ArticlePDF Available

Red and NIR light dosimetry in the human deep brain

Authors:

Abstract and Figures

Photobiomodulation (PBM) appears promising to treat the hallmarks of Parkinson's Disease (PD) in cellular or animal models. We measured light propagation in different areas of PD-relevant deep brain tissue during transcranial, transsphenoidal illumination (at 671 and 808 nm) of a cadaver head and modeled optical parameters of human brain tissue using Monte-Carlo simulations. Gray matter, white matter, cerebrospinal fluid, ventricles, thalamus, pons, cerebellum and skull bone were processed into a mesh of the skull (158 × 201 × 211 voxels; voxel side length: 1 mm). Optical parameters were optimized from simulated and measured fluence rate distributions. The estimated μeff for the different tissues was in all cases larger at 671 than at 808 nm, making latter a better choice for light delivery in the deep brain. Absolute values were comparable to those found in the literature or slightly smaller. The effective attenuation in the ventricles was considerably larger than literature values. Optimization yields a new set of optical parameters better reproducing the experimental data. A combination of PBM via the sphenoid sinus and oral cavity could be beneficial. A 20-fold higher efficiency of light delivery to the deep brain was achieved with ventricular instead of transcranial illumination. Our study demonstrates that it is possible to illuminate deep brain tissues transcranially, transsphenoidally and via different application routes. This opens therapeutic options for sufferers of PD or other cerebral diseases necessitating light therapy.
Content may be subject to copyright.
2921
Physics in Medicine & Biology
Red and NIR light dosimetry in the human
deep brain
APitzschke1, BLovisa1,2, OSeydoux1, MZellweger1,
MPeiderer2, YTardy2 and GWagnières1
1 Federal Institute of Technology (EPFL), Institute of Chemical Sciences
and Engineering (ISIC), 1015 Lausanne, Switzerland
2 Medos International Sàrl, Chemin Blanc 38, 2400 Le Locle, Switzerland
E-mail: andreas.pitzschke@chuv.ch and georges.wagnieres@ep.ch
Received 6 August 2014, revised 24 January 2015
Accepted for publication 12 February 2015
Published 19 March 2015
Abstract
Photobiomodulation (PBM) appears promising to treat the hallmarks of
Parkinson’s Disease (PD) in cellular or animal models. We measured light
propagation in different areas of PD-relevant deep brain tissue during
transcranial, transsphenoidal illumination (at 671 and 808nm) of a cadaver
head and modeled optical parameters of human brain tissue using Monte-
Carlo simulations. Gray matter, white matter, cerebrospinal uid, ventricles,
thalamus, pons, cerebellum and skull bone were processed into a mesh of the
skull (158×201×211 voxels; voxel side length: 1mm). Optical parameters
were optimized from simulated and measured uence rate distributions. The
estimated μeff for the different tissues was in all cases larger at 671 than at
808 nm, making latter a better choice for light delivery in the deep brain.
Absolute values were comparable to those found in the literature or slightly
smaller. The effective attenuation in the ventricles was considerably larger than
literature values. Optimization yields a new set of optical parameters better
reproducing the experimental data. A combination of PBM via the sphenoid
sinus and oral cavity could be benecial. A 20-fold higher efciency of light
delivery to the deep brain was achieved with ventricular instead of transcranial
illumination. Our study demonstrates that it is possible to illuminate deep
brain tissues transcranially, transsphenoidally and via different application
routes. This opens therapeutic options for sufferers of PD or other cerebral
diseases necessitating light therapy.
Keywords: light dosimetry, tissue optics, NIR, brain, photobiomodulation,
Parkinson, substantia nigra
(Some gures may appear in colour only in the online journal)
A Pitzschke et al
Red and NIR light dosimetry in the human deep brain
Printed in the UK
2921
PMB
© 2015 Institute of Physics and Engineering in Medicine
2015
60
Phys. Med. Biol.
PHMBA7
0031-9155
10.1088/0031-9155/60/7/2921
Paper
7
2921
2644
Physics in Medicine & Biology
Institute of Physics and Engineering in Medicine
IOP
0031-9155/15/072921+17$33.00 © 2015 Institute of Physics an d Engineering in Me dicine Printed in the UK
Phys. Med. Biol. 60 (2015) 2921–2937 doi:10.1088/0031-9155/60/7/2921
2922
1. Introduction
Parkinson’s Disease (PD) is a common neurodegenerative disorder of middle-aged and
elderly people, affecting 1.8% of individuals over 65 (Morais and De Strooper 2010) and up
to 4% of the population over 80 (Davie 2008, Lees et al 2009). The pathological hallmark
of PD is the loss of dopaminergic neurons from the pars compacta of the brain’s substantia
nigra (SNpc) (McNaught and Jenner 2001, Samii et al 2004, Davie 2008, Lees et al 2009),
although cell loss in other structures of the brain is also reported (Damier et al 1999). The
cell loss is accompanied by other features, such as intraneuronal inclusions made of abnor-
mal, aggregated forms of the protein α-synuclein called Lewy bodies (McNaught and Jenner
2001, Kordower et al 2008, Lees et al 2009, Morais and De Strooper 2010) and breakdown
of mitochondrial metabolism and functionality (Lees et al 2009, Trimmer et al 2009, Morais
and De Strooper 2010). Whilst PD is an incurable disease, treatments improve quality of
life and functional capabilities of patients. Treatment approaches include oral, injected or
enteric administered drugs, but are often associated with side effects (Davie 2008, Lees
et al 2009). Other treatment options have been explored but have yet to nd their way into the
mainstream treatment of PD (fetal tissue transplantation) or have been abandoned (surgery)
(Samii et al 2004, Temel et al 2006, Davie 2008, Lees et al 2009). Deep-brain stimulation
is now widely accepted for selected patients that are no longer responsive to drug treatment,
yet only symptoms can be treated. To this day, there is no therapeutic option that brings neu-
roprotection or neuroregeneration to patients, although some promising results are reported
on animal models (Shaw et al 2010, Shaw et al 2012). Photobiomodulation (PBM) has been
described to bring positive results by restoring mitochondrial functionality in metabolically
and functionally compromised PD tissue cell lines (Hamblin and Demidova 2006, Trimmer
et al 2009) and in animal models such as the Drosophila (Vos et al 2013) or mouse (Shaw
et al 2010). The hypothesized mechanism is that the red/NIR-light photostimulation of the
diseased tissue improves the electron transport chain function and restores some functional
activity of mitochondria, leading to the restoration of locomotor functions in the animal mod-
els (Moro et al 2013, Vos et al 2013).
The objective of the present study was to explore experimentally the light delivery and
dosimetry for a transcranial, transsphenoidal photostimulation of the SNpc as a possible treat-
ment option for PD patients. We wanted to evaluate how the sphenoid sinus (S.S.) can be
accessed with a light diffuser and how it can be positioned and oriented towards the SNpc,
if necessary with the help of an endoscope. This allowed to quantify the light distribution
within brain tissue when illuminating from the nasal cavity and to dene the best treatment
regimes with a controlled energy deposition. Since light dosimetry depends, among others,
on the tissue optical properties, we had to model the intracranial uence rate distribution by
Monte Carlo simulations based on the real patient skull mesh. An iterative algorithm lead
to a self-consistent set of optical parameters for the model’s tissues. The model was vali-
dated by directly measuring the uence rate distribution in selected intracranial locations of
a human cadaveric head, from the dura mater just behind the S.S., the pituitary gland, to the
most posterior-inferior area of the treatment target, the SNpc. Furthermore, our self-consistent
model enabled us to explore the light delivery to the SNpc using an interventricular/interstitial
approach. The wavelengths of 671 and 808nm chosen in our study correspond to the maxima
in the action spectrum of cytochrome c oxidase, which is the terminal enzyme of the respira-
tory chain driving the ATP production in the mitochondria (Karu 1999). The results of this
work shall provide useful information to determine whether it is feasible to deliver sufcient
illumination power to the relevant brain area to induce a therapeutic effect, whilst preserving
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2923
the structures located between light diffuser and target area. It sheds light onto the exact power
density that reaches specic brain regions.
2. Materials and methods
2.1. Measurement of the uence rate distribution
2.1.1. Human cadaver. The measurement of the uence rate distribution was carried out on
a human cadaveric head (56 years, female, 61kg, primary cause of death: chronic obstructive
pulmonary disease, head dissected from the torso between the 3rd and 4th cervical verte-
bra) with no sign of head injury. The specimen was purchased from Innoved Institute, LLC
(Bensenville, IL, USA). The supplier kept the specimen refrigerated at 4°C post-mortem for
4d and then frozen for 20 months at −20 °C prior to the delivery. No additional antifreeze
agent or perfusion was applied.
Prior to MRI scan at the supplier facility, the specimen was taken from the fridge and
kept at 4°C during 4d. 10h before the scan, the specimen was left at ambient room tem-
perature to ensure thaw. Immediately after imaging, the specimen was placed back into
the freezer at−20°C and then distributed to the measurement lab on dry ice. Upon deliv-
ery, the specimen was kept in the fridge at 8°C for one day to ensure full thaw. Titanium
bone screws (4×1.5mm, Bioplate, Los Angeles, CA) were xed to the skull prior to the
CT scan and optical measurements to serve as reference positions for registration of the
head coordinates. A CT scan was carried out the day before the optical measurement at
the measurement lab. The specimen was placed back in the fridge at 8°C overnight prior
to the optical measurements. The coregistration of MRI and CT image datasets followed
standard protocols of Brainlab’s planning and navigation software (Navigation System
Kolibri 2.0, Navigation Software Cranial 2.1 and Planning Software iPlan RT, Brainlab
AG, Feldkirchen b. M., Germany).
