**Comparison of Cortical Activation During Real Walking and Mental Imagery of Walking – The Possibility of Quickening Walking Rehabilitation by Mental Imaginary of Walking**

Jiang Yinlai1, Shuoyu Wang1, Renpeng Tan1, Kenji Ishida2, Takeshi Ando3 and Masakatsu G. Fujie3 *1Kochi University of Technology 2Kochi University 3Waseda University Japan* 

#### **1. Introduction**

132 Infrared Spectroscopy – Life and Biomedical Sciences

West, J.B. (1992) *Pulmonary Pathophysiology – the essentials*, 4th Ed., Williams & Wilkins,

Windischberger C, et al. (2002), "On the origin of respiratory artifacts in BOLD-EPI of the

human brain". *Magnetic Resonance Imaging*, Vol. 20, pp. 575-582.

*Hemoencephalography*, Haworth Medical Press, pp 5-21.

Baltimore.

*Neurotherapy* 8(3), pp 5-21 and In: Tinius T. (Ed.) *New Developments in Blood Flow* 

Non-invasive brain imaging technologies have become an increasingly important part of research in neurosciences. The thirst for information about brain function is universal, and imaging of the human brain has been used by many as a medium for the discussion. So far, Functional brain imaging with positron emission tomography (PET), functional magnetic resonance imaging (fMRI), electroencephalographic (EEG), and Magnetoencephalography (MEG) have greatly increased scientists' ability to study localized brain activity in humans and carry out studies for better understanding of the neural basis of mental states. They have been used extensively to map regional changes in brain activity, not only in neuroscience researches, as well as in social sciences to objectively and quantitatively evaluate psychological problems. PET and fMRI are based on changes in local circulation and metabolism (Raichle & Mintun, 2006). PET produces detailed three-dimensional images of certain processes in the brain by detecting gamma rays emitted indirectly by radioactive material which has been injected into the person's blood stream prior to scanning. fMRI produces high quality pictures of the brain's delicate soft tissue structures using strong magnets and pulses of radio waves to manipulate the natural magnetic properties of hydrogen, creating useful images of organs and soft tissues. MEG and EEG image electrical activity in the brain. MEG measures magnetic fields generated by small electrical currents in neurons of the brain using arrays of SQUIDs (superconducting quantum interference devices). EEG uses multiple electrodes fixed to the person's scalp to measure the dynamic pattern of electrical fields in the brain. In cognitive neuroscience, researchers use EEG technology to study event-related potentials (ERPs)—brain measurements that are associated with a response to a stimulus.

These methods provide information about changes in electrical, hemodynamic and metabolic activities. Each of these techniques has its advantages and disadvantages, but

Comparison of Cortical Activation During Real Walking and Mental Imagery of

(Plichta et al., 2006; Sato et al., 2005; Sato et al., 2006).

disabilities (Ishida et al., 2008).

Fig. 2. Omnidirectional walkers.

Walking – The Possibility of Quickening Walking Rehabilitation by Mental Imaginary of Walking 135

fNIRS has several unique advantages over current measurement methods. It is noninvasive, and can be used under a variety of conditions with minimal restriction on the examinee. Measurements can be made under more natural conditions, giving more freedom in task design. It also enables simultaneous measurements with other testing modalities such as EEG, fMRI and MEG because near infrared light is not interfered by EEG, fMRI and MEG, and does not interfere them. fNIRS facilitates longitudinal studies and monitoring over extended time periods. Therefore, fNIRS technology allows the design of portable, safe, affordable and accessible monitoring systems. These qualities pose fNIRS as an ideal candidate for monitoring cognitive activity-related hemodynamic changes not only in laboratory settings but also under natural conditions. The reliability of fNIRS signals has in most cases been proven to be sufficient at a group level for observation of brain activity

The authors and their colleagues have been developing machines for walking rehabilitation and walking support. In an aging society with a low birth rate in countries like Japan, people suffering from walking impairments due to illness or accident are increasing and the number of physical therapist cannot meet with the demand for walking rehabilitation. Therefore, rehabilitation machines, which can help with early recovery and relieve burden of physical therapists, have drawn great attentions (Okada et al., 2001; Horst, 2009). In previous studies, we have developed omnidirectional walkers for standing exercise (Tan et al., 2011) and seated exercise, shown in Fig. 2(a) and Fig. 2(b) respectively. The walker for standing exercise is designed for those able to keep standing posture by themselves, and the walker for sitting exercise is designed for the severe patients unable to stand. Omnidirectional walking exercise has been proved effective for early recovery of walking

