A new MRI subject position to explore simultaneous BOLD oscillations of the brain and the body

https://doi.org/10.1016/j.jneumeth.2020.108829Get rights and content

Highlights

  • Arm-over-head MRI position was developed to study EPI BOLD in and outside the brain.

  • Simultaneous FMRI of the brain and limb were collected with decent SNR.

  • Motor imagery tasks revealed ICA components and FC across the brain and limb.

  • Existence of extensive whole-body functional network(s) is rational and possible.

Abstract

Background

Anatomically and physiologically, there is strong relationship between the brain and body. A new MRI platform covering both the brain and the limb would be beneficial for a more thorough understanding of the brain-body interactions.

New Method

A new arm-over-head (AOH) position was developed to collect MRI of the brain and one arm simultaneously. Subject’s tolerability and SNR of both the brain and limb under a serial of seven different TR (250–3000 ms) were tested. Then, blocked motor imagery tasks were performed to test the possible brain-body oscillations.

Results

The new MRI position provided structural images with good quality, and the AOH position had the best SNR under TR 3000 ms (p = 0.03 for the brain; p = 0.064 for the limb). Then, by using both hypothesis-free independent component analysis (ICA) and a priori seed-based functional connectivity (FC) analysis, it is demonstrated during motionless motor imagery tasks there existed possible brain-body BOLD oscillations connecting especially arm flexors to default mode, vision, and sensorimotor networks. The FC appeared at network density as low as 5%.

Comparison with Existing Methods

We have developed a new MRI subject position to explore the possibilities of more extensive neuronal and physiological networks.

Conclusions

The results of this preliminary experiment indicate that functional brain networks might extend outside the brain. A bottom-up circulatory effect might explain this phenomenon. Nonetheless, considering the mechanism of neural top-down control and the nature of complex brain networks, the existence of a more extensive whole-body functional network is rational and possible.

Introduction

MRI has been widely used in medicine since its development in the 1980s. The clinical applications are more on structural diagnosis in various fields, including neurology, cardiology, musculoskeletal, as well as detection of many cancers. Another extended application, namely, FMRI, is more commonly used for research purpose, predominantly in the central nervous system (CNS), and the ground theory of using FMRI as an investigational device is based on BOLD signal changes observed on FMRI images secondary to neuronal activities. BOLD reflects the blood ratio of oxyhemoglobin and deoxyhemoglobin (Ogawa et al., 1990) and is especially highly correlated with local field potentials (LFPs) (Logothetis et al., 2001).

Traditionally, FMRI studies focus on task-related, activation paradigm designed, and case-control comparisons between subjects (Bullmore, 2012). Recently, spontaneous fluctuations and temporal oscillations of BOLD signals during task or nontask status have been observed across the brain. These oscillations over distant but functionally related brain regions have certain patterns, defined as functional connectivity (FC) (Biswal et al., 1995), and may be classified as different “resting-state networks (RSNs)” (van den Heuvel et al., 2009). Many functional brain networks have been discovered, including executive control, frontoparietal, default mode, motor, medial visual and lateral visual, auditory, somatosensory, and cerebellum networks (Beckmann et al., 2005; Fox and Raichle, 2007; Rubinov and Sporns, 2010). These networks may be jeopardized under different pathological conditions (Fornito et al., 2015). Based on many previous works, there is now an attempt to classify neurologic symptoms and signs via the brain connectome (Fox, 2018).

FMRI is also used to explore the BOLD phenomenon in other organ systems, such as to detect ischemic areas in the heart (Manka et al., 2010) and to differentiate myometrium tissues in the uterus (Kido et al., 2007), because the interaction between blood flow and cellular activity is also considered present outside the brain. Due to high vascularity, muscle is also a suitable target to obtain BOLD imaging. BOLD contrast imaging was demonstrated to adequately reflect skeletal muscle perfusion changes after artificial ischemia and reactive hyperemia in the lower leg (Donahue et al., 1998), and the contrast is greater under high field strength (7 T) compared to low field strength (3 T) MRI, with improved signal-to-noise and contrast-to-noise ratio (Towse et al., 2016). Intravascular oxygen saturation and hemoglobin level seem to have a major contribution to the BOLD phenomenon (Meyer et al., 2004). BOLD can be used to explore the effect of skeletal muscle contraction (Towse et al., 2011) and intake of vasomodulative agents (Bulte et al., 2006); furthermore, BOLD is a potential useful tool to study peripheral pathologic conditions such as peripheral artery occlusion disease (PAOD). It was found that in PAOD patients, BOLD signal intensity was significantly lower compared to healthy controls after cuff deflation in lower limbs (Ledermann et al., 2006; Partovi et al., 2012). Moreover, the “task” FMRI concept is applied in other body areas (examples mentioned above), and the concept of RSNs oscillations is explored outside the brain. The spinal cord, an “extended” portion of the CNS outside the brain, has been a suitable research target. Early studies revealed that the BOLD signal responses in cervical spinal cord FMRI are detectable under different stimulation paradigms, such as finger tapping (Govers et al., 2006) and controlled hypercapnia (Cohen-Adad et al., 2010). With improved image quality of the spinal cord area, spatially distinct RSNs have been identified, mainly separated as dorsal and ventral networks mimicking spinal cord functional anatomy (Kong et al., 2014); additionally, bilateral FC communication of motor and sensory horns is noted (Barry et al., 2014).