The measurements were carried out at ambient temperature; no further MRI or CT scans
were performed after the optical measurements. Soft tissue geometry was assumed to be unal-
tered between CT/MRI scans and optical measurements. A rigid coordinate transformation was
used, which could result in some minor positioning uncertainties due to tissue displacement.
2.1.2. Light diffuser and isotropic probes for in situ uence rate measurements. The red and
NIR light was delivered by an optical ber-based frontal light diffuser (FD1, Medlight SA,
Ecublens, Switzerland) introduced into the nasal cavity. This light distributor emitted light
in a cone shape (full aperture angle 34.7 degree in the VIS/NIR wavelength range) towards
the SNpc. The FD1 was coupled to a laser diode at 671nm (FSDL-671, Frankfurter Laser,
Friedrichshof, Germany) or 808nm (RLTMDL-808-5W, 5W, Roithner Lasertechnik GmbH,
Vienna, Austria).
Optical ber-based isotropic probes (IP85, Medlight SA, Switzerland) with a diameter
of 0.85mm were used to measure in situ the uence rate along the trajectories of ve cath-
eters (iCAT-2.0-200, OD 2mm, Medlight SA, Ecublens, Switzerland), which were inserted at
ve positions into the deep brain (gure 1). The inuence of the catheter on the uence rate
measurements, i.e. light attenuation and waveguide effects, was determined by illuminating
the isotropic probes with and without the catheter in an integrating sphere (LMS-200, 20
inch diameter, Labsphere, Inc., USA) or directly with a calibrated collimated light beam at
671 and 808nm. The changes in the measured uence rate due to the presence of a catheter
were smaller than 5%. Only at the catheter tip probe shading became prominent, such that
experimental data acquired close to the tip (usually for probe-tip distances below 3mm) were
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2924
excluded from data analysis. The isotropic probes were coupled to a calibrated multi-channel
optical power meter (OP710-IN, OptoTest Corporation, Camarillo, CA, USA).
2.1.3. Protocol for the ex vivo measurement of the intracranial uence rate distribution. The
position of the SNpc was determined on the pre-op MRI by identifying the Red Nucleus, to
which the SNpc is situated lateral and anterior. The ve target positions (SNpc right posterior,
right anterior, left posterior, left anterior and pituitary gland, see gure1) on the MRI were
taken as input for the planning software. The skull entry points on the frontal lobe were selected
with the help of the navigation system in such a way that the ve catheters were orthogonal to
the FD1 in the sagittal view as well as parallel to one another. Five burr holes were drilled in
the skull at the entry positions and the ve catheters (modied to a length of 160mm, Brainlab
stylet with 1.1mm diameter) were guided by the navigation system to a depth in the brain that
surpassed the four target points of the SNpc by approximately 5mm. The catheter into the
pituitary gland was guided down to the sellaturcica in contact with the sphenoidal bone. The
catheters in the SNpc were found to be within 2mm of the target location selected on the pre-op
MRI. The position of the catheter in the pituitary gland was conrmed by uoroscopy.
In the sphenoidal sinus (S.S.), the placement of the light diffuser was carried out under
endoscopy and uoroscopy. The resulting location and orientation of the diffuser, guided by a
10mm long steel tube (Relieva Sinus Guide Catheter, Acclarent, Menlo Park, CA, USA), were
controlled and documented by the navigation software. The S.S. was prepared by inserting a
balloon sinuplasty system (Relieva, Acclarent, Menlo Park, CA, USA) and resecting the sep-
tum between the left and right S.S. After irrigation and suction to clear the sphenoidal cavity
of remaining ice and mucus, a sinus catheter (Relieva Sinus Guide Catheter, Acclarent, Menlo
Park, CA, USA) was inserted into the S.S. and positioned right after the ostium, thereby keep-
ing a distance of 15–20mm from the posterior sinus mucosa. The FD1 was introduced through
the sinus catheter with its tip just surpassing the tip of the catheter, such that it produced a
circular homogenous light spot of approximately 1cm2 on the sinus mucosa. For alignment
and calibration purposes of the FD1, the light of a green laser pointer was injected into the
ber and its spot position on the posterior wall was veried endoscopically with the help of
a rigid rhinoscope. The laser power from the FD1 was set to 1W at 808nm and to 0.8W at
671nm (maximum output power of the laser diode) with the help of a power meter (detec-
tor 818P-010-12, driver 1918-R, Spectra Physics Newport). The ve isotropic probes were
inserted into the catheters and the background signal was measured with the laser switched
off. With laser on, measurements were performed by withdrawing synchronously each probe
inside the catheters in nine steps of 5mm each with the starting point on the catheters’ distal
Figure 1. Transaxial view at the level of the SNpc with the catheters placed into the
pituitary gland, anterior and posterior SNpc. The SNpc is colorized in brown.
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2925
end. The measurement sequence was repeated ve times to assess typical uence rate varia-
tions for each measurement.
2.2. Modeling of the light propagation
2.2.1. Analytical model based on the diffusion approximation. Many analytical models
describing the propagation of light in biological tissue are based on the diffusion approxi-
mation. The steady-state expression of the photon diffusion within a homogenous medium
(Ishimaru 1978) is given by
μϕ∇− =−qDrr
()(
)(
)/ ,
2eff
2
(1)
where ϕ (r) is the uence rate (mWcm−2), μeff=[3μa(μa+μs(1−g))]1/2 the effective attenua-
tion coefcient (cm−1),
μμ=
D/
aeff
2
the optical diffusivity (cm) and q the diffusive photon den-
sity (mWcm−3). The absorption and scattering coefcients, μa (cm−1) and μs (cm−1), express
the probability for a photon to be absorbed or scattered on a given path length. Analytical solu-
tions of equation(1) for point, line and plane sources can be found in the literature (Tromberg
et al 1996, Jacques 1998).
Radiation in the visible and near-visible range is normally subject to anisotropic scatter-
ing in tissue, which is the result of inhomogeneities of the refractive index. The Henyey–
Greenstein phase function (Henyey and Greenstein 1941) is frequently used to model the
scattering angle probability density function and is given by
θ
πθ
=
+−
pg
gg
() 1
4
(
12cos
)
,
2
23/2
(2)
where g is the anisotropy factor. It is equal to the mean cosine of the scattering angle θ and
ranges from−1, total back scattering, over 0, isotropic (Rayleigh) scattering, to+1, total for-
ward scattering (Mie scattering at large particles). In the diffusion theory, the anisotropic
scattering is simplied to the isotropic case by dening the reduced scattering coefcient
μ
s=μs(1−g)(cm−1).
This analytical model is in general accurate if μaμ
s and if the light distribution is studied
far enough from the light distributor and boundaries, typically at a distance much larger than
1/μ
s. The model becomes unwieldy in complex geometries, e.g. a human skull with different
tissue types or more complex light source geometries. In these more complicated cases, rem-
edy can be found in using numerical approaches, e.g. Monte Carlo methods.
2.2.2. Monte Carlo model of the uence rate. Monte Carlo methods involve a broad class of
computational algorithms that rely on repeated random sampling to obtain numerical results
(Wang et al 1995, Wang and Wu 2007, Zhu and Liu 2013). They are most suited when it is
impossible to obtain a closed-form expression of the uence rate or when deterministic algo-
rithms cannot be applied. Nowadays algorithms are capable to efciently simulate complex
geometries and source emission patterns, boundary conditions, local changes in the refractive
index and optical parameters. In this study, the Monte Carlo algorithm MCX (Fang and Boas
2009a), accelerated by graphics processing units, was used to model the photon migration in
3D. The code is implemented with NVIDIA CUDA programming and was validated, with the
3D skull model, against MMC (Fang 2010) and TracePro (Lambda Research Corporation,
Littletown, MA, USA). The codes were run on an Intel i7-3930K, 32GB RAM and a NVIDIA
GTX 670 graphics card with 2GB RAM.
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2926
The optimal number of photons for a simulation with MCX was determined from a statis-
tical noise analysis. For that purpose, a semi-ininite cube (side length 20cm, optical coef-
cients for the human pons (Yaroslavsky et al 2002): μa=0.6cm−1, μs=80cm−1, g=0.92,
n=1.37 refractive-index-matched) was illuminated by an FD1. The algorithm computed the
uence rate distribution in the cube as function of the photon number. For every number of
photons, ten computations were carried out, each initialized by a different random prime num-
ber to vary the series of random numbers used to compute photon trajectories and absorption
events. Photons with relative energy below 10−3 of the initial energy were truncated (Cassidy
et al 2013). In gure 3, the relative statistical error in the uence rate, δϕ/ϕ, is shown
as function of the number of photons and distance between source and detector, where δϕ
expresses the standard deviation of the uence rate. The results indicate that the relative error
in the uence rate becomes minimal beyond a number of 109 photons even for the longest
distances between source and isotropic probes in the experiment, i.e. 50–60mm. Therefore,
109 photons were chosen for all simulations as a good compromise between accuracy and
computational time.
The 3D model of the head was assembled from the MRI and CT reconstruction (see gure2).
