Frequency-domain fluorescent diffusion tomography:

a finite-element-based algorithm and simulations

Huabei Jiang

We present a finite-element-based algorithm for reconstruction of fluorescence lifetime and yield in turbid
media, using frequency-domain data. The algorithm is based on a set of coupled diffusion equations that
describe the propagation of both excitation and fluorescent emission light in multiply scattering media.
Centered on Newton’s iterative method, we implemented our algorithm by using a synthesized scheme
of Marquardt and Tikhonov regularizations. A low-pass spatial filter is also incorporated into the
algorithm for enhancing image reconstruction. Simulation studies using both noise-free and noisy data
have been performed with the nonzero photon density boundary conditions. Our results suggest that
quantitative images can be produced in terms of fluorescent lifetime and yield values and location, size,
and shape of heterogeneities within a circular background region. © 1998 Optical Society of America
OCIS codes: 170.0170, 170.3010, 170.6960, 170.3650, 170.3660, 170.3880.

1. Introduction experiments indicate that this endogenous optical

Nonionizing, noninvasive optical imaging has great
potential for becoming one of the most promising al-
ternative strategies for breast cancer detection.
This is underscored by the fact that in the last few
years significant progress has been made in under-
standing the fundamental nature of light propaga-
tion in tissue,1 which has led to the emergence of
several advanced techniques for transmission of light
through tissue over distances relevant to breast
imaging.2–10 Considerable effort has been made to-
ward the realization of a new class of breast cancer
detection methods in which these emerging optical
techniques are used.
Both direct imaging and image reconstructions in
which multiply scattered light is used have been dem-
onstrated, largely from simulations and tissue phan-
tom experiments. The majority of these optical
imaging techniques employ methodologies that are
based on differences in the absorption andyor scat-
tering properties between the embedded object and
its surroundings. While in vitro studies show that
the endogenous optical contrast between the tumor
and the normal tissues can be quite low,11,12 in vivo
The author is with the Department of Physics and Astronomy,
Clemson University, Clemson, South Carolina 29634-1911.
Received 5 January 1998; revised manuscript received 20 April
1998.
0003-6935y98y225337-07$15.00y0
© 1998 Optical Society of America
contrast can be as high as fourfold.13 Nonetheless
development of methods for improving both the sen-
sitivity and the specificity for detection is necessary.
Fluorescence spectroscopyyimaging studies of hu-
man tissue suggest that a variety of lesions show
distinct fluorescence spectra compared with those of
normal tissue.14 –17 It has been evidenced that fluo-
rescence spectroscopy of endogenous compounds can
provide enhanced contrast as well as diagnostic in-
formation.18 It has also been shown that exogenous
dyes may offer the best contrast for optical im-
aging.19 –21 Use of exogenous agents would pro-
vide fluorescent markers that could serve to detect
embedded tumors in the breast. In particular, the
ability to monitor the fluorescent yield and lifetime
may also provide biochemical specificity if the fluoro-
phore is sensitive to a specific metabolite, such as
oxygen or glucose.22 This has intrinsic merit that
extends beyond the structural information contained
in the static images associated with x-ray mammog-
raphy, thus providing another strong rationale for its
complementary role in the detectionydiagnosis of
breast cancer. Because tumors exhibit distinctive
metabolite alternations, this ability to extract tissue
functional information may offer a unique way to
noninvasively differentiate between benign and ma-
lignant tumors.
Although current fluorescence imagingyspectro-
scopic techniques may provide sensitive optical sig-
nals that are indicative of disease, they often lose
their specificity to the tissues of interest. For exam-

