Professional Documents
Culture Documents
A Systemic Review of Functional Near-Infrared Spectroscopy For Stroke: Current Application and Future Directions
A Systemic Review of Functional Near-Infrared Spectroscopy For Stroke: Current Application and Future Directions
*Correspondence:
Conclusions: This study provides a preliminary evidence of the application of NIRS in
Pu Wang stroke patients as a monitoring, therapeutic, and research tool. Further studies could
[email protected]
give more emphasize on the combination of NIRS with other techniques and its utility in
the prevention of stroke.
Specialty section:
This article was submitted to Keywords: near-infrared spectroscopy, stroke, motor recovery, cortical function, systemic review
Stroke,
a section of the journal
Frontiers in Neurology
INTRODUCTION
Received: 04 September 2018
Accepted: 16 January 2019 Stroke, which refers to a medical condition in which insufficient brain blood supply results in
Published: 05 February 2019 cell death, is a major cause of death and disability worldwide (1, 2). Survivors are accompanied
Citation: with the deterioration or loss of functions, for example, sensorimotor sequelae including motor
Yang M, Yang Z, Yuan T, Feng W and weakness and impairment of voluntary motor control, spasticity, incoordination, apraxia, sensory
Wang P (2019) A Systemic Review of
loss/numbness, dysphagia, and dysarthria, and stroke could also lead to various cognitive and
Functional Near-Infrared
Spectroscopy for Stroke: Current
psychiatric deficits (3–5). These disfunctions are associated with cortical impairment due to
Application and Future Directions. insufficient blood supply and brain oxygenation. Therefore, monitoring the changes of brain
Front. Neurol. 10:58. circulation and oxygenation could timely reflect rehabilitation and recovery and the effect of
doi: 10.3389/fneur.2019.00058 therapy.
Neuroimaging has been shown to be an effective monitoring blood supply could be observed. The reduction in cerebral
and therapeutic tool, evaluating the evolution of neural oxygenation results in injury of neurovascular unit, decreased
activity and stroke rehabilitation and recovery (6). Traditional neuronal activity and accumulation of anaerobic metabolites.
methods such as functional magnetic resonance imaging NIRS could monitor oxygenation signal changes, thus reflecting
(fMRI), positron emission tomography, electroencephalography this pathophysiological process (1).
(EEG) and magnetoencephalography (MEG)—have provided NIRS is well-established as a safe and effective monitoring
considerable initial insight into brain changes during recovery. tool for stroke recovery, including upper limb, lower limb
However, several shortcomings including a confining monitoring recovery, motor learning, cortical function recovery (8), cerebral
environment, subject head fixation and high cost have limited hemodynamic changes, cerebral oxygenation (14), therapy (15,
applications in tasks which require constant movement or real- 16), clinical researches and evaluation of the risk for stroke (17).
time monitor (7, 8). Brain-computer interfaces (BCIs), a newly emerging tool, use
Near-infrared spectroscopy (NIRS), introduced in 1977 by brain activity to control external devices, facilitating paralyzed
Jöbsis et al. as a monitoring tool of cerebral and myocardial patients to interact with the environment. NIRS combined with
oxygenation (9), has partially overcome these difficulties. NIRS BCI offers great potential as a therapeutic tool (18, 19). In
is a non-invasive neuroimaging tool that has several potential addition, NIRS has used in several clinical studies (20, 21),
advantages including real-time monitor, low price, simplicity, reflecting hemodynamic or oxygenation changes of brain, as
portability, relatively small equipment, and it’s almost completely well as in evaluation of the risks for postoperative stroke and
safe and non-invasive nature (8). muscle metabolism (22). NIRS has shown to be an effective
NIRS can be divided into continuous wave NIRS (CW NIRS), and promising method, however, its clinical value still remains
time domain NIRS (TD NIRS), and frequency domain NIRS controversial (23). Therefore, to address these discrepancies,
(FD NIRS). CW NIRS emits continuous wave and measures the we conducted this systematic review to summarize the existing
changes in the intensity of the light that passed through the tissue, application of NIRS in stroke patients.
whereas TD NIRS utilizes a short pulse of laser light and measures
the arrival times of photons emerging from the tissue. FD NIRS
records the intensity of the detected light as well as the phase METHODS
shift. These signals can then be converted to the concentration
To evaluate how NIRS has been employed for surveillance and
of oxygenated (oxy-Hb) and deoxygenated hemoglobin (deoxy-
treatment of stroke patients, we conducted a systematic literature
Hb). One of the most common used algorithm is Modified
review of all published original research involving NIRS in stroke
Beer–Lambert Law (MBLL). CW NIRS could not measure
patients. The literature search, conducted in July 2018, began
absolute concentrations of oxy-Hb and deoxy-Hb, because this
with an initial search of the NCBI Pubmed database using the
method assumes a homogenous tissue which is not true. This
following search terms: “(‘Spectroscopy, Near-Infrared’ [MeSH
does not change the results of qualitative analysis, but may
Major Topic] OR “near-infrared spectroscopy”[Title/Abstract]
lead to error in quantitative outcome. TD NIRS recording the
OR fNIRS[Title/Abstract]) AND (stroke[MeSH Major Topic]
temporal broadening of the pulse as it penetrates the investigated
OR “cerebrovascular disorders”[MeSH Major Topic] OR “brain
area allows accurate quantification of the concentrations and
infarction”[MeSH Major Topic])”. Additional spot check
has better spatial resolution (10). Neural activity increases
searches stemming from key references in identified works were
oxygen demands, thus increasing cerebral blood flow due to
also carried out to further strengthen the literature review’s
neurovascular coupling. NIRS could capture the changes of
reliability. Only English language articles published in peer-
oxy-Hb and deoxy-Hb to infer changes in brain activity (6,
reviewed journals were included. Then we searched EMBASE
11). Recent years have witnessed rapid development of the
database, Cochrane Library and ScienceDirect database
techniques of NIRS from the single-location measurements to
(Figure 1). All studies involving NIRS in stroke patients of any
two dimensions and then three dimensions. One of its most
age were included. The details of included studies are presented
promising application is in brain-computer interface (BCI)
in Tables 1–7. In total, our literature search identified 66 unique
which was firstly introduced by Coyle et al. (12). BCI use brain
papers relating to stroke recovery, including motor recovery,
activity to control external devices bypassing the peripheral
motor learning, cortical function, hemodynamic changes,
nervous system. NIRS presents as a valuable tool in its brain
cerebral blood oxygen, Near-Infrared Spectroscopy-Based
signal acquisition for its non-invasive nature and real-time
Brain–Computer Interface, and other applications in stroke
monitoring. However, fNIRS-BCI system is still mainly used
patients, which were included for further analysis (Figure 2).
in researches due to slow information transfer rate and low
accuracy (13).
Stroke includes two major types: ischemic, due to lack NIRS FOR MOTOR RECOVERY
of cerebral blood flow, and hemorrhagic, due to bleeding.