In our model, only gray and white matter (GM & WM), cerebrospinal uid (CSF), ventricles
(CSF), thalamus, pons, cerebellum and skull bone were considered. These geometries were
then processed with the help of the iso2mesh toolbox (Fang and Boas 2009b) to produce a
Figure 2. View on the 3D mesh of the skull. The frontal light distributor (located at
the origin of the red cone) placed in the nasal cavity is illuminating the SNpc through
bone ( 2mm, dark mauve), CSF ( 2mm, orange), gray matter ( 2mm, light green),
white matter ( 7mm, dark green) and brainstem ( 4–18mm, light mauve) tissue. The
thalamus is depicted in red, the cerebellum in gray. The catheters are shown as colored
trajectories with circular marks representing the individual positions of the isotropic
probes during the measurement. The catheter color coding is identical with the one
used in gure1.
80
100
120
140 0
50
100
150
0
20
40
60
80
100
120
140
160
180
y [mm]
x [mm]
z [mm]
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2927
tetrahedral and cubic mesh for MCX, MMC and TracePro for cross-verication purposes. The
resulting cubic mesh of the skull had a size of 158×201×211 voxels with a voxel side length
of 1mm. The tetrahedral mesh consisted of around half a million tetrahedra with a distribution
of volumes ranging between 1–30mm3 and a mean around 13mm3. For the eight tissues men-
tioned above, optical parameters, i.e. the scattering coefcient μs, the absorption coefcient μa,
the refractive index n and the anisotropy factor g, all dependent on the wavelength, were taken
from literature (Firbank et al 1993, Bevilacqua et al 1999, Yaroslavsky et al 2002, Koyama
et al 2005, Custo et al 2006).
2.3. Determination of the optical parameters
Ab initio’ computations of the intracranial uence rate distribution with optical parameters
from literature showed signicant differences from what was measured experimentally (see
below in section3). These differences were attributed to various uncertainties in our model:
(1) The optical tissue coefcients for this specimen were unknown, thus initial coefcients had
to be taken from literature. However, these values might be inappropriate when one considers
their large uncertainties reported in literature (Roggan 1997, Tuchin 2007). (2) Furthermore,
tissue storage including freezing and thawing procedures is known to alter optical param-
eters (Roggan et al 1999, Salomatina and Yaroslavsky 2008, Pitzschke et al 2015) making
it unlikely to match the uence rate from an ‘ab initio’ calculation with the measurements.
(3) The positions of the isotropic probes in the catheters during the measurements were only
exact to 1–2mm. Since the uence rate decays approximately exponentially as function of the
distance between source and detector position, small uncertainties in local positioning lead to
non-negligible errors in the uence rate. (4) The mesh construction is, in general, only pre-
cise to a certain extent. The identication of different tissue types on gray scale MRI images,
reconstruction of 3D meshes, smoothing procedures etc yield a skull model, whose geometric
precision was not better than a few millimeters. (5) The computation of the uence rate based
on the statistical Monte Carlo approach has itself uncertainties increasing drastically with the
Figure 3. Relative error in the uence rate, δϕ/ϕ, at different distances between source
and SNpc as function of the number of photons. δϕ expresses the standard deviation of
the uence rate ϕ. The relative error was computed by MCX from ten iterations.
1051061071081091010
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
number of photons
relative statistical error
1 mm
5 mm
10 mm
20 mm
30 mm
40 mm
50 mm
60 mm
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2928
dimensions of the modeled system (see gure3). The error sources (1), (2) and (4) are of sys-
tematic nature and were corrected for, (3) and (5) are uncorrectable random errors.
To remedy this situation, the optical parameters were derived from a χ2-iteration procedure
optimizing the match between simulated and measured uence rate distribution. For this pur-
pose, the reduced χ2-distribution function,
χνϕμ
μϕ=′
rr
1
ln ((,, )) ln (())
,
i
iai
red
2sims exp
2
(3)
was minimized for sets of μa and μs, where ν is the degree of freedom, ϕ (r) the simulated and
experimental uence rate at position r. The natural logarithm of the uence rate was taken to
impose a parabolic behavior in
χred
2
and to avoid that small uence rates, measured far away
from the source, are under-weighted. This approach is equivalent to t directly the optical
parameters instead of the uence rate, since μeff(μa, μ
s)ln(ϕ). The minimization algorithm
was based on the Levenberg-Marquardt-method, which is implemented in MATLAB’s built-
in function lsqnonlin (R2013a, The MathWorks, Inc., Natick, MA, USA).
Several simplications of the model were needed to ensure a better convergence of the t-
ting procedure: (1) The scattering anisotropy was kept constant at g=0.9, which may lead to a
bias in μs, but keeps μ
s approximately constant over a wide range of g and μs. (2) The refractive
indices were set to n=1.37 for tissue and bone and to n=1 for air. (3) The cerebellum was
modeled together with the white matter as a single tissue, since no experimental measure-
ments of the uence rate were carried out in this region of the brain. The optical coefcients
of white matter come closest to what is known for the cerebellum (Yaroslavsky et al 2002,
Tuchin 2007). (4) CSF and ventricles were modeled as a single tissue. (5) Upper and lower
boundaries of μa and μ
s, derived from the optical coefcients' minimal and maximal values
found in literature (Roggan 1997, Yaroslavsky et al 2002, Custo et al 2006, Tuchin 2007),
were set to constrain the t.
These simplications allowed to reduce the number of free parameters from 16 to 12, i.e.
six tissues each with a set of μa and μ
s. Thus, 13 simulations per iteration were necessary to
compute uence rate distribution and its derivative as function of the optical parameters. For
each iteration, the random number generator of MCX was initialized with the same prime
number, such that photon trajectories and absorption as well as absolute values of uence
rate and its derivatives were not subject to statistical noise. Photons with a remaining energy
below 10−3 of the initial value were truncated. The initial step-size in the relative change of
the optical parameters was chosen to be 10% to obtain sensibility of the tting algorithm on
each parameter. Iterations were stopped when either the relative change of
χred
2 or the relative
change in the optical parameters were below 1% ; mainly the latter condition was reached
rst. Several starting points with different optical parameters in the range of± 300% with
respect of the literature values presented in table2 were chosen to ensure global convergence
of the tting algorithm. Usually around ten iterations were sufcient to obtain convergence
and reasonable optical parameters.
3. Results and discussion
3.1. Measured uence rate distribution
The effective attenuation coefcient μe ff was computed from the uence rate measured by the
isotropic probes in the catheter couples anterior-posterior, i.e. 2–3 and 5–4 (see gure1). The
relation
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2929
ϕϕμ
zz() ()e
ij
zd()
ij
eff
(4)
was used to compute μeff(z), where ϕi(z) and ϕj(z) are the uence rates in different catheters
(i, j, i≠j) at a position z of the isotropic probes in the catheter and dij the distances between
the measurement positions. The resulting μeff for the different tissues and wavelengths are
presented in table1 together with typical values from literature.
Equation (4) is a solution of equation(1) that assumes a planar light source and an innite
homogenous medium, which does not exactly correspond to our case. Far from the source,
however, the distance dependence of ϕ is essentially mono-exponential, as has been shown
elsewhere (Bays et al 1997, Jacques 1998). Monte Carlo simulations showed uence rate con-
tours in brainstem and thalamus to be approximately parallel to the catheter trajectories (gure
6(a)), thus validating above taken approximation.
The estimated μeff for the different tissues was, despite the large uncertainties, in all cases
larger at 671 as compared to 808nm. Absolute values were comparable to those found in litera-
ture or slightly smaller, which may be the effect of storing, freezing and thawing or a drain of
liquids from the cadaveric head during the preparation and measurement procedures. The freez-
ing process, especially when performed slowly, is supposed to partially damage cell membranes
leading to changes of the optical parameters (Roggan et al 1999, Salomatina and Yaroslavsky
2008). However, blood drain was shown to have only little effect on the attenuation coefcient
of brain tissue (Pitzschke et al 2015). The effective attenuation in the ventricles was considerably
larger as compared to literature values (Koyama et al 2005). This discrepancy might be due to
impurities in the CSF like debris from tissue degeneration or contamination by blood from the
Arteria pericallosa or other surrounding blood vessels that were damaged by repeated freeze and
thaw of the head. Furthermore, the head resection may have lead to leaking of CSF emptying
the ventricles to some extent. On the MRI data (not shown here), a partial collapse of the left
ventricle was visible, which would explain a somewhat larger μeff for the ventricles.
3.2. Determination of the optical parameters by Monte Carlo simulations
The computation of the uence rate distribution was carried out with a straightforward Monte
Carlo simulation using the tissue optical parameters from the literature (table 2), which were
Table 1. The effective attenuation coefcient
μeff
exp
and its standard deviation (2 SD) in
parentheses at 671 and 808nm was obtained from the uence rate measurements at the
individual positions z (mm) of the isotropic probes in the catheter.
Position z
(mm) Tissue
671nm 808nm
μ
eff
exp
(cm−1)
eff
lit (cm−1)
μ
eff
exp
(cm−1)
eff
lit (cm−1)
0 Pons* 4.3 (1.5) 3.5 2.4 (1.1) 3.6
5 Pons/
thal.*
3.6 (1.8) 2.3 (0.6)
10–20 Thal. 3.4 (0.1) 5.0 2.0 (1.0) 6.0
25 Thal./
ventr.
2.1 (2.5) 1.5 (1.0)
30–40 Ventr. 1.7 (0.9) 1.1 (0.6) 0.5(0.4)
Note: Literature values of
eff
lit for pons and thalamus are taken from Yaroslavsky et al (2002), for
ventricles (CSF) in the NIR from Koyama et al (2005). Slash characters indicate the transition
from one tissue to another around a measurement position. The asterisks indicate positions were
the SNpc was assumed to be located.