1 August 1998 y Vol. 37, No. 22 y APPLIED OPTICS 5337

ple, endogenous fluorescence has been used to distin-
guish between diseased and normal tissues.15,17
However, the correlation between the diseased tis-
sues and measured optical signals has been weak-
ened by background light absorption and scattering,
which are nonspecific to disease. Hence the speci-
ficity and the sensitivity for detection might be im-
proved if these background signal contributions could
be eliminated from the fluorescent, endoscopic mea-
surement. Toward this end, fluorescence lifetime
and yield measurements in tissues demonstrate flu-
orescence lifetime imagesyspectra that are indepen-
dent of tissue absorption and scattering in vivo.
Although a number of groups have begun to explore
the methods needed to extract fluorophore lifetimes
from deep within tissue,23–29 the development of such
techniques has been hampered since absorption and
scattering distort both the intensity and the lifetime
kinetics of reemitted fluorescent light. This distor-
tion or convolution is complex in nature because the
optical property distribution can be quite nonuniform
and in general is not known a priori. This complex-
ity necessitates the use of detailed analysis of the
multiply scattered excitation and fluorescent light to
reconstruct the lifetime images in tissues.
Studies in fluorescent reconstructions in multiply
scattering media or tissues have just begun. Several
groups have recently reported their simulated results
in this regard, using a simple integral method,30
perturbation methods,31,32 and a finite difference al-
gorithm.33 In this paper we report a finite-element-
based reconstruction algorithm for imaging both
fluorescence lifetime and yield in thick tissues, us-
ing frequency-domain data. The algorithm is con-
structed under the same framework of our algorithms
for optical imaging in which excitation data are
used.34,35 The finite-element method has proved to
be powerful in studies of both forward and inverse
photon migration problems by our own research and
others.10,34 – 42 Based on a set of coupled diffusion
equations that describe the propagation of both exci-
tation and fluorescent emission light in multiply scat-
tering media, the algorithm is centered on Newton’s
iterative method where a synthesized scheme of Mar-
quardt and Tikhonov regularizations is used. A low-
pass spatial filter is also incorporated into the
algorithm to enhance image reconstruction. Simu-
lation studies in which both noise-free and noisy ~to
as much as 5% noise! data are used have been per-
formed with the nonzero photon density boundary
conditions. Images reconstructed are quantitative
in terms of fluorescent lifetime and yield values and
location, size, and shape of heterogeneities within a
circular background region.
2. Reconstruction Algorithm
In the frequency domain it is known that propagation
of both excitation and fluorescent emission light in
5338 APPLIED OPTICS y Vol. 37, No. 22 y 1 August 1998
tissues or multiply scattering media can be described
by the following coupled diffusion equations:23,33
F G
iv
¹ z @Dx~r!¹Fx~r, v!# 2 max~r! 2 Fx~r, v!
c
5 2S~r, v!, (1)
F G
iv
¹ z @Dm~r!¹Fm~r, v!# 2 mam~r! 2 Fm~r, v!
c
1 1 ivt~r!
5 2h~r!max3mFx~r, v! , (2)
1 1 v2t~r!2
where Fx,m is the photon density for excitation ~sub-
script x! or fluorescent light ~subscript m!, Dx,m is the
diffusion coefficient, max,m is the absorption coefficient
due to contributions from both nonfluorescing chro-
mophores and fluorescent dye, max3m is the absorption
coefficient for the excitation light due to the contri-
bution from fluorescent dye, v is the modulation fre-
quency, c is the velocity of light in the medium, and h
and t are the fluorescent quantum yield and lifetime.
S~r, v! is the excitation source term in Eq. ~1!, which
for a point source can be written as S 5 S0d~r 2 r0!,
where S0 is the source strength and d~r 2 r0! is the
Dirac-delta function for a source at r0. Note that a
single-exponential fluorescence decay has been as-
sumed in the source term for fluorescent light @right-
hand side of Eq. ~2!#; multiexponential time decay can
be handled by a simple extension. The diffusion co-
efficient can be written as
1
Dx,m~r! 5 , (3)
3@max,m~r! 1 msx,m9~r!#
where msx,m9~r! is the reduced scattering coefficient.
For known optical properties and fluorescent life-
time and yield, Eqs. ~1! and ~2! become standard
boundary value problems for the spatially varying
photon densities of excitation and emission light sub-
ject to appropriate boundary conditions ~BC’s!. In
this paper we use the nonzero photon density BC’s
or type III BC’s, which have proved to be most accu-
rate:35,43 2Dx,m¹Fx,m z n̂ 5 aFx,m, where n̂ is the unit
normal vector for the boundary surface and a is a
coefficient that is related to the internal reflection at
the boundary.
In general, the purpose of fluorescent diffusion to-
mography is to recover all distributions including
Dx,m, max,m, t, and h. We focus on the reconstruction
of t and h distributions in this study. Reconstruc-
tions of other parameters can be easily included in
the algorithm. Thus, making use of finite-element
discretizations of Eqs. ~1! and ~2!, we can obtain two
matrix equations in terms of a discrete set of spatially
distributed fluorescent lifetime and yield and photon
density values:
@Ax#$Fx% 5 $bx%, (4)
@Am#$Fm% 5 $bm%. (5)
Following the procedures outlined in Refs. 34 and 35, that accomplish this: Taylor expansion and least-
one finds that the elements of matrix @Ax# and @Am# squares minimization. These two methods have
are, respectively, proved to be identical.44 We use a Taylor expansion