Moreover, FC has a strong relationship, although not necessarily direct, with structural connectivity (SC) (Honey et al., 2009; van den Heuvel et al., 2009), and this relationship is supported by the striking loss of resting interhemispheric BOLD correlations after corpus callosotomy (Johnston et al., 2008). Similarly, in previous work on RSNs over the spinal cord (Barry et al., 2014; Kong et al., 2014; Wei et al., 2010), the observed spontaneous fluctuations over the spinal cord may be of neural origin, the same as those in the brain, since the spinal cord is part of the CNS and has constant ascending and descending neuronal activities via spinal tracts, which are also structurally related.

Based on results of the previous studies, we hypothesize that since both afferent and efferent fibers are present outside of the brain, the complex BOLD oscillations that have been observed in the brain should exist outside the brain and can be detected, e.g., in the trunk or in the limbs where nerves reside (Byrge et al., 2014; Tang et al., 2011). Like the RSNs in the brain, the spontaneous FMRI oscillations in and out of the brain are speculated to spread throughout the body possibly through nerve fibers, synapses, and later on e.g. neuromuscular junctions. These oscillations might also exist when the whole body is in a resting state, and would be a reflection of the baseline homeostatic status of the body. This also might be different from the well-known corticomuscular coherence that exists during active or pathological muscle contraction (Mima and Hallett, 1999; Timmermann et al., 2002). This speculation might be supported by certain studies (Bashan et al., 2012; Fultz et al., 2019; Kerkman et al., 2018; Klimesch, 2018; Tang et al., 2019). Bashan et al. introduced network physiology by recording electroencephalogram (EEG), electrocardiogram (ECG), respiration, electrooculogram (EOG), and electromyogram (EMG) of the chin and leg (Bashan et al., 2012). Another popular example is the interaction between the brain and gut. There are communications through neuronal, endocrine, and immune-related signals, and are noted to be disturbed in certain diseases (Mayer, 2011).

Nonetheless, differentiating neuronal and non-neuronal activities in MR BOLD studies is challenging and even more serious in FC and network analyses. Brain BOLD signals are coupled with various body signals—although mostly theoretically with the very origin the brain—including heartbeats, respirations, and head motion. Many strategies have been developed to counteract this problem. In seed-based or ROI-based approaches, voxels are chosen based on previous background knowledge. Then, FC and functional networks are calculated and built-up after a series of preprocessing steps, including frequency band filtering and motion correction (Fox and Raichle, 2007; Fox, 2010). Another approach, independent component analysis (ICA), a method of blind source signal separation, is also widely used, taking advantage of its great ability to decompose functional time series into various temporo-spatial components, usually with no need for meticulous preprocessing steps or prior assumptions regarding the selection of prespecified brain regions (Calhoun et al., 2001; Wang and Peterson, 2008). Therefore, ICA is also a powerful tool to explore the possible network organization under the new platform we are proposing.

In summary, it is plausible to speculate that during either a task or nontask status of the brain there are also interactions between the brain and other related body systems, including but not limited to the peripheral nervous system (PNS), and MRI BOLD signals could be used to detect them (please see Fig. 1 diagram). A new platform focusing on both the brain and another body part should be demonstrated to be feasible and stable for different subjects. Optimal parameters are required to obtain images with satisfactory quality. Furthermore at this preliminary stage certain temporal oscillations of functional time series between the brain and another body part are expected. We aimed to establish a platform by developing a new MR subject position and explore simultaneous BOLD signals of the brain and upper limb. We argue that BOLD signal oscillations can be detected in the body and exist in a form correlated to certain existing brain networks. More importantly, we establish a new platform that can be used for further research.

Section snippets

Subjects

A total of 11 right-handed healthy subjects were recruited (5 males/6 females; mean age 26.8 ± 4.8 years old) on the campus. All subjects were screened to exclude past history of neurologic disorder, psychiatric illness, and concomitant drug use. Informed consent was obtained from all subjects in line with the protocol approved by the National Taiwan University Institutional Review Board. Of note, each subject might participate in the experiment more than once (e.g., in the stage 1 and stage 2

Results

We planned to establish a new MR platform to study possible simultaneous BOLD oscillations between the brain and periphery. First, in experiment stage 1, the main objective was to test the tolerability of the new AOH MR position and MR image quality. Then in stage 2, the objective was to further test whether BOLD signals under certain tasks can be detected via this platform. All subjects completed the experiments and thus proved the tolerability of this new MR position. The results of the new