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2930
Table 2. The optical coefcients, μa, μ
s and μeff, tting best the uence rate measurements at 671 and 808nm.
Tissue
671nm 808nm Literature values for 671–808nm
μa (cm−1)μ
s (cm−1)μeff (cm−1)μa (cm−1)μ
s (cm−1)μeff (cm−1)μa (cm−1)μ
s (cm−1)μeff (cm−1)
Skull 0.07 (0.17) 12.7 (1.3) 1.6 (2.6) 0.04 (0.17) 5.4 (1.1) 0.8 (2.6) 0.19–0.50 8.6–17 2.2–5.1
CSF 0.15 (0.04) 11.6 (1.2) 2.3 (0.4) 0.05 (0.02) 5.9 (1.6) 1.0 (0.4) 0.04–0.14 0.1–0.3 0.1–0.4
GM 0.08 (0.17) 7.1 (0.9) 1.4 (2.0) 0.06 (0.18) 5.4 (1.2) 1.0 (2.0) 0.20–0.36 7.0–8.4 2.1–3.1
WM 0.27 (0.26) 27.8 (4.6) 4.8 (3.3) 0.31 (0.36) 20.4 (4.6) 4.4 (3.3) 0.70–1.00 41.0–60.0 9.4–13.5
Pons 0.68 (0.31) 3.7 (0.5) 3.0 (1.0) 0.51 (0.22) 3.6 (0.9) 2.5 (1.0) 0.60–0.80 6.4–8.6 3.5–4.7
Thal. 0.68 (0.17) 5.1 (0.8) 3.4 (0.7) 0.23 (0.06) 5.4 (1.5) 2.0 (0.7) 0.44–0.65 17.6–19.0 4.9–6.2
Note: Values in parenthesis correspond to the parameter uncertainties (2 SD). The coefcients derived from literature (Firbank et al 1993, Bevilacqua et al 1999, Yaroslavsky et al
2002, Koyama et al 2005, Custo et al 2006) for the red/NIR-range are given in the last column.
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2931
chosen for their completeness in wavelengths and tissue types. Figures4(a) and (b) show the
simulation results for 671 and 808nm as dashed lines together with the experimental data
asmarkers. For both wavelengths, simulation results using optical parameters derived from
the literature t only poorly the experimental data. In certain conditions the simulated u-
ence rate showed deviations by one order of magnitude from the experimental one. We could
exclude these deviations that were only due to an imperfect mesh construction. Although
Figure 4. The uence rate (mW cm−2) at different detector positions (markers with
errorbars (2 SD)) and for the calculated optical parameters (solid line) is shown in
gures 4(a) and (b) for 671 and 808 nm, respectively. Dashed lines represent the
uence rate computed with optical parameters derived from literature (table 2). The
zero-position on the x-axis corresponds to the caudal limit of the SNpc. Experimental
and simulated data was normalized such that the tissue irradiance of the conic beam
delivered by the FD1 on the sphenoidal bone was 1Wcm−2. (a) Fluence rate at 671nm.
(b) Fluence rate at 808nm.
0 5 10 15 20 25 30 35 40
10−4
10−3
10−2
10−1
100
101
102
detector position [mm]
fluence rate [mW / cm2]
pituitary gland
SNpc right anterior
SNpc right posterior
SNpc left posterior
SNpc left anterior
0 5 10 15 20 25 30 35 40
10−2
10−1
100
101
102
103
detector position [mm]
fluence rate [mW / cm2]
pituitary gland
SNpc right anterior
SNpc right posterior
SNpc left posterior
SNpc left anterior
(b)
(a)
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2932
absolute values of the uence rate might be biased by the mesh, large local over- or underes-
timation of the uence rate is rather an indication of an inappropriate choice of optical tissue
coefcients. The simulated uence rate showed dip behavior in the region of the thalamus (at
the positions around 15mm for catheters located in the SNpc), in contrast to the experimental
data, suggesting a much lower μeff for this region. The simulated uence rate in the ventricles
(at positions above 30mm) was approximately the same for all catheter positions, which cor-
responds to a light cavity with small μeff and thus semi-homogenous uence rate distribution.
In contrast, the experimental data showed much larger differences in the uence rate between
particular catheters, especially at 671nm, which suggests a larger μeff for the CSF in the ven-
tricles. This is consistent with the determined attenuation coefcients in section3.1, table1.
In other words, the ventricles contained a medium with more scattering and absorbing centers
than is usually the case for CSF.
The χ2-optimization of the optical parameters (section 2.3) yield a new set of optical param-
eters leading to a better match of simulated and measured data. The simulated uence rate is
shown as solid lines in gures4(a) and (b). Deviations of simulated uence rate in thalamus
and ventricles from the experimental data were now much smaller. Nevertheless, these simu-
lation results could not perfectly reproduce the experimental data, which may be due to the
simplicity of the brain model as well as an incomplete or missing set of experimental data for
cerebellum, gray and white matter and for the CSF in the subarachnoid space. The values of
optical tissue parameters derived from the χ2-optimization are given in table2 for both wave-
lengths. As already noted in section3.1, there was a tendency of decreasing value of μeff for
increasing wavelength leading to a higher light penetration depth in the brain tissue at 808
than at 671nm (gures 5(a) and (b)). Both, μa and μ
s were smaller than cited in the literature,
apart from the CSF, which we attributed to be an effect of the freezing and thawing procedures
Figure 5. Colored contours of the log10 uence rate (mW cm−2) computed from the
determined optical parameters are indicated in gures5(a) and (b) for 671 and 808nm,
respectively. The sagittal plane of the skull is shown in gray-scale, the orientation of
the FD1 indicated in magenta and the anterior/posterior location of the SNpc as white
asterisks. Contour lines of experimental and simulated data was normalized such that
the tissue irradiance of the conic beam delivered by the FD1 on the sphenoidal bone
was 1Wcm−2. (a) Fluence rate distribution at 671nm. (b) Fluence rate distribution at
808nm.
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2933
(Roggan et al 1999, Pitzschke et al 2015). The higher value of μ
s for the CSF may result from
the ventricles partially collapsed and/or lled by a medium with higher μeff as noted above.
The standard deviation of the optical tissue parameters presented in table2 were relatively
large for certain values of μa, notably for skull, gray and white matter tissue. In these cases, the
hyper-surface of the χ2-function became relatively at, so that a change in a particular coef-
cient affected only little the χ2-cost-function. In other words, since experimental data was
partially missing for these tissues, the model was less sensitive to the corresponding optical
parameters and, as a consequence, yield larger uncertainties for certain coefcients. In a future
study, this issue should be assessed by uence rate measurements performed more thoroughly
in all considered tissues.
As far as the penetration of light to target the SNpc is concerned, the results obtained from
the specimen suggest that light at 808nm is the better choice. This is due to less absorption
and reduced scattering at 808nm in all considered tissue types. Especially when delivering
light by the S.S., the reduced light attenuation in thalamus and pons at this wavelength is
improving the fraction of uence rate delivered in the SNpc. Furthermore, lower absorption
coefcients lead to lower power deposition within the brain tissue as for 671nm, making it
easier to stay below hyperthermia thresholds. However, for choosing an appropriate wave-
length for PBM, the action spectrum must also be taken into account.
3.3. Modeled uence rate distribution for alternative light delivery routes
To date, there have been no major clinical trials on the therapeutic effect of red/NIR light in
patients with PD (Lapchak 2012). One potential problem associated with this treatment relates
to the effective and reliable penetration of red/NIR light through the human cranium, men-
ingeal layers and brain parenchyma to reach the SNpc in the midbrain. Similar issues occur
in studies for treating strokes (Lapchak and Taboada 2010, Sharma et al 2011) or Alzheimer
(De Taboada et al 2011) by PBM. The lack of precise light dosimetry therefore demands the
evaluation of various light delivery routes for their feasibility and efciency. In the following
we present two illustrative examples of our model to evaluate potential light delivery routes.
3.3.1. Illumination of the SNpc by a light diffuser positioned in the oral cavity. A reduction
of severe Parkinsonian signs was reported for one patient, who underwent light exposure for
a dental treatment (Burchman 2011). The patient received light at 660nm in the oral cavity
for a root canal treatment as well as on the acupuncture point Gallbladder GB#20, situated at
the junction of occipital and nuchal regions (Atlas of acupuncture points 2007), to calm the
patient’s tremors. No information was given about the laser power used for the dental treatment.
We modeled the uence rate distribution corresponding to this case with an FD1 located
in the oral cavity pointing towards the SNpc. The output power of the FD1 was chosen such
that the tissue irradiance was 1 W cm−2. The result, shown in gure 6(a) for the derived
cadaver’s optical coefcients at 808 nm (table 2), indicated that a signicant uence rate
could be attained in the SNpc by this light delivery route, although the efciency compared to
illumination via the S.S. was smaller by more than one order of magnitude (table 3). Absolute
values became even smaller when applying light at 671nm, close to what was used during the
dental treatment. The fraction of uence rate in the SNpc by shining light only on GB #20, i.e.
without an FD1 in the oral cavity, was10−4mWcm−2 at 808nm and even lower at 671nm.
In MPTP-treated mice it was shown that a uence rate of 1–15mWcm−2 within the SNpc
is needed to increase the number of dopaminergic cells (Angell-Petersen et al 2007). Although
the uence rate in the human SNpc delivered by the oral cavity was one order of magnitude
lower, the results indicate that a combination of PBM via the S.S. and oral cavity could be
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2934
benecial. It is conceivable that the tissue irradiance in the oral cavity can be set to a higher
level than in the S.S. without provoking hyperthermia due to a better choice of light spot loca-
tions and moisture cooling the tissue.