During ischemic stroke cerebral blood supply is disrupted by We identified 14 papers (24–37) reporting on NIRS for motor
narrowing of vessels caused by atherosclerosis or embolism. recovery in a total of 399 patients, including upper limb function
In hemorrhagic stroke caused by hypertension or ruptured recovery, lower limb function recovery, balance control and
aneurysm blood flow is reduced due to direct blood loss or motor learning (Table 1). In most patients, their motor function
vessel compression. In either case, a significant decline in was significantly injured due to stroke, resulting in hemiparesis
Study # of Patients character Etiology Target area Measurement Method Clinical applications
patients
(Continued)
TABLE 1 | Continued
Study # of Patients character Etiology Target area Measurement Method Clinical applications
patients
Fujimoto 20 17 men and 3 Subcortical stroke Frontoparietal region 1[Oxy-Hb] and Single-channel Evaluate cortical activation
et al. (36) women; mean (± 1[Deoxy-Hb]. NIRS associated with external
SD) age 60.2 (± 9.5) postural perturbation
years
MOTOR LEARNING
Hatakenaka 12 10 men and 2 Ataxia resulting from Forehead, including 1[Oxy-Hb] and 42-channel NIRS Investigate impact of the
et al. (37) women; mean (± infratentorial stroke SMC, SMA, pre-SMA, 1[Deoxy-Hb]. capacity for motor learning
SD) age 56 (± 16) dorsal PMC, and on motor recovery
years dorsolateral PFC
MCA, middle cerebral artery; SMC, sensorimotor cortex; PMC, premotor cortex; SMA, supplementary motor area; PFC, prefrontal cortex.
Study # of Patients character Etiology Target area Measurement Method Clinical applications
patients
Park et al. 1 Male, 73 Subcortical stroke Bilateral motor area 1[Oxy-Hb] and fNIRS Assess serial changes
(38) 1[Deoxy-Hb]. during CIT
Schytz 108 (four Ischemic stroke MCA 1[Oxy-Hb] and Multi-channel Assess CA by recording
et al. (39) studies) 1[Deoxy-Hb]. NIRS LFO
Han et al. 10 Age: 76.6 ± 8.5 year Cerebral infarction Left and right prefrontal HbO2 signals fNIRS Assess the prefrontal
(40) lobes functional connectivity using
wavelet coherence analysis
of cerebral tissue
oxyhaemoglobin
concentration (Delta [HbO2])
signals in elderly subjects
with cerebral infarction (CI)
during the resting state
Cao et al. 6 3 male, 3 female; Stroke with Sensorimotor cortex, Oxy-Hb, deoxy-Hb, fNIRS Evaluate cortical plasticity
(41) age 10.2 ± 2.1 hemiplegia covering PMC, SMA and total-Hb levels after CIT
and M1/S1
Tan et al. 12 6 men, 6 woman Cerebral infarction Bilateral forehead 1[Oxy-Hb] and Multi-channel Evaluate functional
(42) 1[Deoxy-Hb]. NIRS connectivity
Oldag 20 12 men, 8 women; Unilateral high-grade Bilateral MCA 1[Oxy-Hb] and Multi-channel Assess autoregulation delay
et al. (43) mean age 55.3 ± steno-occlusion of 1[Deoxy-Hb]. continuous wave
12.5 years MCA NIRS
Moriya 11 Seven males, four Stroke Prefrontal cortex 1[Oxy-Hb] and Two-channel NIRS Examine the acute effect of
et al. (44) females, aged 69.6 1[Deoxy-Hb]. physical exercise
± 12.0 years
Saita et al. 2 58-year-old male; Cerebellar stroke Bilateral frontal lobe 1[Oxy-Hb] and fNIRS Evaluate cortical activity
(45) 74-year-old female 1[Deoxy-Hb].
Mori et al. 14 12 men and 2 Stroke Prefrontal cortex 1[Oxy-Hb] and 16-channel NIRS Investigate the association
(46) women; mean age, 1[Deoxy-Hb]. between PFC activity and
61.1 ± 9.3 years; dual-task interference on
range, 36–72 years physical and cognitive
performance
White et al. 2 Infant Perinatal stroke Bilateraloccipital Lobe Oxygenation NIRS oximetry Investigate the functional
(47) organization of the brain in
adults and infants
supplementary motor, and parietal cortical areas in both oxygenation signals in the SMA of the unaffected hemisphere was
hemispheres, involved in balance control in poststroke patients significantly correlated with the gain in balance function (36).
(35). Among which, SMA was then demonstrated to be a Motor learning is essential in function recovery, including
crucial area—the increment of the postural-perturbation-related motor adaptation, and motor sequence learning. Motor
Study # of Patients character Etiology Target area Measurement Method Clinical applications
patients
adaptation refers to the ability to compensate for environmental Two studies have reported significant improvement in
changes such as external force. Motor sequence learning is task performance or cortical plasticity after constraint-induced
defined as learning of sequential motor actions, which is essential movement therapy (38, 41). In addition, fNIRS could assess
for activities of daily living. Impaired motor sequence learning the prefrontal functional connectivity using wavelet coherence
may be correlated with reduced rehabilitation gains for ataxic analysis of cerebral tissue oxy-Hb concentration (40). A
patients with stroke (37). frequency-specific disruption in resting-state connectivity has
been observed in cerebral infarction cases. fNIRS could also
used in neonates as a bedside monitoring tool to early detect
NIRS FOR CORTICAL FUNCTION neurological deficiency and provide prognostic information (47).
RECOVERY These evidence supported NIRS as a promising method for
evaluating functional recovery (42).
The cortical functions including cortical plasticity, functional Post-stroke patients exhibit changes in cerebral autoregulation
connectivity or cerebral autoregulation that could be reflected (CA) which could be assessed by recording spontaneous
based on oxygenation levels are directly related to motor and oscillations in mean arterial blood pressure and cerebral blood
cognitive recovery after stroke. We reviewed 10 papers (38–47) flow. The etiology of oscillations remained unknown, however,
reporting cortical function recovery (Table 2). its significance in reflecting CA has been demonstrated. NIRS
Study # of Patients character Etiology Target area Measurement Method Clinical applications
patients
Nemoto 2 85-yr-old female; Infarction in the Frontal-temporal or Oxygen saturation NIRS oximetry Monitor CBO changes
et al. (62) 62-yr-old female territory of MCA frontal-parietal cortices
Damian 24 15 male, 9 female; Complete MCA Bilateral frontal lobe Oxygen saturation NIRS oximetry Monitor CBO changes
et al. (63) 43-77 stroke and brain
swelling
Sakatani / / / / Oxygen saturation NIRS oximetry Clarify CBO changes
et al. (64) occurring in stroke and
brain tumors
Aries et al. 9 3 male, 6 female; Acute stroke Bilateral forehead Oxygen saturation NIRS oximetry Monitor CBO changes
(65) age 71 ± 10
Mundiyanapurath44 59–82 Acute ischemic Oxygen saturation NIRS oximetry Compare therapeutic effect
et al. (66) stroke
Liebig et al. / / / / Cerebral oxygenation NIRS oximetry Compare therapeutic effect
(67)
Hametner 43 22 male, 21 female; Acute ischemic Forehead Cerebral oxygenation NIRS oximetry Measure rSO2
et al. (68) 65–79 stroke
Hashimoto 1 77-year-old man CI Bilateral frontal lobe Regional cerebral NIRS oximetry Monitor CBO changes
et al. (69) oxygen saturation
(rSO2)
Ritzenthaler 3 56-year-old man; MCA occlusion Bilateral forehead Regional cerebral NIRS oximetry Assess rSO2 changes
et al. (70) 52-year-old man; oxygen saturation
40-year-old (rSO3)
pregnant woman
Nakamura 5 Four male, one Stroke Injured area 1[Oxy-Hb] and NIRS Evaluate effect of
et al. (71) female, 65.6 ± 10.6 1[Deoxy-Hb]. revascularization on the
years abnormal evoked CBO
responses
Vernieri 2 58-year-old man; Ischemic stroke MCA Cerebral oxygenation NIRS oximetry Evaluate the effects of
et al. (72) 61-year-old man hypercapnia
Moreau 5 2 male,3 female; Acute stroke MCA Cerebral oxygenation NIRS oximetry Investigate brain
et al. (73) median age 64 oxygenation in the first few
years old hours of stroke onset
Blokland et al. (15) 10 Stroke with tetraplegia Motor cortex Test the feasibility of a combined
EEG-fNIRS system
Mihara et al. (16) 20 Subcortical stroke Frontoparietal region Investigate the effect of neurofeedback on
motor imagery
with a high time resolution is suggested as a promising tool for therapeutic effect of moderate-intensity aerobic exercise by
measuring oscillations, identifying risk and evaluating treatment improving working memory performance (44).
strategies in patients with carotid artery disease and ischemic
stroke (39, 43).
