fMRI and EEG neurofeedback (rtfMRI-EEG-nf)1-4 allows simultaneous modulation of both electrophysiological (EEG) and hemodynamic (BOLD fMRI) regional brain activities. Here we investigate efficacy and performance of rtfMRI-EEG-nf applied for training of emotion self-regulation in patients with major depressive disorder (MDD). In this study, MDD patients used rtfMRI-EEG-nf during a happy emotion induction task to simultaneously upregulate four target measures: BOLD activity of the left amygdala (LA), BOLD activity of the left rostral anterior cingulate cortex (left rACC), frontal alpha EEG asymmetry (FAA), and frontal high-beta EEG asymmetry (FBA)4. These activity measures are relevant to emotion regulation, and exhibit pronounced changes in MDD5-11. We demonstrate the feasibility and efficacy of their simultaneous modulation. We also examine therapeutic effects of the rtfMRI-EEG-nf and evaluate its potential in treatment of MDD.

Methods

Twenty four unmedicated MDD patients completed the study. Technical details of the rtfMRI-EEG-nf implementation and experimental procedure were described previously4. The experiments were performed on a GE MR750 3T MRI scanner with an 8-channel head coil. A gradient-echo EPI sequence with FOV/slice=240/2.9 mm, TR/TE=2000/30 ms, SENSE R=2, image matrix 96×96, flip=90°, 34 axial slices, was employed for fMRI. Simultaneous EEG recordings were performed using a 32-channel MR-compatible EEG system (Brain Products GmbH) with 0.1 µV resolution and 5 kS/s sampling, relative to FCz reference.

The rtfMRI-EEG-nf (Fig. 1A) was implemented using a custom multimodal real-time system1. The two rtfMRI-nf signals (red, orange bars) were based, respectively, on fMRI activations for the LA and left rACC ROIs (Fig. 1B). The two EEG-nf signals (purple, magenta bars) were based, respectively, on frontal EEG asymmetry changes in the high-beta (beta3) band, [21…30] Hz, and in the alpha band [7.5…12.5] Hz, for channels F3 and F4 (Fig. 1C). Real-time relative alpha asymmetry was defined as A = (P(F4)−P(F3))/(P(F4)+P(F3)), where P is EEG power in the alpha band. Real-time relative high-beta asymmetry was defined as B = (P(F3)−P(F4))/(P(F3)+P(F4)), where P is EEG power in the high-beta band. A custom procedure for improved real-time removal of EEG-fMRI artifacts (Fig. 1E) was implemented in BrainVision RecView software and employed a specially modified MR-compatible EEG cap4. The four nf bars were updated every 2 s.

The experimental protocol (Fig. 1D) included six runs, and each run (except Rest) consisted of 40-s-long blocks of Rest, Happy Memories, and Count conditions9-11. For each Happy Memories condition, the participant was asked to feel happy by evoking happy autobiographical memories, while trying to raise the levels of all four nf bars to the fixed target level (horizontal bar in Fig. 1A), which was raised from run to run. No nf bars were displayed during the Rest and Count conditions, and during the entire Transfer run.

Offline fMRI data analyses were performed in AFNI12, and EEG analyses – in BrainVision Analyzer 2. Normalized asymmetry for the alpha band was defined as FAA = ln(P(F4))−ln(P(F3)), where P is EEG power in the individual upper alpha band11. Normalized asymmetry for the high-beta band was defined as FBA = ln(P(F3))−ln(P(F4)). EEG-informed psychophysiological interaction (PPI)13 analyses were conducted using fMRI regressors based on FAA and FBA time courses1,11.

Results

MDD patients in the experimental group (n=16) were able to upregulate all four target measures (Fig. 2). All the results in Fig. 2 are significant when averaged across the four nf runs. The FAA changes exhibited positive correlations with the patients’ depression severity (HDRS) ratings (Fig. 3A,B), consistent with our previous study11. The FAA-based PPI interaction effect for Happy vs Count contrast was positive and significant for the LA (Fig. 3C, Fig. 4). It was also positive and significant for regions of the frontoparietal control network (FPCN)14, including the right middle frontal gyrus (BA 8,6,9), inferior parietal lobule (BA 40), precuneus, and others (Fig. 4). Similarly, the FBA-based PPI interaction effect was positive and significant for the LA (Fig. 3D, Fig. 5) and for the same FPCN regions (Fig. 5).

Conclusions

These results demonstrate that MDD patients can successfully learn to upregulate two fMRI and two EEG target measures relevant to MDD using the rtfMRI-EEG-nf. The EEG-informed fMRI analyses indicate that both target EEG asymmetries, FAA and FBA, were modulated simultaneously with the LA BOLD activity. Furthermore, the FAA- and FBA-based analyses independently reveal engagement of the FPCN regions, involved in goal-directed cognition14, and their interactions with the amygdala and related network. Upregulation of the FAA or FBA was more specifically associated with reduced activation of the right prefrontal regions, which may be beneficial to MDD patients8. Our findings show potential of the rtfMRI-EEG-nf in treatment of depression.