3.3.2. Illumination of the SNpc by a cylindrical light diffuser inserted in the third ventricle. One
remedy to the low efciency of transcranial illumination of the SNpc is the intracranial illu-
mination with a light source inserted in the third ventricle, i.e. in the vicinity of pons and
thalamus. Not only is the power load on the tissue located around the light source drastically
reduced, but the uence rate distribution can be chosen to be much more localized around the
SNpc in order to minimize possible negative side effects of the light on cellular mechanism
Figure 6. Figures 6(a) and (b) show, respectively, the illumination of the SNpc (anterior/
posterior location as white asterisks) by an FD1 (magenta arrow) in the oral cavity and a
cylindrical diffuser (magenta line) in the third ventricle. Colored contour lines represent
the log10 uence rate computed from the optimized set of optical tissue coefcients at
808nm. The tissue irradiance by each of the light sources was normalized to 1Wcm−2,
i.e. at the surface tissue—light cone or tissue—cylindrical light diffuser. (a) Fluence
rate distribution at 808nm for an FD1 (34.7 degree full aperture angle) in the oral
cavity. (b) Fluence rate distribution at 808nm for a cylindrical light diffuser (20mm
length and 1mm diameter) inserted in the third ventricle.
Table 3. The average uence rate in the SNpc is given for the different light delivery
routes and wavelengths.
Light delivery route
Average uence rate in the SNpc (mWcm−2)
671nm 808nm
Third ventricle 40 95
Sphenoidal sinus 0.3 3.6
Oral cavity 5×10−3 0.2
Note: Absolute values of uence rate are computed with the derived cadaver’s optical coefcients
for a tissue irradiance of 1Wcm−2, i.e. at the surface where the FD1 light cone enters the tissue
or the surface of the cylindrical light diffuser.
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2935
elsewhere. Moro et al (2013) are currently undertaking experiments to implant light sources
into the third ventricle of monkeys to show if PBM preserves behaviour and midbrain dopa-
minergic cells from MPTP toxicity.
We chose a cylindrical diffuser of 20mm length and 1mm diameter as light source to be
placed into the ventricle. The diffuser was modeled as a line of 21 equidistant isotropic point
sources (Jacques 1998), one every 1mm. The output power of the cylindrical diffuser was
chosen such that the tissue irradiance was 1Wcm−2. It is noteworthy that the simplied brain
mesh used for simulation did not contain the third ventricle due to constrains in the voxel size,
yet its location could be accurately identied on the original mesh from the MRI.
The uence rate distribution, shown in gure6(b) for the derived cadaver’s optical param-
eters at 808nm given in table2, indicates that a sufciently high uence rate can be achieved
within the SNpc with a low power output of the source. Comparing the values of uence rate
in table3 clearly shows the about 20 fold higher efciency of this light delivery route with
respect to any transcranial illumination approach.
4. Conclusions
The objective of the present study was to rene a possible treatment option for PD patients,
in a human cadaver model. We wanted to validate the practicalities of light delivery by illu-
minating the SNpc transcranially and transsphenoidally. We also wanted to validate the light
dosimetry by quantifying light distribution within the target brain tissue when illumination
takes place from the nasal cavity under endoscopic guidance. To this end, we measured u-
ence rate distribution in selected intracranial locations while delivering light at two different
wavelengths from a location outside the skull and accessible by an endoscope. Our study
demonstrates that this approach is possible.
We showed experimentally that the transsphenoidal delivery of light to the relevant target
structures of the brain is possible. Our measurements allowed us to validate the model by
directly measuring the uence rate within the target brain structures and to experimentally
determine the optical parameters of the brain tissues in this specimen. These will add to an
existing body of data and we hope they will be useful to rene published parameters that are
notoriously subject to large variations due to experimental conditions or to the specics of a
given specimen. Finally, our study allowed us to determine the light dose delivered to the deep
brain for different application routes, a result that could open a number of therapeutic options
for PD patients as well as for sufferers of other cerebral disease necessitating light therapy
delivered in a non-invasive way.
Acknowledgments
This work was supported in part by the CTI projects 13758.1 and 14660.1, the Swiss National
Science Foundation, project 205320_147141/1 and the J Jacobi grant.
The authors wish to thank Drs H Lewine and J Blue for anatomical and instrumentation
guidance and J Petersen, head of the cadaver lab facility at Acclarent (MenloPark, CA), for
providing any relevant help during the measurement campaign. We would also like to thank
the Brainlab staff P Patel and N Wright for the on-site support and F Vollmer and I Thiemann
for data preparation and analysis. The present work beneted from the input of Professor
H van den Bergh, who provided valuable comments to the writing and undertaking of the
research summarized here.
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2936
References
Angell-PetersenE, HirschbergH and MadsenS J 2007 Determination of uence rate and temperature
distributions in the rat brain; implications for photodynamic therapy J. Biomed. Opt. 12 014003
BaysR, Wagnières G, RobertD, Braichotte D, SavaryJ F, MonnierP and Van Bergh H 1997 Light
dosimetry for photodynamic therapy in the esophagus Lasers Surg. Med. 20 290303
BevilacquaF, PiguetD, MarquetP, Gross J D, TrombergB J and DepeursingeC 1999 In vivo local
determination of tissue optical properties: applications to human brain Appl. Opt. 38 493950
BurchmanM A 2011 Using photobiomodulation on a severe Parkinson’s patient to enable extractions,
root canal treatment and partial denture fabrication J. Laser Dent. 19 297–300
Cassidy J, Lilge L and Betz V 2013 Fullmonte: a framework for high-performance Monte Carlo
simulation of light through turbid media with complex geometry Proc. SPIE 8592 85920H
CustoA, WellsW M, BarnettA H, HillmanE M and BoasD A 2006 Effective scattering coefcient of
the cerebral spinal uid in adult head models for diffuse optical imaging Appl. Opt. 45 474755
DamierP, HirschE, AgidY and GraybielA 1999 The substantia nigra of the human brain: II. Patterns of
loss of dopamine-containing neurons in Parkinson’s disease Brain 122 143748
DavieC 2008 A review of Parkinson’s disease Br. Med. Bull. 86 10927
De TaboadaL, YuJ, El-AmouriS, Gattoni-CelliS, RichieriS, McCarthyT, StreeterJ and KindyM 2011
Transcranial laser therapy attenuates amyloid-beta peptide neuropathology in amyloid-beta protein
precursor transgenic mice J. Alzheimers Dis. 23 521–35
FangQ and BoasD A 2009a Monte carlo simulation of photon migration in 3d turbid media accelerated
by graphics processing units Opt. Express 17 2017890
FangQ and BoasD A 2009b Tetrahedral mesh generation from volumetric binary and gray-scale images
Proc. 6. IEEE Int. Conf. Symp. Biomed. Imag.: From Nano to Macro pp 1142–5
FangQ 2010 Mesh-based Monte Carlo method using fast raytracing in Plücker coordinates Biomed. Opt.
Express 1 16575
FirbankM, HiraokaM, EssenpreisM and DelpyD T 1993 Measurement of the optical properties of the
skull in the wavelength range 650–950nm Phys. Med. Biol. 38 50310
HamblinM and DemidovaT 2006 Mechanisms of low level light therapy Proc. SPIE vol 6140
HenyeyL G and GreensteinJ L 1941 Diffuse radiation in the Galaxy Astrophys. J. 93 7083
IshimaruA 1978 Wave Propagation and Scattering in Random Media (New York: Academic)
JacquesS L 1998 Light distributions from point, line and plane sources for photochemical reactions and
uorescence in turbid biological tissues Photochem. Photobiol. 67 2332
Karu T 1999 Primary and secondary mechanisms of action of visible to near-IR radiation on cells
J. Photochem. Photobiol. B: Biol. 49 17
KordowerJ, ChuY, HauserR, FreemanT and OlanowC 2008 Lewy body-like pathology in long-term
embryonic nigral transplants in Parkinson’s disease Nat. Med. 14 5046
KoyamaT, IwasakiA, OgoshiY and OkadaE 2005 Practical and adequate approach to modeling light
propagation in an adult head with low-scattering regions by use of diffusion theory Appl. Opt.
44 2094103
LapchakP A and Taboada L D 2010 Transcranial near infrared laser treatment (NILT) increases cortical
adenosine-5’-triphosphate (ATP) content following embolic strokes in rabbits Brain Res. 1306 1005
LapchakP 2012 Transcranial near-infrared laser therapy applied to promote clinical recovery in acute
and chronic neurodegenerative diseases Exp. Rev. Med. Dev. 9 7183
LeesA J, HardyJ and ReveszT 2009 Parkinson’s disease Lancet 373 205566
McNaughtK and Jenner P 2001 Proteasomal function is impaired in substantia nigra in Parkinson’s
disease Neurosci. Lett. 297 1914
MoraisV A and De StrooperB 2010 Mitochondria dysfunction and neurodegenerative disorders: cause
or consequence J. Alzheimers Dis. 20 S255–63
MoroC, Torres N, El Massri N, Ratel D, Johnstone D, Stone J, MitrofanisJ and Benabid A L 2013
Photobiomodulation preserves behaviour and midbrain dopaminergic cells from MPTP toxicity:
evidence from two mouse strains BMC Neurosci. 14 21942
PitzschkeA, LovisaB, SeydouxO, Haenggi M, OertelM F, ZellwegerM, TardyY and Wagnières G
2015 Optical properties of rabbit brain in the red and nir: changes observed under in-vivo, post-
mortem, frozen and formalin-xated conditions J. Biomed. Opt. 20 025006
RogganA 1997 Dosimetrie Thermischer Laseranwendungen in der Medizin 16 (ecomed)
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
2937
RogganA, SchädelD, NetzU, RitzJ P, GermerC T and MüllerG 1999 Effect of preparation technique
on the optical parameters of biological tissue Appl. Phys. B: Lasers Opt. 69 44553
SalomatinaE and YaroslavskyA 2008 Evaluation of the in vivo and ex vivo optical properties in a mouse
ear model Phys. Med. Biol. 53 2797–807
SamiiA, NuttJ G and RansomB R 2004 Parkinson’s disease Lancet 363 178393
SharmaS K, KharkwalG B, SajoM, HuangY Y, De TaboadaL, McCarthyT and HamblinM R 2011
Dose response effects of 810nm laser light on mouse primary cortical neurons Lasers Surg. Med.