K S D L
here; i.e., we Taylor expand both the real and the
iv imaginary parts of Fm about an assumed ~t, h! dis-
~ax!ij 5 2Dx¹cj z ¹ci 2 max 2 cj ci , (6)
c tribution that is a perturbation away from some other
distribution ~t̃, h̃! so that a discrete set of photon

K
~am!ij 5 2Dm¹cj z ¹ci 2 mam 2 S iv
c
D L
cj ci , (7)
density values can be expressed as

]Fm~R! ]Fm~R!
and the entries in column vectors $bx,m% and $Fx,m% are Fm~R!~t̃, h̃! 5 Fm~R!~t, h! 1 Dt 1 Dh 1 · · ·,
]t ]h
M (11)
~bx!i 5 2^Sci & 1 a ( ~F ! r c c ds,
x j j i (8)
]Fm~I! ]Fm~I!
Fm~I!~t̃, h̃! 5 Fm~I!~t, h! 1
j51
Dt 1 Dh 1 · · ·,
]t ]h

7 D8
L

K N 1 2 iv ( tc l l (12)
( (
S(
~bm!i 5 2 hkckmax3m ~Fx!j cj ci l51
L 2 where Dt 5 t̃ 2 t and Dh 5 h̃ 2 h. Fm~R! and Fm~I!
k51 j51
11v 2
tlcl are the real and the imaginary parts of Fm. If the
l51
assumed lifetime and yield distributions are close to
M
the true profiles, the left-hand side of Eqs. ~11! and
1a ( ~F
j51
! r cj cids,
m j (9)
~12! can be considered as true data ~either imposed or
observed! and the relationship truncated to produce
Fx,m 5 $~Fx,m!1, ~Fx,m!2, · · · ,~Fx,m!N%T, (10)
)Dx 5 Fmo 2 Fmc, (13)
where ^ & indicates integration over the problem do-
main and Fx,m, t, and h have been expanded as the where ) is the Jacobian matrix consisting of the de-
sum of coefficients multiplied by a set of locally spa- rivatives of Fm with respect to t or h at each boundary
tially varying Lagrangian basis functions cj , cl, and observation node34,35; Dx is the vector that expresses
ck. r expresses integration over the boundary sur- perturbations of t and h; and Fmo and Fmc are the
face where type III BC’s have been applied. ~Fx,m!i observed and the computed fluorescent photon den-
is the photon density at node i, N is the node number sity at the boundary. All these matrix and vectors
of a finite-element mesh, and M is the boundary node can be written as

3 4
]Fm,1~R! ]Fm,1~R! ]Fm,1~R! ]Fm,1~R! ]Fm,1~R! ]Fm,1~R!
··· ···
]t1 ]t2 ]tL ]h1 ]h2 ]hK
]Fm,1~I! ]Fm,1~I! ]Fm,1~I! ]Fm,1~I! ]Fm,1~I! ]Fm,1~I!
··· ···
]t1 ]t2 ]tL ]h1 ]h2 ]hK
· · ·· · · · ·· ·
)5 ·
·
·
· · ·
·
·
·
·
· · ·
· , (14)
]Fm,M~R! ]Fm,M~R! ]Fm,M~R! ]Fm,M~R! ]Fm,M~R! ]Fm,M~R!
··· ···
]t1 ]t2 ]tL ]h1 ]h2 ]hK
]Fm,M~I! ]Fm,M~I! ]Fm,M~I! ]Fm,M~I! ]Fm,M~I! ]Fm,M~I!
··· ···
]t1 ]t2 ]tL ]h1 ]h2 ]hK
Dx 5 ~Dt1, Dt2, · · · DtL, Dh1, Dh2, · · · DhK!T, (15)
~R! ~I! ~R! ~I! ~R! ~I!
Fm 5 $@Fm #1 , @Fm #1 , @Fm #2 , @Fm #2 · · · @Fm #M , @Fm #M % ,
o o o o o o o T
(16)
Fmc 5 $@Fm~R!#1c, @Fm~I!#1c, @Fm~R!#2c, @Fm~I!#2c · · · @Fm~R!#Mc, @Fm~I!#Mc%T, (17)