Discussion

To the best of our knowledge, we are the first to design the new AOH MR position, verify the image quality, and use this new simultaneous brain-periphery platform to illustrate possible functional connections between the brain and periphery in an MRI scanner (Tang et al., 2019). Our results support the following: (1) the new AOH MR position is feasible and tolerable for subjects in regular MR experiments; (2) MRI can be used to study the brain and periphery in the same plane with adequate SNR

Conclusion

In conclusion, mapping symptoms to brain networks with the human connectome has been used in various neurologic and psychiatric diseases (Fox, 2018), but pathophysiologically, the findings are not expected to be restricted to the brain. Considering the to and fro interactions between the central and periphery, it is reasonable to speculate that studying the brain alone is insufficient. Our new platform for simultaneous brain-body MRI may contribute to new knowledge of interactions between human

CRediT authorship contribution statement

I-Ning Tang: Conceptualization, Methodology, Formal analysis, Investigation, Writing - original draft, Writing - review & editing. Tun Jao: Conceptualization, Methodology, Software, Formal analysis, Investigation, Writing - original draft, Writing - review & editing, Visualization, Funding acquisition. Yun-An Huang: Methodology, Software, Validation, Formal analysis, Investigation. Chia-Wei Li: Methodology, Validation, Investigation. Ya-Chih Yu: Methodology, Validation, Investigation. Jyh-Horng

Acknowledgments

This research was funded by the Ministry of Science and Technology (NSC 99-2321-B-002-007 and MOST 103-2321-B-002-097). T.J. is supported by the Ministry of Science and Technology (MOST 107-2314-B-002-070) and National Taiwan University Hospital (NTUH 107-N4022). We would also like to thank Dr. Che-Wei Chang of National Taiwan University for providing the illustration of the AOH position in Fig. 2A.

References (63)

  • M.J. Rohrer et al.

    Axillary artery compression and thrombosis in throwing athletes

    J. Vasc. Surg.

    (1990)
  • M. Rubinov et al.

    Complex network measures of brain connectivity: uses and interpretations

    Neuro. Image

    (2010)
  • A. Weissenbacher et al.

    Correlations and anticorrelations in resting-state functional connectivity MRI: a quantitative comparison of preprocessing strategies

    Neuro. Image

    (2009)
  • S. Achard et al.

    A resilient, low-frequency, small-world human brain functional network with highly connected association cortical hubs

    J. Neurosci.

    (2006)
  • A. Alexander-Bloch et al.

    Disrupted modularity and local connectivity of brain functional networks in childhood-onset schizophrenia

    Front. Syst. Neurosci.

    (2010)
  • J. Ashburner et al.

    SPM12 manual. Functional imaging laboratory

    Wellcome Trust Centre for Neuroimaging

    (2018)
  • R.L. Barry et al.

    Resting state functional connectivity in the human spinal cord

    Elife

    (2014)
  • A. Bashan et al.

    Network physiology reveals relations between network topology and physiological function

    Nat. Commun.

    (2012)
  • D. Bassett et al.

    Small-world brain networks

    Neuroscientist

    (2006)
  • C. Beckmann et al.

    Investigations into resting-state connectivity using independent component analysis

    Philos. Trans. Biol. Sci.

    (2005)
  • B. Biswal et al.

    Functional connectivity in the motor cortex of resting human brain using echo-planar MRI

    Magn. Reson. Med.

    (1995)
  • E. Bullmore

    The future of functional MRI in clinical medicine

    Neuro. Image

    (2012)
  • D.P. Bulte et al.

    Vasomodulation of skeletal muscle BOLD signal

    J. Magn. Reson. Imaging

    (2006)
  • V. Calhoun et al.

    A method for making group inferences from functional MRI data using independent component analysis

    Hum. Brain Mapp.

    (2001)
  • V.D. Calhoun et al.

    ICA of functional MRI data: an overview

  • J. Damoiseaux et al.

    Consistent resting-state networks across healthy subjects

    Proc. Natl. Acad. Sci.

    (2006)
  • K.M. Donahue et al.

    Simultaneous gradient‐echo/spin‐echo EPI of graded ischemia in human skeletal muscle

    J. Magn. Reson. Imaging

    (1998)
  • A. Fornito et al.

    The connectomics of brain disorders

    Nat. Rev. Neurosci.

    (2015)
  • M.D. Fox

    Clinical applications of resting state functional connectivity

    Front. Syst. Neurosci.

    (2010)
  • M.D. Fox

    Mapping symptoms to brain networks with the human connectome

    N. Engl. J. Med.

    (2018)
  • M. Fox et al.

    Spontaneous fluctuations in brain activity observed with functional magnetic resonance imaging

    Nat. Rev. Neurosci.

    (2007)
  • Cited by (0)

    1

    Co-first authors.

    View full text