Mori et al. investigated dual-task interference on physical NIRS FOR HEMODYNAMIC CHANGES
and cognitive performance in stroke patients. In consistent
with previous studies, during dual-task walking, PFC activation Fourteen papers (48–61) focusing on the applications of NIRS
might prioritize physical demands in stroke patients, but might in monitoring hemodynamic changes in 158 stroke patients
prioritize cognitive demands in healthy subjects (46). Another (Table 3). The probe usually positioned on the injured area of the
study examined the acute effect of physical exercise on prefrontal brain, NIRS can detect cerebral perfusion timely in patients with
cortex activity in post-stroke patients and demonstrated the acute infarction (49, 50) or steno-occlusive disease (56) at bedside
and demand and assess functional state of the brain. In infarcted Doppler recording subcortical changes and NIRS—cortical
non-metabolizing brain, oxygen saturation may be near normal arterioles and capillary modifications (72).
due to sequestered cerebral venous blood in capillaries and
decreased consumption. In regionally or globally ischemic, but
metabolizing brain, saturation decreases because oxygen supply NIRS AS A THERAPEUTIC TOOL
is insufficient to meet metabolic demand (62). Several studies
demonstrated good responsiveness of NIRS signal to oxygenation Despite several efficient therapies, neurorehabilitation is an
changes in stroke and brain tumors (63–65). Unlike fMRI which intricate process requiring time and numerous factors. Given this
provides information mainly about concentration changes of circumstance, BCI which use brain activity to control external
deoxy-Hb that is paramagnetic,. NIRS detect both oxy-Hb and devices may be the only solution for patients with severe function
deoxy-Hb changes, thus providing more information (64). In deficiency, failing to interact with the environment and sustain
addition, bilateral NIRS seems to be more useful on cerebral normal daily lives.
oxygenation than unilateral measurements and may help in BCI can be divided into two types, assistive BCI and
prediction of the worsening of brain swelling (63). rehabilitative BCI system. The assistive BCI system aims
One of the most important predictors of clinical outcome after to substitute lost functions, for example, to achieve high-
acute ischemic stroke is recanalization. Endovascular treatment dimensional movement control of robotic devices. In contrast,
is one of major therapies of stroke patients. NIRS may be useful the rehabilitative BCI system aims to facilitate the restoration of
to monitor oxygenation during management of acute ischemic brain function by normalizing the neurophysiologic activity (84).
stroke (70). Nakamura et al. demonstrated that revascularization The patient’s brain signals obtained via invasive or non-invasive
could improve the abnormal evoked-CBO response in ischemic means are acquired via amplifiers, filtered, and decoded using
stroke patients (71). Studies were designed to optimize clinical an online classification algorithm to control external devices.
treatment with NIRS as a monitoring tool in measuring cerebral Distinct techniques have been explored to control a BCI, such as
oximetry. However, whether patients suffering from acute EEG and more recently NIRS.
ischemic stroke and undergoing endovascular recanalization NIRS is a relatively new measurement modality that offers a
should be treated under general anesthesia (GA) or conscious portable, low-cost, sensitive option for BCI development. The
sedation (CS) remained a matter of debate. A prospective study idea of using NIRS as an optical BCI has been introduced by
with forty-four patients investigated their effects. The result Coyle in 2004 (85). Since then a number of studies (84–90)
showed CS groups required less vasopressor medication and had further investigated its application in BCI, either examining the
a higher mean blood pressure (66). Mechanical thrombectomy resulting signals for motor imagery or classifying the NIRS signals
might have higher recanalization rates than intravenous or intra- directly. Despite significant advantages because of non-invasive,
arterial thrombolysis, resulting in a better clinical outcome (67). portable nature of NIRS, there are several seconds of delay
Treatment of endovascular stroke often impedes neurologic between neural activation and NIRS signal changes. To overcome
assessment in patients who need sedation or general anesthesia. these disadvantages of NIRS-based BCI systems, Siamac et al.
NIRS may fulfill this gap by measuring regional oxygen combined EEG and NIRS measurements in BCI (91). Recently
saturation (rSO2). Bi-channel rSO2-NIRS has potential for this hybrid EEG–fNIRS brain computer interface has attracted
guiding neuroanesthesia and predicting outcome (68). extensive attention and shown enhanced accuracy (92–97). Most
Cerebral hyperperfusion syndrome (CHS) was described of these studies, however, are limited in healthy population.
in a case report of a 77-year-old man with acute cerebral Blokland et al. demonstrated feasibility and accuracy of using a
infarction, which is known to be a rare but devastating combined EEG-fNIRS sensorimotor rhythms-based brain switch
complication of carotid artery revascularization. The patient in patients with tetraplegia. A combined EEG-fNIRS system
underwent successful recanalization with thrombectomy by might be especially beneficial for patients whose classification
using a stent retriever for the middle cerebral artery and performance based on EEG only are low (15). The recent study
stent placement for the origin of the internal carotid artery. of Khan et al. Has shown the possibility of this hybrid EEG-
However, 12 h after treatment acute intracranial hemorrhage fNIRS system in controlling a quadcopter with fNIRS decoding
happened, which was considered to have been caused by mental arithmetic, counting, rotation, and word formation, EEG
CHS under the evaluation of NIRS. This case emphasized decoding eyeblinks and movement (92). The combination of EEG
the significance of performing routine monitoring of regional and fNIRS increases the number of control commands, improve
cerebral oxygen saturation by NIRS to timely recognize CHS classification accuracy and reduce the signal detection time (93).