number. Note that the expansions used to repre- and @Fm~R,I!#i0 and @Fm~R,I!#ic are observed and cal-
sent the lifetime and yield profiles in Eq. ~9! are K and culated real ~or imaginary! parts of the fluorescent
L terms long where K Þ L Þ N in general; however, photon density @based on the estimated ~t, h! distri-
in the research reported here K 5 L 5 N. bution# for i 5 1, 2, . . . , M boundary locations. tl for
To form images from presumably uniform initial l 5 1, 2, . . . , L and hk for k 5 1, 2, . . . , K are the
estimates of the fluorescent lifetime and yield distri- reconstruction parameters for the lifetime and the
butions, we need a way of updating t and h from their yield profiles.
starting values. There are typically two methods Left multiplying Eq. ~13! by the transpose of ) and

1 August 1998 y Vol. 37, No. 22 y APPLIED OPTICS 5339

invoking regularization methods to stabilize the de-
composition of the square system of equations, we
obtained a matrix equation for updating t and h, as
described similarly in Refs. 34 and 35 for optical im-
aging:

~)T) 1 lI!Dx 5 )T~Fmo 2 Fmc!, (18)

where I is the identity matrix and l may be a scalar
or a diagonal matrix.
In this study we have used a low-pass spatial filter
to smooth the reconstructed parameters at each iter-
ative process. As in the case of optical imaging, we
have found that the use of this low-pass filter not only
enhanced the visual quality of the reconstructed im- Fig. 1. Geometry of the test case under study. Transects ~ AB,
ages but also improved the images quantitatively. CD, EF, GH! used to quantify imaging performance are also
This filter acts to average the t and h values of a given shown.
node with values of the surrounding nodes in a
weighted manner so that the influence of the sur-
rounding nodal values can be systematically con- sion light. The radial location of each source was
trolled. This filtering is realized point by point with positioned inside the physical boundary by a dis-
tance, d 5 1yms9, for the point source excitation used
u N* in the computational algorithm.34,35,39 Type III BC’s
xnewi 5 ~1 2 u!xoldi 1
N* (x
j51
oldj , (19) were applied. The finite-element mesh used in this
study consisted of 504 nodes and 992 triangle ele-
ments. The final images reported are the result of
where u is a factor between 0 and 1 and the summa-
iteration until the initial sum of squared errors be-
tion is over the values of the N* nodes directly con-
tween measured and computed intensity and phase
nected to node i. We have found that u 5 0.25
values at the measurement site locations is reduced
appears to give an optimal result for the cases studied
by 5 orders of magnitude. Reaching this level of
to date.
reduction in the initial sum of squared errors typi-
3. Results cally required 30 iterations at a cost of 30 syiteration
for the finite-element mesh used herein in a Sun
In this section our reconstruction algorithm de-
Ultra 30 workstation.
scribed in Section 2 is used to conduct a group of
In each simulation a forward diffusion model with
simulation tests. In these simulation studies we
the exact lifetime and the yield in place generated the
show simulated results that demonstrate a working
implementation of the reconstruction algorithm in
conditions of no measurement noise and with 5%
added noise for the ac intensity and the phase shift of
both excitation and emission light. The modulation
frequency chosen in the simulation is 150 MHz. We
use an image rms error in the fluorescent property
values ~defined below! and the location, size, and
shape of the target to quantify the reconstructed im-
ages.
The test case, shown in Fig. 1, consists of a circular
background region ~radius 5 21.5 mm! with an em-
bedded circular target ~radius 5 6.25 mm! offsetting
5 mm. The optical properties for both the back-
ground and the target are msx,m9 5 1.0 mm21, max 5
0.008 mm21, mam 5 0.005 mm21, and max3m 5 0.003
mm21. The background medium has the fluorescent
properties of t 5 4 ns and h 5 0.15, whereas the
target region has the fluorescent properties of t 5 8 ns
and h 5 0.30. Multiple excitation and measurement
positions were used to produce the boundary infor-
mation used in the reconstructions. Specifically, we
Fig. 2. Simulated simultaneous reconstruction of both fluores-
used 16 excitation positions ~equally spaced around cent lifetime ~t, nanoseconds! and yield ~h, dimensionless! in dif-
the circular circumference! and 16 measurement lo- ferent noise conditions: ~a! t, reconstruction with no noise added;
cations ~also equally spaced around the circular cir- ~b! t, reconstruction with 5% random noise added; ~c! h, reconstruc-
cumference but with a shift relative to the excitation tion with no noise added; ~d! h, reconstruction with 5% random
positions! for detection of both excitation and emis- noise added.