43 8519
ShawV E, Peoples C, Spana S, AshkanK, BenabidA L, StoneJ, BakerG E and Mitrofanis J 2012
Patterns of cell activity in the subthalamic region associated with the neuroprotective action of
near-infrared light treatment in mptp-treated mice Park. Dis. 2012 296875
ShawV E, SpanaS, AshkanK, BenabidA L, StoneJ, BakerG E and MitrofanisJ 2010 Neuroprotection
of midbrain dopaminergic cells in mptp-treated mice after near-infrared light treatment J. Comp.
Neurol. 518 2540
TemelY, KesselsA, TanS, TopdagA, BoonP and Visser-VandewalleV 2006 Behavioural changes after
bilateral subthalamic stimulation in advanced Parkinson disease: a systematic review Park. Rel.
Disord. 12 26572
TrimmerP A, SchwartzK M, BorlandM K, De TaboadaL, StreeterJ and OronU 2009 Reduced axonal
transport in Parkinson’s disease cybrid neurites is restored by light therapy Mol. Neurodeg. 4 26
TrombergB J, SvaasandL O, FehrM K, MadsenS J, WyssP, SansoneB and TadirY 1996 A mathematical
model for light dosimetry in photodynamic destruction of human endometrium Phys. Med. Biol.
41 223
Tuchin V 2007 Tissue Optics: Light Scattering Methods and Instruments for Medical Diagnosis
vol PM166, 2nd edn (Bellingham, WA: SPIE Pr. Monograph) doi:10.1117/3.684093
VosM, LovisaB, GeensA, MoraisV A, WagnièresG, Van Den BerghH, GinggenA, De StrooperB,
Tardy Y and Verstreken P 2013 Near-Infrared 808 nm light boosts complex IV-dependent
respiration and rescues a Parkinson-related pink1 Model PLoS One 8 e78562
WangL, JacquesS L and Zheng L 1995 MCML: Monte Carlo modeling of light transport in multi-
layered tissues Comput Methods Programs Biomed. 47 13146
WangL and WuH 2007 Biomedical Optics: Principles and Imaging (New York: Wiley)
www.AcupunctureProducts.com 2007 Atlas of acupuncture points: point locations
YaroslavskyA N, SchulzeP C, YaroslavskyI V, SchoberR, UlrichF and SchwarzmaierH J 2002 Optical
properties of selected native and coagulated human brain tissues in vitro in the visible and near
infrared spectral range Phys. Med. Biol. 47 205973
ZhuC and Liu Q 2013 Review of Monte Carlo modeling of light transport in tissues J. Biomed. Opt.
18 050902
A Pitzschke et al
Phys. Med. Biol. 60 (2015) 2 921
... While the application of PBM to the body is a relatively easy task, application to the nervous system, especially the brain, is more challenging. Some success has been achieved when applied through a sinus [13] or via a transcranial approach [14]. PBM wavelengths can penetrate up to 5 cm depending on the selected wavelength [15][16][17][18]. ...
... They discovered that weekly treatment for 2 months improved cognitive capacity and spatial learning that was accompanied by a 68% reduction in Aβ load in the hippocampus [28]. Pitzschke and colleagues compared the penetration of 671 nm and 810 nm light into cadaver brain that was delivered either via a transcranial or a transsphenoidal approach [13]. Results revealed that the best combination for delivery to the brain was 810 nm NIR light administered via the transsphenoidal route [13]. ...
... Pitzschke and colleagues compared the penetration of 671 nm and 810 nm light into cadaver brain that was delivered either via a transcranial or a transsphenoidal approach [13]. Results revealed that the best combination for delivery to the brain was 810 nm NIR light administered via the transsphenoidal route [13]. Intranasal PBM has received wide attention from those concerned with diseases and therapeutics affecting the brain and has been recently reviewed by Salehpour and colleagues [60]. ...
Article
Full-text available
Alzheimer’s disease is a growing global crisis in need of urgent diagnostic and therapeutic strategies. The current treatment strategy mostly involves immunotherapeutic medications that have had little success in halting disease progress. Hypotheses for pathogenesis and development of AD have been expanded to implicate both organ systems as well as cellular reactions. Non-pharmacologic interventions ranging from minimally to deeply invasive have attempted to address these diverse contributors to AD. In this review, we aim to delineate mechanisms underlying such interventions while attempting to provide explanatory links between the observed differences in disease states and postulated metabolic or structural mechanisms of change. The techniques discussed are not an exhaustive list of non-pharmacological interventions against AD but provide a foundation to facilitate a deeper understanding of the area of study.
... These variables can add endless combinations of parameters to the field where the same disease can be approached with various wavelengths of light, adjusted power of emission, or pulse rate for delivery to a target tissue, and of course, can add variables with treatment protocol like duration of each session, a number of applications, duration of therapy and so on creating confusion in the field and total disarray. That is where dosimetry can be very helpful in standardizing different approaches to common ground and influencing the order of things to get to the goal of the treatment outcome (Pitzschke et al., 2015). The purpose of dosimetry is to help deliver a particular amount of light stimulation per area of the target tissue to trigger the same signaling cascade or cell receptors activation to initiate a beneficial cascade of effects in the target tissue despite the difference in physical characteristics of a PBM light. ...
... The adjustments can be made to light stimulation duration time or light emission power which can be reflected in similar outcomes across the board. This is where standardization of treatment across the entire field can become a reality for investigators with different approaches to achieve the same result on a known scale of quantitative light measurement for the light wavelength (Pitzschke et al., 2015;Khan and Arany, 2016). ...
Article
Full-text available
Light is a natural agent consisting of a range of visible and invisible electromagnetic spectrum travels in waves. Near-infrared (NIR) light refers to wavelengths from 800 to 2,500 nm. It is an invisible spectrum to naked eyes and can penetrate through soft and hard tissues into deep structures of the human body at specific wavelengths. NIR light may carry different energy levels depending on the intensity of emitted light and therapeutic spectrum (wavelength). Stimulation with NIR light can activate intracellular cascades of biochemical reactions with local short- and long-term positive effects. These properties of NIR light are employed in photobiomodulation (PBM) therapy, have been linked to treating several brain pathologies, and are attracting more scientific attention in biomedicine. Transcranial brain stimulations with NIR light PBM in recent animal and human studies revealed a positive impact of treatment on the progression and improvement of neurodegenerative processes, management of brain energy metabolism, and regulation of chronic brain inflammation associated with various conditions, including traumatic brain injury. This scientific overview incorporates the most recent cellular and functional findings in PBM with NIR light in treating neurodegenerative diseases, presents the discussion of the proposed mechanisms of action, and describes the benefits of this treatment in neuroprotection, cell preservation/detoxification, anti-inflammatory properties, and regulation of brain energy metabolism. This review will also discuss the novel aspects and pathophysiological role of the glymphatic and brain lymphatics system in treating neurodegenerative diseases with NIR light stimulations. Scientific evidence presented in this overview will support a combined effort in the scientific community to increase attention to the understudied NIR light area of research as a natural agent in the treatment of neurodegenerative diseases to promote more research and raise awareness of PBM in the treatment of brain disorders.
... This pulse rate also correlates with electroencephalogram (EEG) alpha brain wave "entrainment" (41) and cellular light absorption (42). Additionally, the 810 nm wavelength has shown several benefits for mental applications compared to other wavelengths used in PBM, such as 633 nm, 655 nm, etc. (43). Finally, the question of whether or not low-level NIR light can penetrate the skull has been studied in cadaver experiments (43)(44)(45) and studies with ex vivo skulls (46), as well as in silico using Monte Carlo (MC) simulations (47,48). ...
... Additionally, the 810 nm wavelength has shown several benefits for mental applications compared to other wavelengths used in PBM, such as 633 nm, 655 nm, etc. (43). Finally, the question of whether or not low-level NIR light can penetrate the skull has been studied in cadaver experiments (43)(44)(45) and studies with ex vivo skulls (46), as well as in silico using Monte Carlo (MC) simulations (47,48). Together, these studies suggest that a low energy 810 nm light source pulsed at a frequency of 10 Hz can penetrate the skull and could have a potential positive effect on mental health. ...
Article
Full-text available
We report the results of experimental investigations involving photobiomodulation (PBM) of living cells, tubulin, and microtubules in buffer solutions exposed to near-infrared (NIR) light emitted from an 810 nm LED with a power density of 25 mW/cm2 pulsed at a frequency of 10 Hz.