(69). Further improvement in its efficacy requires proper brain region
NIRS may be more useful in combination with other identification and feature selection (94).
techniques. Vernieri et al. combined NIRS with transcranial In addition, NIRS was used in neurofeedback. Neurofeedback
Doppler to investigate the effects of hypercapnia on healthy is a type of biofeedback that measures brain waves to produce
individuals and patients of ischemic stroke (76). The results a signal that can be used as feedback to modulate brain
revealed that in healthy hemisphere scalp vessels dilated in function voluntarily by the patients themselves. Real-time
response to hypercapnia, thus increasing cerebral blood flow, feedback of neural activity to the subjects may modulate plastic
whereas in ischemic side this changes was negligible. This hybrid reorganization without external stimulation of the brain. NIRS,
method could provide more information, with transcranial as a neurofeedback tool detecting blood flow signals, could
improve cortical plasticity and enhance the therapeutic effect NIRS (TD NIRS) to examine muscular metabolism on post-
of motor imagery. In a pilot study twenty hemiplegic patients stroke patients and the result indicated metabolic dysfunction of
with subcortical stroke were analyzed and the results suggested muscles seemed to be local and unilateral (22).
that NIRS-mediated neurofeedback may enhance the efficacy
of mental practice with motor imagery and augment motor
recovery (16). DISCUSSION AND FUTURE DIRECTIONS
NIRS, a non-invasive neuroimaging technique with the
OTHER APPLICATIONS advantages of low price, simplicity, portability and small devices,
has a wide range of utility (48). The existing clinical literature
Apart from applications above, NIRS has been used in evaluation suggests NIRS plays an important role in monitoring motor
of the risk for stroke, clinical studies and monitoring muscle recovery, including upper limb (24–28), lower limb recovery
metabolic indexes (Tables 6, 7). (29–34), balance control (35, 36), motor learning (37), cortical
Continuously monitoring oxygenation levels by NIRS during function recovery (38–47), cerebral hemodynamic changes (48–
operation could diagnose and predict the occurrence of 61), cerebral oxygenation (62–73), therapy (15, 16), and other
perioperative stroke. In a study of 46 patients underwent applications (20–22, 67, 74–83). In particular, its ambulatory
selective antegrade cerebral perfusion of the right subclavian real-time measurement without the fixation of position provides
artery, decreased regional cerebral tissue oxygen saturation great value in tasks requiring constant movement, for example,
was observed. Monitoring of regional cerebral oxygenation gait control.
by NIRS allows detection of clinically important cerebral NIRS is applied in healthy population at first and then
desaturation, and could predict perioperative neurologic sequelae used in stroke subjects. By monitoring of NIRS could reflect
(74). Two studies evaluated perioperative stroke after carotid the function recovery and the therapeutic effect in patients;
endarterectomy (75, 77). NIRS could monitor cerebral changes compared with normal people, it can be used to study the
during the surgery, providing immediate feedback to the treating pathophysiological mechanism of disease and predict the risk,
physician and allowing prompt corrections. Multichannel NIRS serving as an effective monitoring and research tool. In addition,
detects watershed-type posterior perfusion defects, superior to it can be used as therapeutic tool combined with BCI and
single-channel (75). NIRS is also of great importance during neurofeedback, but the current studies mainly limited in healthy
cardiopulmonary bypass (78), cardiac surgery (76), and vessel people, however, the results have shown potential value, requiring
occlusion (67, 79). Treatment of intracranial aneurysms by further investigations in stroke patients.
surgical clipping carries a risk of intraoperative ischemia, caused Despite its non-invasive, portable and ambulatory nature,
mainly by prolonged temporary occlusion of cerebral arteries. its accuracy is influenced by several factors, thus limiting its
NIRS monitoring of cerebral blood flow (CBF) during surgery further application. fNIRS detects neurovascular coupling and its
may decrease the risk (67). Another study proposed multichannel accuracy may be impaired by extracerebral and systemic changes
CW NIRS during endovascular neuroradiologic interventions such as blood pressure and concentration of CO2 (98, 99).
requiring temporary balloon occlusion of arteries supplying the This problem confounds all hemodynamic-based neuroimaging
cerebral circulation (79). Based on these studies, NIRS plays methods, like fMRI. However, due to its high sensitivity, the slight
an essential role in monitoring perioperative stroke, however, blood fluctuations in the scalp may cause “false positives and
a combination of several monitoring modalities may enhance negatives.” We generally assumed the signal changes are only
accuracy and require further investigations. related to neural activity. However, in reality, the fNIRS signals
Studies have also evaluated the risk for stroke in other diseases. are composed of six components: evoked/non-evoked cerebral
We reviewed three papers. Viola et al. assessed hemodynamic neuronal changes and evoked/non-evoked cerebral/extracerebral
changes in migraine patients. The results showed a mild systemic changes (23, 100). These shortages could be minimized
vasoconstriction, which may be the possible explanation of the by careful study design, improved fNIRS techniques and
association between migraine and an increased risk for ischemic statistical processing (23). A modified fNIRS techniques with
stroke (80). The spontaneous cerebral oscillations based on the higher resolution is needed. The newly proposed fNIRS-based
wavelet transform of NIRS signals were assessed to evaluate bundled-optode method used 32 optodes could increase spatial
risk for atherosclerotic stroke (81). Sleep-disordered breathing resolution and this method can be extended for 3D fNIRS
(SDB) could lead to cerebral deoxygenation, resulting in stroke imaging (101, 102). As to obtain higher temporal resolution
outcome. By monitoring oxygenated state of the brain, NIRS for fNIRS-BCI, several studies demonstrated the detection of
could recognize these conditions at early stage (82). the initial dip may be useful. The initial dip refers to an early
In addition, NIRS presents as an effective research tool in small decrease of the concentration of oxy-Hb and is closely
clinical studies. The concentration of oxy-Hb and deoxy-Hb associated with neuronal metabolism. fNIRS-based BCI using
obtained by NIRS could be measured, calculated and analyzed, initial dip detection could reduce the command generation
evaluating hemodynamic, oxygenation changes and cortical time and improve temporal resolution (103, 104). In addition,
function, thus widely used in assessing the effects of different longitudinal studies, for example, measuring with NIRS/fNIRS
therapies or the evolution of diseases (20, 21). Ferrante et al. on one subject every day for a week could provide deeper
combined functional electrical stimulation and time-domain insight the dynamic changes of brain, and as time prolonged
the influence of confounding factors could be minimized. to be effective in many neurological disorders, including stroke,
Furthermore, stroke has specific circadian and circannual Parkinson’s disease, schizophrenia, depression and addiction
rhythms. Most of strokes occurred in the morning between (109). Several studies has suggested its ability to change
7 a.m. and noon (105). Therefore, the time of measurement oscillatory neural activity in resting state, functional connectivity
may influence the efficacy of NIRS and determining the and event related desynchronization/synchronization which are
optimal time window for measurements might be one aspect often injured in patients with motor disorders (110). It could
for optimization of NIRS. Apart from “false positives and also influence membrane potentials of pyramidal neurons and
negatives,” NIRS provides no anatomical information, thus must induce long lasting effects on cortical plasticity by modifying N-
use scalp anatomy to locate the position where the signals methyl-D-aspartate (NMDA) receptor and revealed changes in
arise (52, 106). At present, inhibitory and excitatory activity resting state oscillatory neural activity, functional connectivity,
cannot be discriminated, which requiring the support of other and event related activity during cognitive tasks (111, 112).
techniques (107). These results could partially explain its therapeutic effects, but
Considering the inevitable shortcomings, one promising its exact mechanism still remain unclear. Neuroimaging tool,
prospect is the combination of different neuroimaging including fMRI, PET, EEG and NIRS could partially illuminate
techniques, including diffusion-tensor imaging (DTI), the neurophysiological changes, providing valuable insights
magnetic resonance spectroscopy (MRS), ligand-based into brain changes during electrical stimulation. However, the
positron emission tomography (PET), single-photon emission application of fMRI and PET are restricted due to bulky
computed tomography (SPECT), functional magnetic and immobile properties. EEG are portable and suitable in
resonance imaging (fMRI), NIRS, and EEG. Conjugated combination with tDCS, but its accuracy may be diminished by
with other techniques, the surveillance might be more disturbance of artifacts. fNIRS, based on the measurements of
comprehensive and accurate. Moreover, at present most light intensity, are not affected by electrically induced artifacts
studies have focused on its application in stroke recovery, and might be most desirable tool in tDCS researches (113).
its ability to assess the occurrence of stroke, for example,
during operation and in people with higher risk, is of great AUTHOR CONTRIBUTIONS
significance. Further researches could emphasize NIRS in
the prevention of stroke as well as stroke recovery and PW contributed to the conception of the study. MY contributed
prognosis. significantly to manuscript preparation and wrote the
NIRS could also used in combination with neurostimulation manuscript. ZY, TY, and WF revised the manuscript and
tools. Medical communities have been interested in electrical approved the final version.
stimulation for a long time. Recently transcranial direct current
stimulation (tDCS), which was firstly used in humans in 2,000 FUNDING
by Nitsche et al. (108), has shown to be an appealing therapeutic
tool in neurological disorders. It emits small amounts of electric Gaofeng Medical Technology Talents Project (grant number:
current to modulate the excitability of neuron and has shown 2018061).
brain switch control: an offline study in patients with tetraplegia. IEEE Trans signature of the posture first hypothesis. Brain Topogr. (2015) 29:334–43.