5340 APPLIED OPTICS y Vol. 37, No. 22 y 1 August 1998

 Table 2. Image rms Errorsa for Reconstructed Fluorescent Properties
in Different Noise Conditionsb

Lifetime Yield

Noise Target Target

Condition Background ~Exact Size! Background ~Exact Size!

0% Noise 0.041 0.18 0.048 0.24

5% Noise 0.063 0.26 0.12 0.35
a
See text for definition.
b
The true values for the background are t 5 4 ns, h 5 0.15; for
the target region, t 5 8 ns, h 5 0.30.

determined the image rms errors, which are defined

as

F (S DG
N 2 1y2
1 xexact 2 xreconstructed
,
N i51 xexact

where N is the number of sampled positions along the

Fig. 3. Comparison of exact and simulated reconstructions along
transect AB shown in Fig. 1 with different noise levels: ~a! t
profiles, ~b! h profiles. The horizontal axes indicate transect AB
with millimeter units.
measured data. Figure 2 shows the lifetime and the
yield images reconstructed in conditions of no noise
and with 5% added noise for the measured intensity
and phase shift of both excitation and emission light.
As can be seen, the images formed are qualitatively
correct, even for those with a 5% noise level. Figure
3 provides a more quantitative assessment of these
images, where the reconstructed fluorescent property
distribution is displayed along one transect running
through the centers of both the target and the back-
ground regions ~transect AB in Fig. 1! for the no noise
and 5% noise conditions compared with the exact
values. We note that the images do appear to be
quantitatively recovered. To obtain quantitative in-
formation about the reconstructed images further, we
reconstructed t or h profiles. We also calculated the
location, the size, and the shape of the target. We
estimated these parameters by calculating the
FWHM of the reconstructed fluorescent property pro-
files along the two transects ~see transects AB and
CD in Fig. 1!. In Tables 1 and 2 we present the
results from these calculations for all the images dis-
played in Fig. 2.
4. Discussion and Conclusions
Both qualitative and quantitative useful information
can be obtained from the results in Section 3. The
simulations have shown that the methodology out-
lined in Section 2 leads to a reconstruction algorithm
that can be implemented at a reasonable computa-
tional cost in a workstation computing environment.
Importantly, it has been demonstrated that absolute
fluorescent reconstructions can be obtained quantita-
tively with this approach in terms not only of the
location, the size, and the shape of the heterogeneity
but also of the fluorescent property values them-
selves. Figures 2 and 3 clearly support these con-
clusions. From Fig. 2 we note that the t images
present a better overall recovery of the shape and the
size of the heterogeneity than the h images. This is
particularly true in the case of noisy data. Table 1
provides a quantitative verification of this observa-
tion. Nonetheless the simulated data presented in

Table 1. Geometric Information Derived from the Reconstructed Images in Different Noise Conditionsa

Target Location Target Size Target Shape

t Image h Image t Image h Image t Image h Image

Noise
Condition X Y X Y EF GH EF GH EFyGH EFyGH

0% Noise 4.5 0.2 4.6 0.3 12.8 12.7 12.9 13.1 1.01 0.98
5% Noise 4.3 0.4 4.2 0.8 13.0 13.2 12.2 11.5 0.98 1.06
a
X and Y refer to the x and y coordinates ~in millimeters! of the target center, respectively. EF and GH are the transect length ~i.e.,
the recovered target diameter in millimeters! of the target region along the x and the y directions, respectively ~see Fig. 1!.