Acknowledgements

This work was supported by the Laureate Institute for Brain Research and the William K. Warren Foundation.

References

1. Zotev V, Phillips R, Yuan H, et al. Self-regulation of human brain activity using simultaneous real-time fMRI and EEG neurofeedback. NeuroImage 2014; 85:985.

2. Mano M, Lécuyer A, Bannier E, et al. How to build a hybrid neurofeedback platform combining EEG and fMRI. Front Neurosci 2017; 11:140.

3. Perronnet L, Lécuyer A, Mano M, et al. Unimodal versus bimodal EEG-fMRI neurofeedback of a motor imagery task. Front Hum Neurosci 2017; 11:193.

4. Zotev V, Phillips R, Misaki M, et al. Simultaneous real-time fMRI and EEG neurofeedback for emotion regulation training in depressed patients. Proc ISMRM 2016; 24:1399.

5. Price JL, Drevets WC. Neurocircuitry of mood disorders. Neuropsychopharmacol 2010; 35:192-216.

6. Pizzagalli DA. Frontocingulate dysfunction in depression: toward biomarkers of treatment response. Neuropsychopharmacol 2011; 36:183-206.

7. Thibodeau R, Jorgensen RS, Kim S. Depression, anxiety, and resting frontal EEG asymmetry: a meta-analytic review. J Abnorm Psychol 2006; 115:715-729.

8. Pizzagalli DA, Nitschke JB, Oakes TR, et al. Brain electrical tomography in depression: the importance of symptom severity, anxiety, and melancholic features. Biol Psychiatry 2002; 52:73-85.

9. Zotev V, Krueger F, Phillips R, et al. Self-regulation of amygdala activation using real-time fMRI neurofeedback. PLoS ONE 2011; 6:e24522.

10. Zotev V, Phillips R, Young KD, et al. Prefrontal control of the amygdala during real-time fMRI neurofeedback training of emotion regulation. PLoS ONE 2013; 8:79184.

11. Zotev V, Yuan H, Misaki M, et al. Correlation between amygdala BOLD activity and frontal EEG asymmetry during real-time fMRI neurofeedback training in patients with depression. NeuroImage Clin 2016; 11:224-238.

12. Cox RW. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res 1996; 29:162-173.

13. Friston KJ, Buechel C, Fink GR, et al. Psychophysiological and modulatory interactions in neuroimaging. NeuroImage 1997; 6:218-229.

14. Spreng RN, Sepulcre J, Turner GR, et al. Intrinsic architecture underlying the relations among the default, dorsal attention, and frontoparietal control networks of the human brain. J Cogn Neurosci 2013; 25:74-86.

Figures

Figure 1. A) Multimodal neurofeedback GUI screen with two variable-height EEG-nf bars and two rtfMRI-nf bars. B) Left amygdala (LA) and left rostral anterior cingulate cortex (L rACC) target ROIs for rtfMRI-nf, defined as 14-mm-diameter spheres centered at (−21,−5,−16) and (−3,+34,+5) in the Talairach space. C) EEG channels used to provide EEG-nf. D) Experimental protocol with six runs, abbreviated as RE, PR, R1, R2, R3, TR. The five task runs consist of Rest (R), Happy Memories (H), and Count (C) condition blocks. E) Custom real-time EEG-fMRI artifact removal procedure. F) Modified 32-ch EEG cap for improved EEG-nf during fMRI.

Figure 2. Mean changes in the EEG-nf and rtfMRI-nf target activity measures during the Happy Memories (H) conditions relative to control conditions (Rest, Count) for the four rtfMRI-EEG-nf runs and the Transfer run without nf. The results are from offline data analyses. FAA – frontal EEG asymmetry, ln(P(F4))−ln(P(F3)), in the upper alpha band [IAF…IAF+2] Hz, where IAF is the individual alpha peak frequency. FBA – frontal EEG asymmetry, ln(P(F3))−ln(P(F4)), in the high-beta band [21…30] Hz. LA – left amygdala target ROI. L rACC – left rACC target ROI.

Figure 3. A) Positive correlation between the FAA changes and individual depression severity (HDRS) ratings during the Practice run. B) Positive correlation between the FAA changes, averaged across the four nf runs, and the HDRS ratings. C) Mean values of the FAA-based PPI interaction effect for Happy vs Count contrast for the LA ROI. D) Mean values of the FBA-based PPI interaction effect for the LA ROI.

Figure 4. Statistical map of the FAA-based PPI interaction effect for the Happy vs Count contrast. A positive effect means a stronger temporal correlation between the FAA time course (upper alpha band) and BOLD activity during Happy conditions with rtfMRI-EEG-nf compared to Count conditions. The results are shown in the Talairach space. Green arrows point brain regions with overlapping effects in Figs. 4 and 5.

Figure 5. Statistical map of the FBA-based PPI interaction effect for the Happy vs Count contrast. A positive effect means a stronger temporal correlation between the FBA time course (high-beta band) and BOLD activity during Happy conditions with rtfMRI-EEG-nf compared to Count conditions. The results are shown in the Talairach space. Green arrows point brain regions with overlapping effects in Figs. 4 and 5.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)
0185