Neural Syst Rehabil Eng. (2014) 22:222–9. doi: 10.1109/TNSRE.2013.2292995 doi: 10.1007/s10548-015-0465-z
16. Mihara M1, Hattori N, Hatakenaka M, Yagura H, Kawano T, Hino T, et al. 35. Mihara M, Miyai I, Hattori N, Hatakenaka M, Yagura H, Kawano T,
Near-infrared spectroscopy-mediated neurofeedback enhances efficacy of et al. Cortical control of postural balance in patients with hemiplegic
motor imagery-based training in poststroke victims: a pilot study. Stroke stroke. NeuroReport (2012) 23:314–9. doi: 10.1097/WNR.0b013e32835
(2013) 44:1091–8. doi: 10.1161/STROKEAHA.111.674507 1757b
17. Pennekamp CW, Bots ML, Kappelle LJ, Moll FL, de Borst GJ. The value 36. Fujimoto H, Mihara M, Hattori N, Hatakenaka M, Kawano T,
of near-infrared spectroscopy measured cerebral oximetry during carotid Yagura H, et al. Cortical changes underlying balance recovery in
endarterectomy in perioperative stroke prevention. A review. Eur J Vasc patients with hemiplegic stroke. Neuroimage (2014) 85:547–54.
Endovasc Surg. (2009) 38:539. doi: 10.1016/j.ejvs.2009.07.008 doi: 10.1016/j.neuroimage.2013.05.014
18. Daly JJ, Wolpaw JR. Brain-computer interfaces in neurological rehabilitation. 37. Hatakenaka M, Miyai I, Mihara M, Yagura H, Hattori N. Impaired motor
Lancet Neurol. (2008) 7:1032–43. doi: 10.1016/S1474-4422(08)70223-0 learning by a pursuit rotor test reduces functional outcomes during
19. Dobkin BH. Brain-computer interface technology as a tool to augment rehabilitation of poststroke ataxia. Neurorehabil Neural Repair (2012) 26:293.
plasticity and outcomes for neurological rehabilitation. J Physiol. (2007) doi: 10.1177/1545968311412053
579:637–42. doi: 10.1113/jphysiol.2006.123067 38. Park SW, Butler AJ, Cavalheiro V, Alberts JL, Wolf SL. Changes in serial
20. Bönöczk P, Panczel G, Nagy Z. Vinpocetine increases cerebral blood optical topography and TMS during task performance after constraint-
flow and oxygenation in stroke patients: a near infrared spectroscopy induced movement therapy in stroke: a case study. Neurorehabil Neural
and transcranial Doppler study. Eur J Ultrasound (2002) 15:85–91. Repair (2004) 18:95–105. doi: 10.1177/0888439004265113
doi: 10.1016/S0929-8266(02)00006-X 39. Schytz HW, Hansson A, Phillip D, Selb J, Boas DA, Iversen HK, et al.
21. De Vis JB, Petersen ET, Kersbergen KJ, Alderliesten T, de Vries LS, Spontaneous low-frequency oscillations in cerebral vessels: applications in
van Bel F, et al. Evaluation of perinatal arterial ischemic stroke using carotid artery disease and ischemic stroke. J Stroke Cerebrovasc Dis. (2010)
noninvasive arterial spin labeling perfusion MRI. Pediat Res. (2013) 74:307– 19:465–74. doi: 10.1016/j.jstrokecerebrovasdis.2010.06.001
13. doi: 10.1038/pr.2013.111 40. Han Q, Zhang M, Li W, Gao Y, Xin Q, Wang Y, et al. Wavelet coherence
22. Ferrante S, Contini D, Spinelli L, Pedrocchi A, Torricelli A, Molteni F, analysis of prefrontal tissue oxyhaemoglobin signals as measured using near-
et al. Monitoring muscle metabolic indexes by time-domain near-infrared infrared spectroscopy in elderly subjects with cerebral infarction. Microvasc
spectroscopy during knee flex-extension induced by functional electrical Res. (2014) 95:108–15. doi: 10.1016/j.mvr.2014.08.001
stimulation. J Biomed Optics (2009) 6629:044011. doi: 10.1117/1.3183802 41. Cao J, Khan B, Hervey N, Tian F, Delgado MR, Clegg NJ, et al. Evaluation
23. Tachtsidis I, Scholkmann F. False positives and false negatives in functional of cortical plasticity in children with cerebral palsy undergoing constraint-
NIRS: issues, challenges and the way forward. Neurophotonics (2016) induced movement therapy based on functional near-infrared spectroscopy.
3:030401. doi: 10.1117/1.NPh.3.3.030401 J Biomed Optics (2015) 20:046009. doi: 10.1117/1.JBO.20.4.046009
24. Kato H, Izumiyama M, Koizumi H, Takahashi A, Itoyama Y. Near-infrared 42. Tan Q, Zhang M, Wang Y, Zhang M, Wang Y, Xin Q, et al. Frequency-
spectroscopic topography as a tool to monitor motor reorganization after specific functional connectivity revealed by wavelet-based coherence analysis
hemiparetic stroke: a comparison with functional MRI. Stroke (2002) in elderly subjects with cerebral infarction using NIRS method. Med Phys.
33:2032–6. (2015) 42:5391–403. doi: 10.1118/1.4928672
25. Takeda K1, Gomi Y, Imai I, Shimoda N, Hiwatari M, Kato H. Shift of motor 43. Oldag A, Neumann J, Goertler M, Hinrichs H, Heinze HJ, Kupsch A, et al.
activation areas during recovery from hemiparesis after cerebral infarction: Near-infrared spectroscopy and transcranial sonography to evaluate cerebral
a longitudinal study with near-infrared spectroscopy. Neurosci Res. (2007) autoregulation in middle cerebral artery steno-occlusive disease. J Neurol.
59:136–44. doi: 10.1016/j.neures.2007.06.1466 (2016) 263:2296–301. doi: 10.1007/s00415-016-8262-5
26. Hara Y, Obayashi S, Tsujiuchi K, Muraoka Y. The effects of 44. Moriya M, Aoki C, Sakatani K. Effects of physical exercise on working
electromyography-controlled functional electrical stimulation on memory and prefrontal cortex function in post-stroke patients. Adv Exp Med
upper extremity function and cortical perfusion in stroke patients. Biol. (2016) 923:203. doi: 10.1007/978-3-319-38810-6_27
Clin Neurophysiol. (2013) 124:2008–15. doi: 10.1016/j.clinph.2013.03.030 45. Saita K, Ogata T, Watanabe J, Tsuboi Y, Takahara M, Inoue T, et al.