1 August 1998 y Vol. 37, No. 22 y APPLIED OPTICS 5341

Figs. 2 and 3 have shown that our reconstruction
algorithm is largely resistant to random noise.
From Figs. 2 and 3 note that the recovered fluores-
cent property values in both the background and the
target regions are quite smooth. Compared with the
results of others ~e.g., Refs. 30 and 33!, unphysically
high values of both lifetime and yield have appeared
in their reconstructions30,33 where the grid points
with these large artifacts are typically on or close to
the tissue periphery. These spurious values were
replaced by the average background lifetime and
yield in their simulation studies. The image errors
shown in Table 2 further suggest that overall good
quantitative accuracies in the recovery of the fluores-
cent property values have been achieved.
In summary, we have demonstrated frequency-
domain fluorescent image reconstructions by using
simulated data. Simultaneous reconstruction of
both fluorescent lifetime and yield has been achieved
quantitatively. Note that the quantitative recon-
structed images have been realized from absolute im-
age reconstruction procedures that do not use
differencing techniques to enhance target definition
by first subtracting the reconstruction of a homoge-
neous region. Although the quality of these images
is promising, we anticipate improvedyenhanced im-
ages after we add our image enhancement schemes,
such as total variation minimization and dual mesh-
ing, into the algorithms, as we have experienced in
optical image reconstructions.36 –38,44 These image
enhancement schemes become particularly impor-
tant when we attempt to reconstruct images of life-
time and yield from in vitro or in vivo measurement
data.
This study was supported in part by a National
Institutes of Health Individual National Research
Service Award ~CA73107!, the Greenville Hospital
SystemyClemson University Biomedical Coopera-
tive, and the Clemson University Research Grant
Committee.
References
1. B. C. Wilson, E. M. Sevick, M. S. Patterson, and B. Chance,
“Time-dependent optical spectroscopy and imaging for biomed-
ical applications,” Proc. IEEE 80, 918 –930 ~1992!.
2. Y. Yamishita and M. Kaneko, “Infrared diaphanoscopy for
medical diagnosis,” in Medical Optical Tomography: Func-
tional Imaging and Monitoring, G. Muller, B. Chance, R. Al-
fano, S. Arridge, J. Beuthan, E. Gratton, M. Kaschke, B.
Masters, S. Svanberg, and P. Van der Zee, ~SPIE Press, Bel-
lingham, Wash., 1993!, pp. 283–316.
3. R. L. Barbour, H. Graber, Y. Wang, J. Chang, and R. Aronson,
“Perturbation approach for optical diffusion tomography using
continuous-wave and time-resolved data,” in Medical Optical
Tomography: Functional Imaging and Monitoring, G. Mul-
ler, B. Chance, R. Alfano, S. Arridge, J. Beuthan, E. Gratton,
M. Kaschke, B. Masters, S. Svanberg, and P. Van der Zee, eds.
~SPIE Press, Bellingham, Wash., 1993!, pp. 87–120.
4. J. C. Hebden and K. S. Wong, “Time-resolved optical tomog-
raphy,” Appl. Opt. 32, 372–380 ~1993!.
5. R. Berg, S. Anderson-Engels, O. Jarlman, and S. Svanberg,
“Tumor detection using time-resolved light transillumination,”
5342 APPLIED OPTICS y Vol. 37, No. 22 y 1 August 1998
in Future Trends in Biomedical Applications of Lasers, L. O.
Svaasand, ed., Proc. SPIE 1525, 59 – 67 ~1991!.
6. B. B. Das, K. M. Yoo, and R. R. Alfano, “Ultrafast time-gated
imaging in thick tissues—a step toward optical mammogra-
phy,” Opt. Lett. 18, 1092–1094 ~1993!.
7. E. Gratton, W. Mantulin, M. J. van de Ven, J. Fishkin, M.
Maris, and B. Chance, “A novel approach to laser tomography,”
Bioimaging 1, 40 – 46 ~1993!.
8. E. M. Sevick, J. K. Frisoli, C. L. Burch, and J. R. Lakowicz,
“Localization of absorbers in scattering media by use of fre-
quency domain measurements of time-dependent photon mi-
gration,” Appl. Opt. 33, 3562–3570 ~1994!.
