Biomedical Engineering Reference
In-Depth Information
where
k
is
=
−µω
c
i
a
k
(9.34)
cD
To reduce the effect of the mismatch between measured and calculated data as
well as of unknown system parameters, such as different gains and time/phase delays
for each source-detector pair and heterogeneity of background media on the recon-
struction image, the fluorescence intensity has to be normalized to the background
signal. Better results have been obtained by using the normalized ratio of fluores-
cence emission signal to the unfiltered diffused signal, coming from the excitation
light (eq. 9.35). In this case, the unknown system parameters that are common in
both fluorescence and unfiltered diffused signal will be canceled out. unfiltered dif-
fused signal has better signal-to-noise ratio and provides more information about the
heterogeneity and optical properties of the background tissues than just the background
fluorescence signal:
ϕ
ϕ
()
()
i
i
ϕ
ϕ
()
i
=
[
] [
]
=… =…
em measured
_
em calculated
_
−
Wij
().
,
∆α
( )
j
i Mj
1
,
,
and
1
,
,
N
()
i
ex measured
_
ex calcul
_
ated
(9.35)
where
φ
em _ measured
(
i
) and
φ
em _ calculated
(
i
) are the measured and calculated fluorescence
field and
φ
ex _ measured
(
i
) and
φ
ex _ calculated
(
i
) are the measured and the calculated excitation
field, respectively, all associated with the source-detector pair
i
[72, 73].
Different optimization algorithms can be used to evaluate the unknown fluores-
cence parameters
Δα
in equation 9.30. The most common are Newton/Marquardt-
levenberg algorithms and the total least-squares method [71, 74].
The primary difficulty of diffuse optical tomography (DoT) is related to the fact
that multiple scattering dominates NIr light propagation in tissue, making 3D local-
ization of targets and accurate quantification of their optical parameters difficult.
Scattering properties of the media and a limited number of measurements compared
to the number of voxels make equation 9.35 ill posed and underdetermined.
Coregistration of fluorescence imaging with high-resolution imaging modalities such
as CT, MrI, and uS can provide anatomical structures to reduce the uncertainties of
unknown variables in the fluorescence reconstruction algorithms and to improve the
two-dimensional (2D) and 3D images of the fluorescence probes [75-80].
Since absorption and scattering properties of the tissue can affect the detected
fluorescence yield and lifetime in the deep tissue imaging, they should be taken into
account in the imaging reconstruction algorithms. These properties of the tissue can
be estimated by DoT methods [81]. In general, the quality of image reconstruction
depends upon many parameters, including signal-to-noise ratio, measurement geom-
etry, and depth of the target from the source and detector. A more thorough review of
fluorescence optical tomography is provided in [47, 82].
Search WWH ::
Custom Search