27. Gobbo M, Gaffurini P, Vacchi L, Lazzarini S, Villafane J, Orizio C, et al. Contralateral cerebral hypometabolism after cerebellar stroke: a functional
Hand passive mobilization performed with robotic assistance: acute effects near-infrared spectroscopy study. J Stroke Cerebrovasc Dis. (2017) 26:e69–71.
on upper limb perfusion and spasticity in stroke survivors. BioMed Res Int. doi: 10.1016/j.jstrokecerebrovasdis.2017
(2017) 2017:1–6. doi: 10.1155/2017/2796815 46. Mori T, Takeuchi N, Izumi SI. Prefrontal cortex activation during
28. Brunetti M, Morkisch N, Fritzsch C, Mehnert J, Steinbrink J, Niedeggen a dual task in patients with stroke. Gait Posture (2018) 59:193–8.
M, et al. Potential determinants of efficacy of mirror therapy in stroke doi: 10.1016/j.gaitpost.2017.09.032
patients – A pilot study. Restorat Neurol Neurosci. (2015) 33:421–34. 47. White BR, Liao SM, Ferradal SL, Inder TE, Culver JP. Bedside optical
doi: 10.3233/RNN-140421 imaging of occipital resting-state functional connectivity in neonates.
29. Miyai I, Yagura H, Oda I, Konishi I, Eda H, Suzuki T, et al. Premotor cortex Neuroimage (2012) 59:2529–38. doi: 10.1016/j.neuroimage.2011.08.094
is involved in restoration of gait in stroke. Ann Neurol. (2010) 52:188–94. 48. Saitou H, Yanagi H, Hara S, Tsuchiya S, Tomura S. Cerebral blood
doi: 10.1002/ana.10274 volume and oxygenation among poststroke hemiplegic patients: effects of 13
30. Miyai I, Yagura H, Hatakenaka M, Oda I, Konishi I, Kubota K. Longitudinal rehabilitation tasks measured by near-infrared spectroscopy. Arch Phys Med
optical imaging study for locomotor recovery after stroke. Stroke (2003) Rehabil. (2000) 81:1348–56. doi: 10.1053/apmr.2000.9400
34:2866. doi: 10.1161/01.STR.0000100166.81077.8A 49. Terborg C, Bramer S, Harscher S, Simon M, Witte OW. Bedside assessment
31. Mihara M, Miyai I, Hatakenaka M, Kubota K, Sakoda S. Sustained prefrontal of cerebral perfusion reductions in patients with acute ischaemic stroke
activation during ataxic gait: a compensatory mechanism for ataxic stroke?. by near-infrared spectroscopy and indocyanine green. J Neurol Neurosurg
Neuroimage (2007) 37:1338–45. doi: 10.1016/j.neuroimage.2007.06.014 Psychiat. (2004) 75:38–42.
32. Lin PY, Chen JJ, Lin SI. The cortical control of cycling exercise in 50. Terborg C, Gröschel K, Petrovitch A, Ringer T, Schnaudigel S, Witte OW,
stroke patients: an fNIRS study. Hum Brain Mapp. (2014) 34:2381–90. et al. Noninvasive assessment of cerebral perfusion and oxygenation in
doi: 10.1002/hbm.22072 acute ischemic stroke by near-infrared spectroscopy. Eur Neurol. (2009)
33. Rea M, Rana M, Lugato N, Terekhin P, Gizzi L, Brötz D, et al. Lower limb 62:338–43. doi: 10.1159/000239794
movement preparation in chronic stroke: a pilot study toward an fNIRS- 51. Durduran T, Zhou C, Edlow BL, Yu G, Choe R, Kim MN. Transcranial
BCI for gait rehabilitation. Neurorehabil Neural Repair (2014) 28:564–75. optical monitoring of cerebrovascular hemodynamics in acute stroke
doi: 10.1177/1545968313520410 patients. Optics Express (2009) 17:3884–902.
34. Holtzer R, Verghese J, Allali G, Izzetoglu M, Wang C, Mahoney 52. Muehlschlegel S, Selb J, Patel M, Diamond SG, Franceschini MA, Sorensen
JR. Neurological gait abnormalities moderate the functional brain AG, et al. Feasibility of NIRS in the neurointensive care unit: a pilot study in
stroke using physiological oscillations. Neurocritical Care (2009) 11:288–95. cerebral artery occlusions. World Neurosurg. (2018) 110, 145–51.
doi: 10.1007/s12028-009-9254-4 doi: 10.1016/j.wneu.2017.11.023
53. Li Z, Wang Y, Li Y, Wang Y, Li J, Zhang L. Wavelet analysis of 70. Ritzenthaler T, Cho TH, Luis D, Berthezene Y, Nighoghossian N. Usefulness
cerebral oxygenation signal measured by near infrared spectroscopy of near-infrared spectroscopy in thrombectomy monitoring. J Clin Monit
in subjects with cerebral infarction. Microvasc Res. (2010) 80:142–7. Comput. (2015) 29:585–9. doi: 10.1007/s10877-014-9636-9
doi: 10.1016/j.mvr.2010.02.004 71. Nakamura S, Sakatani K, Kano T, Hoshino T, Fujiwara N, Murata Y,
54. Leistner S, Sander-Thoemmes T, Wabnitz H, Moeller M, Wachs M, et al. Effects of revascularisation on evoked cerebral blood oxygenation
Curio G, et al. Non-invasive simultaneous recording of neuronal and responses in stroke patients. Adv Exp Med Biol. (2010) 662, 525–30.
vascular signals in subacute ischemic stroke. Biomed Tech. (2011) 56:85–90. doi: 10.1007/978-1-4419-1241-1_76
doi: 10.1515/bmt.2011.002 72. Vernieri F, Rosato N, Pauri F, Tibuzzi F, Passarelli F, Rossini PM. Near
55. Keller E, Froehlich J, Muroi C, Sikorski C, Muser M. Neuromonitoring infrared spectroscopy and transcranial Doppler in monohemispheric stroke.
in intensive care: a new brain tissue probe for combined monitoring of Eur Neurol. (1999) 41:159–62. doi: 10.1159/000008041
intracranial pressure (ICP) cerebral blood flow (CBF) and oxygenation. Acta 73. Moreau F, Yang R, Nambiar V, Demchuk AM, Dunn JF. Near-infrared
Neurochirurg Suppl. (2011) 110:217–20. doi: 10.1007/978-3-7091-0356-2_39 measurements of brain oxygenation in stroke. Neurophotonics (2016)
56. Oldag A, Goertler M, Bertz AK, Schreiber S, Stoppel C, Heinze HJ, 3:031403. doi: 10.1117/1.NPh.3.3.031403
et al. Assessment of cortical hemodynamics by multichannel near-infrared 74. Olsson C, Thelin S. Regional cerebral saturation monitoring with
spectroscopy in steno-occlusive disease of the middle cerebral artery. Stroke near-infrared spectroscopy during selective antegrade cerebral perfusion:
(2012) 43:2980. doi: 10.1161/STROKEAHA.112.656710 diagnostic performance and relationship to postoperative stroke. J Thorac
57. Han Q, Li Z, Gao Y, Li W, Xin Q, Tan Q, et al. Phase synchronization Cardiovasc Surg. (2006) 131:371–9. doi: 10.1016/j.jtcvs.2005.08.068
analysis of prefrontal tissue oxyhemoglobin oscillations in elderly subjects 75. Nakamura S, Kano T, Sakatani K, Hoshino T, Fujiwara N, Murata Y, et al.
with cerebral infarction. Med Phys. (2014) 41:102702. doi: 10.1118/1.4896113 Optical topography can predict occurrence of watershed infarction during
58. Zirak P, Delgado-Mederos R, Dinia L, Carrera D, Martí-Fàbregas J, carotid endarterectomy: technical case report. Surg Neurol. (2009) 71:540–2.