9. M. A. O’Leary, D. A. Boas, B. Chance, and A. G. Yodh, “Ex-
perimental images of heterogeneous turbid media by
frequency-domain diffusing-photon tomography,” Opt. Lett.
20, 426 – 428 ~1995!.
10. H. Jiang, K. D. Paulsen, U. L. Osterberg, B. W. Pogue, and
M. S. Patterson, “Simultaneous reconstruction of absorption
and scattering maps in turbid media from near-infrared
frequency-domain data,” Opt. Lett. 20, 2128 –2130 ~1995!.
11. V. G. Peters, D. R. Wyman, M. S. Patterson, and G. L. Frank,
“Optical properties of normal and diseased human breast tis-
sue in the visible and near infrared,” Phys. Med. Biol. 35,
1317–1334 ~1990!.
12. T. Troy, D. Page, and E. Sevick-Muraca, “Optical properties of
normal and diseased breast tissues: prognosis for optical
mammography,” J. Biomed. Opt. 1, 342–355 ~1996!.
13. B. J. Tromberg, O. Coquoz, J. Fishkin, T. Pham, E. Anderson,
J. Butler, M. Cahn, J. Gross, V. Venugopalan, and D. Pham,
“Noninvasive measurements of breast tissue optical properties
using frequency-domain photon migration,” Philos. Trans. R.
Soc. London B 352, 661– 668 ~1997!.
14. R. Richards-Kortum, R. P. Rava, M. Fitzmaurice, L. L. Tong,
N. Ratliff, J. Kramer, and M. S. Feld, “A one layer model of
laser-induced fluorescence for diagnosis of disease in human
tissue: applications to atherosclerosis,” IEEE Trans. Biomed.
Eng. 36, 1222–1232 ~1989!.
15. R. R. Alfano, G. C. Tang, A. Pradhan, W. Lam, D. S. Choy, and
E. Opher, “Fluorescence spectra from cancerous and normal
human breast and lung tissues,” IEEE J. Quantum Electron.
QE-23, 1806 –1811 ~1987!.
16. E. S. Andersson, J. Johansson, K. Svanberg, and S. Svanberg,
“Fluorescence imaging and point measurements of tissue:
applications to the determination of malignant tumors and
atherosclerotic lesions from normal tissue,” Photochem. Pho-
tobiol. 53, 807– 814 ~1991!.
17. J. Wu, Y. Wang, L. Perelman, I. Itzkan, R. R. Dasari, and M. S.
Feld, “Time-resolved multichannel imaging of fluorescent ob-
jects embedded in turbid media,” Opt. Lett. 20, 489 – 491
~1995!.
18. Y. Yang, “Fluorescence spectroscopy as a photonic pathology
method for detecting colon cancer,” Lasers Life Sci. 6, 259 –276
~1995!.
19. T. L. Troy, L. Nelson-Larry, C. L. Hutchinson, and E. M.
Sevick-Muraca, “Investigation of exogenous contrast agents
for biomedical optical imaging,” in Biomedical Optical Spec-
troscopy and Diagnostics, E. Sevick-Muraca and D. Benaron,
eds., Vol. 3 of OSA Trends in Optics and Photonics Series
~Optical Society of America, Washington, D.C., 1996!, pp. 104 –
106.
20. E. Sevick-Muraca, G. Lopez, T. Troy, J. Reynolds, and C.
Hutchinson, “Fluorescence and absorption contrast mecha-
nisms for biomedical optical imaging using frequency-domain
techniques,” Photochem. Photobiol. 66, 55– 64 ~1997!.
21. S. Zhao, Y. Yang, S. Nioka, and B. Chance, “Human breast
tumor detection using contrast agent,” in Integration of Med-
ical Optical Imaging and Spectroscopy and Magnetic Reso-
 nance Imaging Symposium Abstracts ~University of Penn-
sylvania, Philadelphia, Pa., 1994!.
22. J. R. Lakowicz, Principles of Fluorescence Spectroscopy ~Ple-
num, New York, 1983!.
23. X. Li, M. O’Leary, D. Boas, B. Chance, and A. Yodh, “Fluores-
cent diffuse photon density waves in homogeneous and heter-
ogeneous turbid media: analytic solutions and applications,”
Appl. Opt. 35, 3746 –3758 ~1996!.
24. A. E. Cerussi, J. S. Maier, S. Fantini, M. A. Franceschini,
W. W. Mantulin, and E. Gratton, “Experimental verification of
a theory for the time-resolved fluorescence spectroscopy of
thick tissues,” Appl. Opt. 36, 116 –124 ~1997!.
25. E. M. Sevick-Muraca and C. L. Burch, “Origin of phosphores-
cence signals reemitted from tissues,” Opt. Lett. 19, 1928 –
1930 ~1994!.
26. M. S. Patterson and B. W. Pogue, “Mathematical model for