(a) For standing exercise (b) For seated exercise

The causes for walking disabilities include not only muscle weakness but also neural dysfunctions due to stroke or Alzheimer's disease. 58% of walking disabilities are caused by problems in the neural system. However, up to now, most of the developed walking rehabilitation machines aim at enhancing muscle strength, neglecting the recovery of the neural system. Thus it is necessary to consider the brain activities besides muscle strength in walking rehabilitation, in order to improve the efficiency of walking rehabilitation. Furthermore, for the severe patients who are completely bed-ridden, it is important to activating their neural system related to walking movement. A hybrid walking

helps to elucidate certain aspects of the capacity of neural networks to process information. MEG and EEG provide unique insights into the dynamic behaviour of the human brain as they are able to follow changes in neural activity on a millisecond timescale. In comparison, PET and fMRI are limited in temporal resolution to time scales on the order of one second by physiological and signal-to-noise considerations. On the other hand, MEG, PET and fMRI provide high resolution brain images. The resolution of fMRI is about 2-3 mm at present, limited by the spatial spread of the hemodynamic response to neural activity. However, MEG, PET and fMRI techniques are very expensive, highly sensitive to motion artifact, confine participants to restricted positions, which severely limits their application in daily use outside hospitals and research centers. Although EEG is much cheaper, it is highly sensitive to artifacts whose amplitude can be quite large relative to the size of amplitude of the cortical signals of interest. The artifacts include both biological artifacts such as eye-induced artifacts, cardiac artifacts and muscle activation induced artifacts, and environmental artifacts such as body movement, settling of the electrodes, and electrical appliances. Therefore, EEG, MEG, PET and fMRI are difficult to measure brain activity in natural environment. A brain imaging technology, which is non-invasive, less constrictive, low-cost, with a relatively higher temporal and spatial resolution, is desirable.

Functional near-infrared spectroscopy (fNIRS) is an emerging brain imaging technology monitoring concentration changes of oxygenated hemoglobin (oxy-Hb) and deoxygenated hemoglobin (deoxy-Hb) at the cortex by measuring the absorption of near infrared light between 650nm and 950nm through the intact skull (Chance et al., 1993; Villringer et al, 1993). Specifically, the transmission and absorption spectra of oxy-Hb and deoxy-Hb are distinct in this wavelength region. The fundamentals of the optical topography system utilize the phenomenon, using the better penetrating near infrared light, rather than visible light, to measure changes in blood hemoglobin concentrations in the brain. As shown in Fig.1, a laser optode is illuminated onto head from optical fibers attached to the scalp. The near infrared light passes through the skull and reaches the cerebral cortex. It is scattered by hemoglobin in the blood. The light is partially reflected back through the scalp. The reflected light back on the scalp contains the information about the cerebral cortex. When a specific area of the brain is activated, the localized blood volume in that area changes quickly. It can thus be detected, where and how active the specific regions of the brain are, by continuously monitoring the blood hemoglobin levels according to the absorption level of near infrared light, while having the examinee do some specific action or task paradigm.

Fig. 1. Cortical activation measurement with near infrared light.

fNIRS has several unique advantages over current measurement methods. It is noninvasive, and can be used under a variety of conditions with minimal restriction on the examinee. Measurements can be made under more natural conditions, giving more freedom in task design. It also enables simultaneous measurements with other testing modalities such as EEG, fMRI and MEG because near infrared light is not interfered by EEG, fMRI and MEG, and does not interfere them. fNIRS facilitates longitudinal studies and monitoring over extended time periods. Therefore, fNIRS technology allows the design of portable, safe, affordable and accessible monitoring systems. These qualities pose fNIRS as an ideal candidate for monitoring cognitive activity-related hemodynamic changes not only in laboratory settings but also under natural conditions. The reliability of fNIRS signals has in most cases been proven to be sufficient at a group level for observation of brain activity (Plichta et al., 2006; Sato et al., 2005; Sato et al., 2006).