... The putative depth that can be reliably imaged (especially with commercial devices) is estimated to be in the 2-3 cm range (below the scalp surface) on average [45][46][47]. Further depths of 4-5 cm have been reported using higher wavelengths (808 nm or more) [48][49][50]. Due to its more advanced signal properties and sophisticated software interface, DOT is able to more consistently produce images in the 4-5 cm depth range, compared to fNIRS [25,38,51]. Additionally, recent experiments using timeor frequency-domain detection modes appear to mitigate some of the depth sensitivity limitations encountered with the standard continuous-wave design [52]. ...
Article
Full-text available
Transcranial magnetic stimulation (TMS) has been established as an important and effective treatment for various psychiatric disorders. However, its effectiveness has likely been limited due to the dearth of neuronavigational tools for targeting purposes, unclear ideal stimulation parameters, and a lack of knowledge regarding the physiological response of the brain to TMS in each psychiatric condition. Modern optical imaging modalities, such as functional near-infrared spectroscopy and diffuse optical tomography, are promising tools for the study of TMS optimization and functional targeting in psychiatric disorders. They possess a unique combination of high spatial and temporal resolutions, portability, real-time capability, and relatively low costs. In this mini-review, we discuss the advent of optical imaging techniques and their innovative use in several psychiatric conditions including depression, panic disorder, phobias, and eating disorders. With further investment and research in the development of these optical imaging approaches, their potential will be paramount for the advancement of TMS treatment protocols in psychiatry.
... This review does not comprise the available evidence for applications of PBM in chronic TBI, though a review summarising this data is available in the literature 30 Transcranial approaches to PBM have a clear benefit in animal models, as illustrated here, but PBM delivery via this route is inherently impeded by the thickness of scalp and skull when translating this approach to humans. Thorough clarification of the effective dose window will be invaluable, alongside computational simulation, in determining whether this approach is the optimal route for clinical practice, or whether more elaborate or novel approaches may improve delivery and functional outcomes [90][91][92][93][94][95] . ...
Article
Full-text available
Photobiomodulation (PBM) is a therapeutic modality which has gained increasing interest in neuroscience applications, including acute traumatic brain injury (TBI). Its proposed mechanisms for therapeutic effect when delivered to the injured brain include anti-apoptotic and anti-inflammatory effects. This systematic review summarises the available evidence for the value of PBM in improving outcomes in acute TBI and presents a meta-analysis of the pre-clinical evidence for neurological severity score (NSS) and lesion size in animal models of TBI. A systematic review of the literature was performed, with searches and data extraction performed independently in duplicate by two authors. Eighteen published articles were identified for inclusion: seventeen pre-clinical studies of in vivo animal models; and one clinical study in human patients. The available human study supports safety and feasibility of PBM in acute moderate TBI. For pre-clinical studies, meta-analysis for NSS and lesion size were found to favour intervention versus control. Sub-group analysis based on PBM parameter variables for these outcomes was performed. Favourable parameters were identified as: wavelengths in the region of 665 nm and 810 nm; time to first administration of PBM ≤ 4 hours; total number of daily treatments ≤3. No differences were identified between pulsed and continuous wave modes or energy delivery. Mechanistic sub-studies within included in vivo studies are presented and were found to support hypotheses of anti-apoptotic, anti-inflammatory and pro-proliferative effects, and a modulation of cellular metabolism. This systematic review provides substantial meta-analysis evidence of the benefits of PBM on functional and histological outcomes of TBI in in vivo mammalian models. Consideration of study design and PBM parameters should be closely considered for future human clinical studies.
... Specifically, Transcranial LED therapy (TCLT) defines the limited application of LED therapy to the brain. The LED light travels through the layers of the scalp and skull to reach brain cells [10][11][12]. The brain is commonly irradiated with red (RL) or near-infrared (NIR) light (600-1100 nm), with a total output power of 1-10,000 mW, a power density that has no thermal effects [9]. ...
Chapter
Full-text available
The hippocampus is an integral portion of the limbic system and executes a critical role in spatial and recognition learning, memory encoding, and memory consolidation. Hippocampal aging showed neurobiological alterations, including increased oxidative stress, altered intracellular signaling pathways, synaptic impairment, and organelle deterioration such as mitochondrial dysfunction. These alterations lead to hippocampal cognitive decline during aging. Therefore, the search for new non-invasive therapies focused on preserving or attenuating age-related hippocampal memory impairment could have of great impact on aging, considering the increasing life expectancy in the world. Red light Transcranial LED therapy (RL-TCLT) is a promising but little explored strategy, which involves red light LED irradiation without surgical procedures, safe and at a low cost. Nevertheless, the precise mechanism involved and its real impact on age-related cognitive impairment is unclear, due to differences in protocol, wavelength applied, and time. Therefore, in this chapter, we will discuss the evidence about RL-TCLT and its effects on the hippocampal structure and function, and how this therapy could be used as a promising treatment for memory loss during aging and in age-related diseases such as Alzheimer’s Disease (AD). Finally, we will mention our advances in Red 630-light-Transcranial LED therapy on the hippocampus in aging and AD.
... Since our study clearly demonstrates that PBM stimulates the endogenous production of PpIX in glioblastoma cells, a logical extrapolation of our finding consists to study the PBM potential to treat other cell types, in which the PBM mechanisms are likely to follow other pathways, that are at the origin of various conditions affecting the brain. Cells involved in neurodegenerative conditions such as Parkinson's, Huntington's, and Alzheimer's diseases, in particular dopaminergic neurons, have very good chances to respond positively to PBM providing that optimal and controlled radiometric and spectral parameters are used [56]. Such an optimization study of PBM, possibly conducted while integrating the combined use of already approved exogenous drugs to treat neurodegenerative conditions, looks promising in terms of clinical impact in the near future. ...
Article
Protoporphyrin IX (PpIX) is produced in the mitochondria and used as fluorescent contrast agent or photosensitizer after exogenous 5-aminolevulinic acid (ALA) delivery in cancer photodynamic detection and therapy (PDT). Although routinely used in the clinics, the stimulated production of PpIX is often insufficient and/or heterogeneous within the lesions, thereby limiting the PDT performances. Since photobiomodulation, which is based on the illumination of the tissues with sub-thermal radiometric conditions in the red or near-infrared, is known to stimulate the cell metabolism, we have optimized these conditions in vitro. Some of them lead to the homogenization and strong stimulation of the PpIX endogenous production. Interestingly, combined sequentially, PBM enhanced significantly the potency of PpIX-based PDT in vitro and in vivo in tumors grown on the chicken embryo chorioallantoic membrane. These results are in excellent agreement with other assays based on measurements of the cell survival/death, the production of reactive oxygen species, including singlet oxygen, and the mitochondrial membrane potential.
... Phototherapeutics that respond to wavelengths within the optical window of tissue (600-900 nm) (Fig. 1) are a relatively recent innovation, with potentially profound clinical ramifications. For example, studies have demonstrated that 800-nm photons display a brain penetrance (through the skull) of up to 3 cm (reFs [31][32][33][34][35][36][37][38] ), thereby, exemplifying potential biomedical opportunities. We note that a large number of reviews have appeared on the general design and properties of photoactivatable compounds, and a few representative comprehensive articles are cited here 7,[39][40][41][42][43][44][45][46][47][48] and in appropriate places below. ...
Article
More than four decades have passed since the first example of a light-activated (caged) compound was described. In the intervening years, a large number of light-responsive derivatives have been reported, several of which have found utility under a variety of in vitro conditions using cells and tissues. Light-triggered bioactivity furnishes spatial and temporal control, and offers the possibility of precision dosing and orthogonal communication with different biomolecules. These inherent attributes of light have been advocated as advantageous for the delivery and/or activation of drugs at diseased sites for a variety of indications. However, the tissue penetrance of light is profoundly wavelength-dependent. Only recently have phototherapeutics that are photoresponsive in the optical window of tissue (600–900 nm) been described. This Review highlights these recent discoveries, along with their limitations and clinical opportunities. In addition, we describe preliminary in vivo studies of prospective phototherapeutics, with an emphasis on the path that remains to be navigated in order to translate light-activated drugs into clinically useful therapeutics. Finally, the unique attributes of phototherapeutics is highlighted by discussing several potential disease applications.
Chapter
Photobiomodulation (PBM) is the use of near-infrared (NIR) or red light (600–1100 nm) to positively affect brain function. PBM treatment of healthy human young and elderly volunteers has demonstrated improvements in mood, cognitive function, and memory, with increases in regional cerebral blood flow and tissue oxygenation. Improvements in sleep, executive function, and working memory have been reported in patients with chronic traumatic brain injury (TBI).
Article
Glioblastoma Multiforme (GBM) is a multifaceted and complex disease, which has experienced no changes in treatment for nearly two decades and has a 5-year survival rate of only 5.4%. Alongside challenges in delivering chemotherapeutic agents across the blood brain barrier (BBB) to the tumour, the immune microenvironment is also heavily influenced by tumour signalling. Immunosuppression is a major aspect of GBM; however, evidence remains conflicted as to whether pro-inflammatory or anti-inflammatory therapies are the key to improving GBM treatment. To address both of these issues, particle delivery systems can be designed to overcome BBB transport while delivering a wide variety of immune-stimulatory molecules to investigate their effect on GBM. This review explores literature from the past 3 years that combines particle delivery systems alongside immunotherapy for the effective treatment of GBM.