Durduran T. Transcranial diffuse optical monitoring of microvascular doi: 10.1016/j.surneu.2007.11.012
cerebral hemodynamics after thrombolysis in ischemic stroke. J Biomed. 76. Kellermann K, Jungwirth B. Avoiding stroke during cardiac
Optics (2014) 19:18002. doi: 10.1117/1.JBO.19.1.018002 surgery. Semin Cardiothorac Vasc Anesth. (2010) 14:95–101.
59. Dutta A, Jacob A, Chowdhury SR, Das A, Nitsche MA. EEG-NIRS based doi: 10.1177/1089253210370902
assessment of neurovascular coupling during anodal transcranial direct 77. Pennekamp CW, Moll FL, de Borst GJ. The potential benefits and the role
current stimulation - a stroke case series. J Med Syst. (2015) 39:36. of cerebral monitoring in carotid endarterectomy. Curr Opin Anaesthesiol.
doi: 10.1007/s10916-015-0205-7 (2011) 24:693–7. doi: 10.1097/ACO.0b013e32834c7aa1
60. Mitra S, Bale G, Meek J, Mathieson S, Uria C, Kendall G, et al. In vivo 78. Ono M, Joshi B, Brady K, Easley RB, Zheng Y, Brown C, et al. Risks
measurement of cerebral mitochondrial metabolism using broadband near for impaired cerebral autoregulation during cardiopulmonary bypass and
infrared spectroscopy following neonatal stroke. Adv Exp Med Biol. (2016) postoperative stroke. Br J Anaesth. (2012) 109:391–8. doi: 10.1093/bja/aes148
876:493. doi: 10.1007/978-1-4939-3023-4_62 79. Rummel C, Zubler C, Schroth G, Gralla J, Hsieh K, Abela E, et al. Monitoring
61. Cooper RJ, Hebden JC, O’Reilly H, Mitra S, Michell AW, Everdell NL, et al. cerebral oxygenation during balloon occlusion with multichannel NIRS. J
Transient haemodynamic events in neurologically compromised infants: a Cereb Blood Flow Metab. (2014) 34:347–56. doi: 10.1038/jcbfm.2013.207
simultaneous EEG and diffuse optical imaging study. Neuroimage (2011) 80. Viola S, Viola P, Litterio P, Buongarzone MP, Fiorelli L. Stroke risk and
55:1610–6. doi: 10.1016/j.neuroimage.2011.01.022 migraine: near-infrared spectroscopy study. Neurol Sci. (2012) 33(Suppl.
62. Nemoto EM, Yonas H, Kassam A. Clinical experience with cerebral 1):S173–5. doi: 10.1007/s10072-012-1077-y
oximetry in stroke and cardiac arrest. Critical Care Med. (2000) 28:1052–4. 81. Li Z, Zhang M, Xin Q, Chen G, Liu F, Li J. Spectral analysis of near-infrared
doi: 10.1097/00003246-200004000-00023 spectroscopy signals measured from prefrontal lobe in subjects at risk for
63. Damian MS, Schlosser R. Bilateral near infrared spectroscopy in space- stroke. Med Phys. (2012) 39:2179–85. doi: 10.1118/1.3696363
occupying middle cerebral artery stroke. Neurocritical Care (2007) 6:165–73. 82. Pizza F, Biallas M, Kallweit U, Wolf M, Bassetti CL. Cerebral hemodynamic
doi: 10.1007/s12028-007-0010-3 changes in stroke during sleep-disordered breathing. Stroke (2012) 43:1951–
64. Sakatani K, Murata Y, Fujiwara N, Hoshino T, Nakamura S, Kano T, 3. doi: 10.1161/STROKEAHA.112.656298
et al. Comparison of blood-oxygen-level-dependent functional magnetic 83. Selb J, Yucel MA, Phillip D, Schytz HW, Iversen HK, Vangel M, et al.
resonance imaging and near-infrared spectroscopy recording during Effect of motion artifacts and their correction on near-infrared spectroscopy
functional brain activation in patients with stroke and brain tumors. J Biomed oscillation data: a study in healthy subjects and stroke patients. J Biomed Opt.
Opt. (2007) 12:062110. doi: 10.1117/1.2823036 (2015) 20:56011. doi: 10.1117/1.JBO.20.5.056011
65. Aries MJ, Coumou AD, Elting JW, van der Harst JJ, Kremer BP, Vroomen PC. 84. Chaudhary U, Birbaumer N, Ramos-Murguialday A. Brain-computer
Near infrared spectroscopy for the detection of desaturations in vulnerable interfaces in the completely locked-in state and chronic stroke. Progr Brain
ischemic brain tissue: a pilot study at the stroke unit bedside. Stroke (2012) Res. (2016) 228:131–61. doi: 10.1016/bs.pbr.2016.04.019
43:1134–6. doi: 10.1161/STROKEAHA.111.636894 85. Coyle S, Ward, T, Markham C. On the suitability of near-infrared (NIR)
66. Mundiyanapurath S, Schonenberger S, Rosales ML, Carrilho Romeiro systems for next-generation brain–computer interfaces. Physiol Measure.
AM, Mohlenbruch M, Bendszus M, et al. Circulatory and respiratory (2004) 25:815–822. doi: 10.1088/0967-3334/25/4/003
parameters during acute endovascular stroke therapy in conscious sedation 86. Coyle SM, Ward TE, Markham CM. Brain–computer interface using a
or general anesthesia. J Stroke Cerebrovasc Dis. (2015) 24:1244–9. simplified functional near-infrared spectroscopy system. J Neur Eng. (2007)
doi: 10.1016/j.jstrokecerebrovasdis.2015.01.025 4:219–26. doi: 10.1088/1741-2560/4/3/007
67. Liebig T, Gralla J, Schroth G. Endovascular treatment of acute stroke: 87. Sitaram R, Zhang H, Guan C, Thulasidas M, Hoshi Y, Ishikawa A, et al.
evolution and selection of techniques and instruments based on Temporal classification of multichannel near-infrared spectroscopy signals
thrombus imaging. Clin Neuroradiol. (2015) 25(Suppl. 2), 299–306. of motor imagery for developing a brain–computer interface. Neuroimage
doi: 10.1007/s00062-015-0435-6 (2007) 34:1416–27. doi: 10.1016/j.neuroimage.2006.11.005
68. Hametner C, Stanarcevic P, Stampfl S, Rohde S, Veltkamp R, Bösel J. 88. Lisa H, Martin W. Single-trial classification of motor imagery differing in
Noninvasive cerebral oximetry during endovascular therapy for acute task complexity: a functional near-infrared spectroscopy study. J NeuroEng
ischemic stroke: an observational study. J Cereb Blood Flow Metab. (2015) Rehabil. (2011) 8:34–34. doi: 10.1186/1743-0003-8-34
35:1722–8. doi: 10.1038/jcbfm.2015.181 89. Naseer N, Hong KS. Classification of functional near-infrared spectroscopy
69. Hashimoto T, Matsumoto S, Ando M, Chihara H, Tsujimoto A, Hatano signals corresponding to the right- and left-wrist motor imagery for
T. Cerebral hyperperfusion syndrome after endovascular reperfusion development of a brain–computer interface. Neurosci Lett. (2013) 553:84–9.