The authors and their colleagues have been developing machines for walking rehabilitation and walking support. In an aging society with a low birth rate in countries like Japan, people suffering from walking impairments due to illness or accident are increasing and the number of physical therapist cannot meet with the demand for walking rehabilitation. Therefore, rehabilitation machines, which can help with early recovery and relieve burden of physical therapists, have drawn great attentions (Okada et al., 2001; Horst, 2009). In previous studies, we have developed omnidirectional walkers for standing exercise (Tan et al., 2011) and seated exercise, shown in Fig. 2(a) and Fig. 2(b) respectively. The walker for standing exercise is designed for those able to keep standing posture by themselves, and the walker for sitting exercise is designed for the severe patients unable to stand. Omnidirectional walking exercise has been proved effective for early recovery of walking disabilities (Ishida et al., 2008).

(a) For standing exercise (b) For seated exercise

Fig. 2. Omnidirectional walkers.

134 Infrared Spectroscopy – Life and Biomedical Sciences

helps to elucidate certain aspects of the capacity of neural networks to process information. MEG and EEG provide unique insights into the dynamic behaviour of the human brain as they are able to follow changes in neural activity on a millisecond timescale. In comparison, PET and fMRI are limited in temporal resolution to time scales on the order of one second by physiological and signal-to-noise considerations. On the other hand, MEG, PET and fMRI provide high resolution brain images. The resolution of fMRI is about 2-3 mm at present, limited by the spatial spread of the hemodynamic response to neural activity. However, MEG, PET and fMRI techniques are very expensive, highly sensitive to motion artifact, confine participants to restricted positions, which severely limits their application in daily use outside hospitals and research centers. Although EEG is much cheaper, it is highly sensitive to artifacts whose amplitude can be quite large relative to the size of amplitude of the cortical signals of interest. The artifacts include both biological artifacts such as eye-induced artifacts, cardiac artifacts and muscle activation induced artifacts, and environmental artifacts such as body movement, settling of the electrodes, and electrical appliances. Therefore, EEG, MEG, PET and fMRI are difficult to measure brain activity in natural environment. A brain imaging technology, which is non-invasive, less constrictive, low-cost, with a relatively higher temporal and

Functional near-infrared spectroscopy (fNIRS) is an emerging brain imaging technology monitoring concentration changes of oxygenated hemoglobin (oxy-Hb) and deoxygenated hemoglobin (deoxy-Hb) at the cortex by measuring the absorption of near infrared light between 650nm and 950nm through the intact skull (Chance et al., 1993; Villringer et al, 1993). Specifically, the transmission and absorption spectra of oxy-Hb and deoxy-Hb are distinct in this wavelength region. The fundamentals of the optical topography system utilize the phenomenon, using the better penetrating near infrared light, rather than visible light, to measure changes in blood hemoglobin concentrations in the brain. As shown in Fig.1, a laser optode is illuminated onto head from optical fibers attached to the scalp. The near infrared light passes through the skull and reaches the cerebral cortex. It is scattered by hemoglobin in the blood. The light is partially reflected back through the scalp. The reflected light back on the scalp contains the information about the cerebral cortex. When a specific area of the brain is activated, the localized blood volume in that area changes quickly. It can thus be detected, where and how active the specific regions of the brain are, by continuously monitoring the blood hemoglobin levels according to the absorption level of near infrared light, while having the examinee do some specific action

Emitter Detector

Fig. 1. Cortical activation measurement with near infrared light.

Skull

Cortex

spatial resolution, is desirable.

or task paradigm.

The causes for walking disabilities include not only muscle weakness but also neural dysfunctions due to stroke or Alzheimer's disease. 58% of walking disabilities are caused by problems in the neural system. However, up to now, most of the developed walking rehabilitation machines aim at enhancing muscle strength, neglecting the recovery of the neural system. Thus it is necessary to consider the brain activities besides muscle strength in walking rehabilitation, in order to improve the efficiency of walking rehabilitation. Furthermore, for the severe patients who are completely bed-ridden, it is important to activating their neural system related to walking movement. A hybrid walking

Comparison of Cortical Activation During Real Walking and Mental Imagery of

possible to quicken walking rehabilitation by mental imagery of walking.

was suggested.

**2. fNIRS measurement** 

Fig. 4. ETG-7100 system and its shell to hold 4×4 optodes.