Article
Full-text available
The outcome of light-based therapeutic approaches depends on light propagation in biological tissues, which is governed by their optical properties. The objective of this study was to quantify optical properties of brain tissue in vivo and postmortem and assess changes due to tissue handling postmortem. The study was carried out on eight female New Zealand white rabbits. The local fluence rate was measured in the VIS/NIR range in the brain in vivo, just postmortem, and after six weeks’ storage of the head at −20°C or in 10% formaldehyde solution. Only minimal changes in the effective attenuation coefficient μeff were observed for two methods of sacrifice, exsanguination or injection of KCl. Under all tissue conditions, μeff decreased with increasing wavelengths. After long-term storage for six weeks at −20°C, μeff decreased, on average, by 15 to 25% at all wavelengths, while it increased by 5 to 15% at all wavelengths after storage in formaldehyde. We demonstrated that μeff was not very sensitive to the method of animal sacrifice, that tissue freezing significantly altered tissue optical properties, and that formalin fixation might affect the tissue’s optical properties.
Article
Full-text available
The use of low levels of visible or near infrared light for reducing pain, inflammation and edema, promoting healing of wounds, deeper tissues and nerves, and preventing tissue damage has been known for almost forty years since the invention of lasers. Originally thought to be a peculiar property of laser light (soft or cold lasers), the subject has now broadened to include photobiomodulation and photobiostimulation using non-coherent light. Despite many reports of positive findings from experiments conducted in vitro, in animal models and in randomized controlled clinical trials, LLLT remains controversial. This likely is due to two main reasons; firstly the biochemical mechanisms underlying the positive effects are incompletely understood, and secondly the complexity of rationally choosing amongst a large number of illumination parameters such as wavelength, fluence, power density, pulse structure and treatment timing has led to the publication of a number of negative studies as well as many positive ones. In particular a biphasic dose response has been frequently observed where low levels of light have a much better effect than higher levels. This introductory review will cover some of the proposed cellular chromophores responsible for the effect of visible light on mammalian cells, including cytochrome c oxidase (with absorption peaks in the near infrared) and photoactive porphyrins. Mitochondria are thought to be a likely site for the initial effects of light, leading to increased ATP production, modulation of reactive oxygen species and induction of transcription factors. These effects in turn lead to increased cell proliferation and migration (particularly by fibroblasts), modulation in levels of cytokines, growth factors and inflammatory mediators, and increased tissue oxygenation. The results of these biochemical and cellular changes in animals and patients include such benefits as increased healing in chronic wounds, improvements in sports injuries and carpal tunnel syndrome, pain reduction in arthritis and neuropathies, and amelioration of damage after heart attacks, stroke, nerve injury and retinal toxicity.
Article
Full-text available
Emerging clinical applications including bioluminescence imaging require fast and accurate modelling of light propagation through turbid media with complex geometries. Monte Carlo simulations are widely recognized as the standard for high-quality modelling of light propagation in turbid media, albeit with high computational requirements. We present FullMonte: a flexible, extensible software framework for Monte Carlo modelling of light transport from extended sources through general 3D turbid media including anisotropic scattering and refractive index changes. The problem geometry is expressed using a tetrahedral mesh, giving accurate surface normals and avoiding artifacts introduced by voxel approaches. The software uses multithreading, Intel SSE vector instructions, and optimized data structures. It incorporates novel hardware-friendly performance optimizations that are also useful for software implementations. Results and performance are compared against existing implementations. We present a discussion of current state-of-the-art algorithms and accelerated implementations of the modelling problem. A new parameter permitting accuracy-performance tradeoffs is also shown which has significant implications including performance gains of over 25% for real applications. The advantages and limitations of both CPU and GPU implementations are discussed, with observations important to future advances. We also point the way towards custom hardware implementations with potentially large gains in performance and energy efficiency.
Article
Full-text available
Mitochondrial electron transport chain (ETC) defects are observed in Parkinson's disease (PD) patients and in PD fly- and mouse-models; however it remains to be tested if acute improvement of ETC function alleviates PD-relevant defects. We tested the hypothesis that 808 nm infrared light that effectively penetrates tissues rescues pink1 mutants. We show that irradiating isolated fly or mouse mitochondria with 808 nm light that is absorbed by ETC-Complex IV acutely improves Complex IV-dependent oxygen consumption and ATP production, a feature that is wavelength-specific. Irradiating Drosophila pink1 mutants using a single dose of 808 nm light results in a rescue of major systemic and mitochondrial defects. Time-course experiments indicate mitochondrial membrane potential defects are rescued prior to mitochondrial morphological defects, also in dopaminergic neurons, suggesting mitochondrial functional defects precede mitochondrial swelling. Thus, our data indicate that improvement of mitochondrial function using infrared light stimulation is a viable strategy to alleviate pink1-related defects.
Article
Full-text available
Background We have shown previously that near-infrared light (NIr) treatment or photobiomodulation neuroprotects dopaminergic cells in substantia nigra pars compacta (SNc) from degeneration induced by 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) in Balb/c albino mice, a well-known model for Parkinson’s disease. The present study explores whether NIr treatment offers neuroprotection to these cells in C57BL/6 pigmented mice. In addition, we examine whether NIr influences behavioural activity in both strains after MPTP treatment. We tested for various locomotive parameters in an open-field test, namely velocity, high mobility and immobility. Results Balb/c (albino) and C57BL/6 (pigmented) mice received injections of MPTP (total of 50 mg/kg) or saline and NIr treatments (or not) over 48 hours. After each injection and/or NIr treatment, the locomotor activity of the mice was tested. After six days survival, brains were processed for TH (tyrosine hydroxylase) immunochemistry and the number of TH+ cells in the substantia nigra pars compacta (SNc) was estimated using stereology. Results showed higher numbers of TH+ cells in the MPTP-NIr groups of both strains, compared to the MPTP groups, with the protection greater in the Balb/c mice (30% vs 20%). The behavioural tests revealed strain differences also. For Balb/c mice, the MPTP-NIr group showed greater preservation of locomotor activity than the MPTP group. Behavioural preservation was less evident in the C57BL/6 strain however, with little effect of NIr being recorded in the MPTP-treated cases of this strain. Finally, there were differences between the two strains in terms of NIr penetration across the skin and fur. Our measurements indicated that NIr penetration was considerably less in the pigmented C57BL/6, compared to the albino Balb/c mice. Conclusions In summary, our results revealed the neuroprotective benefits of NIr treatment after parkinsonian insult at both cellular and behavioural levels and suggest that Balb/c strain, due to greater penetration of NIr through skin and fur, provides a clearer model of protection than the C57BL/6 strain.
Article
To achieve accuracy in studying the patterns of loss of midbrain dopamine-containing neurons in Parkinson's disease, we used compartmental patterns of calbindin D 28K immunostaining to subdivide the substantia nigra with landmarks independent of the degenerative process. Within the substantia nigra pars compacta, we identified dopamine-containing neurons in the calbindin-rich regions ('matrix') and in five calbindin-poor pockets ('nigrosomes') defined by analysis of the three-dimensional networks formed by the calbindin-poor zones. These zones were recognizable in all of the brains, despite severe loss of dopamine-containing neurons. The degree of loss of dopamine-containing neurons in the substantia nigra pars compacta was related to the duration of the disease, and the cell loss followed a strict order. The degree of neuronal loss was significantly higher in the nigrosomes than in the matrix. Depletion was maximum (98%) in the main pocket (nigrosome 1), located in the caudal and mediolateral part of the substantia nigra pars compacta. Progressively less cell loss was detectable in more medial and more rostral nigrosomes, following the stereotyped order of nigrosome 1 > nigrosome 2 > nigrosome 4 > nigrosome 3 > nigrosome 5. A parallel, but lesser, caudorostral gradient of cell loss was observed for dopamine-containing neurons included in the matrix. This pattern of neuronal loss was consistent from one parkinsonian substantia nigra pars compacta to another. The spatiotemporal progression of neuronal loss related to disease duration can thus be drawn in the substantia nigra pars compacta for each Parkinson's disease patient: depletion begins in the main pocket (nigrosome 1) and then spreads to other nigrosomes and the matrix along rostral, medial and dorsal axes of progression.
Conference Paper
We report a general purpose mesh genretor for creating finite-element surface or volumetric mesh from 3D binary or gray-scale medical images. This toolbox incorporates a number of exisiting free mesh processing utilities and enables researchers to perform a range of mesh processing tasks for image-based mesh generation, including raw image processing, surface mesh extraction, surface re-sampling, and multi-scale/adaptive tetrahedral mesh generation open-surfaces and sub-region labeling. Atomic meshin utilities for each processing step can be accessed with simple interfaces, which can be streamlined or executed independently. The toolbox is compatible with Matlab or GNU Octave. We demonstrate the applications of this toolbox for meshing a range of challenging geometrics including complex vessel network, human brain and breast.
Article
A general survey is provided on the capability of Monte Carlo (MC) modeling in tissue optics while paying special attention to the recent progress in the development of methods for speeding up MC simulations. The principles of MC modeling for the simulation of light transport in tissues, which includes the general procedure of tracking an individual photon packet, common light-tissue interactions that can be simulated, frequently used tissue models, common contact/noncontact illumination and detection setups, and the treatment of time-resolved and frequency-domain optical measurements, are briefly described to help interested readers achieve a quick start. Following that, a variety of methods for speeding up MC simulations, which includes scaling methods, perturbation methods, hybrid methods, variance reduction techniques, parallel computation, and special methods for fluorescence simulations, as well as their respective advantages and disadvantages are discussed. Then the applications of MC methods in tissue optics, laser Doppler flowmetry, photodynamic therapy, optical coherence tomography, and diffuse optical tomography are briefly surveyed. Finally, the potential directions for the future development of the MC method in tissue optics are discussed.