therapy in a patient with acute internal carotid artery and middle doi: 10.1016/j.neulet.2013.08.021
90. Hong KS, Naseer N, Kim YH. Classification of prefrontal and motor 104. Zafar A, Hong KS. Neuronal activation detection using vector phase analysis
cortex signals for three-class fNIRS–BCI. Neurosci Lett. (2015) 587:87–92. with dual threshold circles: a functional near-infrared spectroscopy study.
doi: 10.1016/j.neulet.2014.12.029 Neural Syst. (2018) 24:1850031. doi: 10.1142/S0129065718500314
91. Fazli S, Mehnert J, Steinbrink J, Curio G, Villringer A, Müller KR, 105. Gallerani M, Manfredini R, Ricci L, Cocurullo A, Goldoni C, Bigoni M, et al.
et al. Enhanced performance by a hybrid NIRS–EEG brain computer Chronobiological aspects of acute cerebrovascular diseases. Acta Neurologica
interface. Neuroimage (2012) 59:519–29. doi: 10.1016/j.neuroimage.2011. Scand. (2010) 87:482–7. doi: 10.1111/j.1600-0404.1993.tb04141.x
07.084 106. Calderon-Arnulphi M, Alaraj A, Amin-Hanjani S, Mantulin WW, Polzonetti
92. Khan MJ, Hong KS. Hybrid EEG–fNIRS-based eight-command decoding CM, Gratton E, et al. Detection of cerebral ischemia in neurovascular
for BCI: application to quadcopter control. Front Neurorobot. (2017) 11:6. surgery using quantitative frequency-domain near-infrared spectroscopy. J
doi: 10.3389/fnbot.2017.00006 Neurosurg. (2007) 106:283–90. doi: 10.3171/jns.2007.106.2.283
93. Hong KS, Khan MJ. Hybrid brain–computer interface techniques for 107. Eliassen JC, Boespflug EL, Lamy M, Allendorfer J, Chu WJ, Szaflarski
improved classification accuracy and increased number of commands: a JP. Brain-mapping techniques for evaluating poststroke recovery
review. Front Neurorobot. (2017) 11:35. doi: 10.3389/fnbot.2017.00035 and rehabilitation: a review. Topics Stroke Rehabil. (2008) 15:427–50.
94. Hong KS, Khan MJ, Hong MJ. Feature extraction and classification methods doi: 10.1310/tsr1505-427
for hybrid fNIRS-EEG brain-computer interfaces. Front Hum Neurosci. 108. Nitsche MA, Paulus W. Excitability changes induced in the human motor
(2018) 12:246. doi: 10.3389/fnhum.2018.00246 cortex by weak transcranial direct current stimulation. J Physiol. (2000)
95. Koo B, Lee HG, Nam Y, Kang H, Koh CS, Shin HC, et al. A 527:633–9.
hybrid NIRS-EEG system for self-paced brain computer interface with 109. Roy A, Baxter B, He B. High-definition transcranial direct current
online motor imagery. J Neurosci Methods (2015) 244(Sp. Iss. SI):26–32. stimulation induces both acute and persistent changes in broadband cortical
doi: 10.1016/j.jneumeth.2014.04.016 synchronization: A simultaneous tDCS-EEG study. IEEE Trans. Biomed.
96. Yin X, Xu B, Jiang C, Fu Y, Wang Z, Li H, et al. A hybrid BCI based on EEG Eng. (2014). 61:1967–78. doi: 10.1109/TBME.2014.2311071
and fNIRS signals improves the performance of decoding motor imagery 110. Kasashima Y, Fujiwara T, Matsushika Y, Tsuji T, Hase K, Ushiyama
of both force and speed of hand clenching. J Neural Eng. (2015) 12:036004. J, et al. Modulation of event-related desynchronization during motor
doi: 10.1088/1741-2560/12/3/036004 imagery with transcranial direct current stimulation (tDCS) in patients
97. Buccino AP, Keles HO, Omurtag A. Plos ONE (2016) with chronic hemiparetic stroke. Exp Brain Res. (2012) 221:263–8.
doi: 10.1371/journal.pone.0146610.11:e0146610. doi: 10.1007/s00221-012-3166-9
98. Kirilina E, Jelzow A, Heine A, Niessing M, Wabnitz H, Brühl R, 111. Rahman A, Reato D, Arlotti M, Gasca F, Datta A, Parra LC, et al.
et al. The physiological origin of task-evoked systemic artefacts in Cellular effects of acute direct current stimulation: somatic and
functional near infrared spectroscopy. Neuroimage (2012) 61:70–81. synaptic terminal effects. J Physiol (Lond) (2013) 591(Pt 10):2563–78.
doi: 10.1016/j.neuroimage.2012.02.074 doi: 10.1113/jphysiol.2012.247171
99. Caldwell M, Scholkmann F, Wolf U, Wolf M, Elwell C, Tachtsidis 112. Jones KT, Gözenman F, Berryhill ME. The strategy and
I. Modelling confounding effects from extracerebral contamination motivational influences on the beneficial effect of neurostimulation:
and systemic factors on functional near-infrared spectroscopy. a tDCS and fNIRS study. Neuroimage (2015) 105:238–47.
NeuroImage (2016) 143:91–105. doi: 10.1016/j.neuroimage. doi: 10.1016/j.neuroimage.2014.11.012
2016.08.05 113. Yaqub MA, Woo SW, Hong KS. Effects of HD-tDCS on resting-state
100. Zimeo Morais GA, Scholkmann F, Balardin JB, Furucho RA, de Paula RCV, functional connectivity in the prefrontal cortex: an fNIRS study. Complexity
Biazoli CE, et al. Non-neuronal evoked and spontaneous hemodynamic (2018) 2018:1613402. doi: 10.1155/2018/1613402
changes in the anterior temporal region of the human head may lead
to misinterpretations of functional near-infrared spectroscopy signals. Conflict of Interest Statement: The authors declare that the research was
Neurophotonics (2018) 5:011002. doi: 10.1117/1.NPh.5.1.011002 conducted in the absence of any commercial or financial relationships that could
101. Nguyen HD, Hong KS. Bundled-optode implementation for 3D imaging be construed as a potential conflict of interest.
in functional near-infrared spectroscopy. Biomedical Optics Express (2016)
7:3491. doi: 10.1364/BOE.7.003491 Copyright © 2019 Yang, Yang, Yuan, Feng and Wang. This is an open-access article
102. Nguyen HD, Hong KS, Shin YI. Bundled-optode method in distributed under the terms of the Creative Commons Attribution License (CC BY).
functional near-infrared spectroscopy. PLoS ONE (2016) 11:e0165146. The use, distribution or reproduction in other forums is permitted, provided the
doi: 10.1371/journal.pone.0165146 original author(s) and the copyright owner(s) are credited and that the original
103. Zafar A, Hong KS. Detection and classification of three-class initial publication in this journal is cited, in accordance with accepted academic practice.
dips from prefrontal cortex. Biomedical Optics Express (2016) 8:367. No use, distribution or reproduction is permitted which does not comply with these
doi: 10.1364/BOE.8.000367 terms.