Walking – The Possibility of Quickening Walking Rehabilitation by Mental Imaginary of Walking 137

In the first experiment, we compared the activation in motor area of the brain during real walking (RW) and walking observation (WO) (Jiang et al., 2010). Two subjects participated in the experiment. During WO, the subjects were instructed to imagine that they were walking with the same pace to a person in the video being shown to the subjects. As a result, the concentration of oxygenated hemoglobin in motor area during WO was higher than that during RW in both of the subjects. This was because that it was not necessary to pay attention to the movements of the legs and feet during normal walking on the plain road without any obstacles, while movement planning was required when the subjects imagined their walking in the same way to another person. The experiment result indicated that it is

In the second experiment, we compared the activation in motor area during RW, virtual walking (VW), and WO. Subjects stood on a treadmill throughout the experiment (Jiang et al., 2011). In the VW, subjects were shown moving scenes of a virtual visual environment in which subjects easily imagined as if they were actually walking from the first-person perspective. In the WO, subjects were instructed to imagine that they were walking with the same pace to a person in the video being shown to the subjects (third-person perspective). Four subjects participated in the experiment. As a result, the oxy-Hb in motor area during both VW and WO were higher than that during RW on the average. This was because that it was not necessary to pay attention to the movements of the legs and feet during normal walking, while movement planning was required when the subjects imagined their walking according to the videos. There was no significant difference between the oxy-Hb during VW and that during WO. The importance of stimulus diversity in mental imagery of walking

Regional hemodynamic changes in brain tissue were monitored using fNIRS system ETG-7100 (Hitachi Medical Corporation) (Fig. 4). This system uses two wavelengths of nearinfrared light (695 nm and 830 nm) to separate the two types of hemoglobin concentration

rehabilitation system is proposed which includes both muscle strength enhancement by walking rehabilitation machines and neurological rehabilitation by imaginary of walking. This system has the following most prominent advantages compared with traditional rehabilitation methods considering only physical rehabilitation.


However, the neural mechanism of neurological rehabilitation is yet to be elucidated and there is no standard method to carry out neurological rehabilitation.

Recently, motor imagery, as a method of neurological rehabilitation, is drawing more and more attention.Motor imagery is widely used in sport to improve performance, which raises the possibility of applying it as a rehabilitation method. The effectiveness of motor imagery training at restoring motor function after stroke has been indicated by several studies (Sharma et al., 2006; Dickstein et al., 2004). However, the underlying mechanism of motor imagery training-induced improved performance remains unexplained. Understanding the effect of rehabilitative techniques on brain plasticity is potentially important in providing a neural substrate to underpin rehabilitation and hence in developing novel rehabilitation strategies. An fMRI study has shown that premotor cortex (PM) and supplementary motor area (SMA), as shown in Fig. 3, are involved in the observation of gait and related conditions in combination with motor imagery of gait (Iseki et al., 2008; Wagner et al., 2008). However, since the MRI environment excluded real gait movement, the comparison between brain activities involved in walking and imaginary walking was still insufficient. In this chapter, we compared the activation in motor area of the brain during real walking and imaginary walking by means of fNIRS. Two experiments were conducted.

Fig. 3. Activated brain regions by mental imagery of walking (Iseki et al., 2008;).

In the first experiment, we compared the activation in motor area of the brain during real walking (RW) and walking observation (WO) (Jiang et al., 2010). Two subjects participated in the experiment. During WO, the subjects were instructed to imagine that they were walking with the same pace to a person in the video being shown to the subjects. As a result, the concentration of oxygenated hemoglobin in motor area during WO was higher than that during RW in both of the subjects. This was because that it was not necessary to pay attention to the movements of the legs and feet during normal walking on the plain road without any obstacles, while movement planning was required when the subjects imagined their walking in the same way to another person. The experiment result indicated that it is possible to quicken walking rehabilitation by mental imagery of walking.

In the second experiment, we compared the activation in motor area during RW, virtual walking (VW), and WO. Subjects stood on a treadmill throughout the experiment (Jiang et al., 2011). In the VW, subjects were shown moving scenes of a virtual visual environment in which subjects easily imagined as if they were actually walking from the first-person perspective. In the WO, subjects were instructed to imagine that they were walking with the same pace to a person in the video being shown to the subjects (third-person perspective). Four subjects participated in the experiment. As a result, the oxy-Hb in motor area during both VW and WO were higher than that during RW on the average. This was because that it was not necessary to pay attention to the movements of the legs and feet during normal walking, while movement planning was required when the subjects imagined their walking according to the videos. There was no significant difference between the oxy-Hb during VW and that during WO. The importance of stimulus diversity in mental imagery of walking was suggested.