time-resolved and frequency-domain fluorescence spectros-
copy in biological tissues,” Appl. Opt. 33, 1963–1974 ~1994!.
27. B. J. Tromberg, S. Madsen, C. Chapman, L. Svaasand, and R.
Haskell, “Fluorescence energy transfer studies on the macro-
phage scavenger receptor,” in Advances in Optical Imaging
and Photon Migration, R. R. Alfano, ed., Vol. 21 of 1994 OSA
Proceedings Series ~Optical Society of America, Washington,
D.C., 1994!, p. 93.
28. A. J. Durkin, S. Jailkumar, and R. Richards-Kortum, “Opti-
cally dilute, absorbing, and turbid phantoms for fluorescence
spectroscopy of homogeneous and inhomogeneous samples,”
Appl. Spectrosc. 47, 2114 –2121 ~1993!.
29. J. Wu, M. S. Feld, and R. P. Rava, “Analytical model for ex-
tracting intrinsic fluorescence in turbid media,” Appl. Opt. 32,
3585–3595 ~1993!.
30. M. A. O’Leary, D. A. Boas, X. D. Li, B. Chance, and A. G. Yodh,
“Fluorescence lifetime imaging in turbid media,” Opt. Lett. 21,
158 –160 ~1996!.
31. J. Chang, H. L. Graber, and R. L. Barbour, “Luminescence
optical tomography of dense scattering media,” J. Opt. Soc.
Am. A 14, 288 –299 ~1997!.
32. J. Chang, H. Graber, and R. L. Barbour, “Imaging of fluores-
cence in highly scattering media,” IEEE Trans. Biomed. Eng.
44, 810 – 822 ~1997!.
33. D. Y. Paithankar, A. U. Chen, B. W. Pogue, M. S. Patterson,
and E. M. Sevick-Muraca, “Imaging of fluorescent lifetime and
yield from multiple scattered light reemitted from tissues and
other random media,” Appl. Opt. 36, 2260 –2272 ~1997!.
34. K. D. Paulsen and H. Jiang, “Spatially-varying optical prop-
erty reconstruction using a finite element diffusion equation
approximation,” Med. Phys. 22, 691–702 ~1995!.
35. H. Jiang, K. D. Paulsen, U. L. Osterberg, B. W. Pogue, and
M. S. Patterson, “Optical image reconstruction using
frequency-domain data: simulations and experiments,” J.
Opt. Soc. Am. A 13, 253–266 ~1996!.
36. K. D. Paulsen and H. Jiang, “Enhanced frequency-domain
optical image reconstruction in tissues through total variation
minimization,” Appl. Opt. 35, 3447–3458 ~1996!.
37. H. Jiang, K. D. Paulsen, U. L. Osterberg, and M. S. Patterson,
“Frequency-domain optical image reconstruction in heteroge-
neous media: an experimental study of single-target detect-
ability,” Appl. Opt. 36, 52– 63 ~1997!.
38. H. Jiang, K. D. Paulsen, U. L. Osterberg, and M. S. Patterson,
“Frequency-domain optical image reconstruction for breast im-
aging: initial evaluation in multitarget tissuelike phantoms,”
Med. Phys. 25, 183–193 ~1998!.
39. S. R. Arridge, M. Schweijer, M. Hiraoka, and D. T. Delpy, “A
finite element approach for modeling photon transport in tis-
sue,” Med Phys. 20, 299 –308 ~1993!.
40. M. Schweijer and S. R. Arridge, “Direct calculation with a
finite-element method of the Laplace transform of the distri-
bution of photon time of flight in tissue,” Appl. Opt. 36, 9042–
9049 ~1997!.
41. S. Takahashi, D. Imai, and Y. Yamada, “Fundamental 3D
FEM analysis of light propagation in head model toward 3D
optical tomography,” in Optical Tomography and Spectroscopy
of Tissue: Theory, Instrumentation, Model and Human Stud-
ies II, B. Chance and R. Alfano, eds., Proc. SPIE 2979, 130 –138
~1997!.
42. Y. Yamada and Y. Hasegawa, “Time-dependent FEM analysis
of photon migration in random media,” in Optical Tomography
and Spectroscopy of Tissue: Theory, Instrumentation, Model
and Human Studies, B. Chance and R. Alfano, eds., Proc. SPIE
1888, 167–178 ~1993!.
43. R. C. Haskell, L. O. Svaasand, T. Tsay, T. C. Feng, M. S.
McAdams, and B. J. Tromberg, “Boundary conditions for the
diffusion equation in radiative transfer,” J. Opt. Soc. Am. A 11,
2727–2741 ~1994!.
44. H. Jiang, “Reconstructed near infrared diffusion imaging for
breast cancer detection,” Ph.D. dissertation ~Dartmouth Col-
lege, Hanover, N.H., 1995!.
1 August 1998 y Vol. 37, No. 22 y APPLIED OPTICS 5343