没有合适的资源?快使用搜索试试~ 我知道了~
Brief review on learning-based methods for optical tomography
0 下载量 189 浏览量
2021-01-26
05:30:43
上传
评论
收藏 427KB PDF 举报
温馨提示
试读
14页
Learning-based methods have been proved to perform well in a variety of areas in the biomedical field, such as biomedical image segmentation, and histopathological image analysis. Deep learning, as the most recently presented approach of learning-based methods, has attracted more and more attention. For instance, massive researches of deep learning methods for image reconstructions of computed tomography (CT) and magnetic resonance imaging (MRI) have been reported, indicating the great potential
资源详情
资源评论
资源推荐
Brief review on learning-based methods for optical
tomography
Lin Zhang and Guanglei Zhang
*
Beijing Advanced Innovation Center for Biomedical Engineering
Beihang University
Beijing 100191, P. R. China
School of Biological Science and Medical Engineering
Beihang University
Beijing 100191, P. R. China
*
Received 2 February 2019
Accepted 13 July 2019
Published 19 August 2019
Learning-based methods have been proved to perform well in a variety of areas in the biomedical
¯eld, such as biomedical image segmentation, and histopathological image analysis. Deep
learning, as the most recently presented approach of learning-based methods, has attracted more
and more attention. For instance, massive researches of deep learning methods for image
reconstructions of computed tomography (CT) and magnetic resonance imaging (MRI) have
been reported, indicating the great potential of deep learning for inverse problems. Optical
technology-related medical imaging modalities including di®use optical tomography (DOT),
°uorescence molecular tomography (FMT), bioluminescence tomography (BLT), and photo-
acoustic tomography (PAT) are also dramatically innovated by introducing learning-based
methods, in particular deep learning methods, to obtain better reconst ruction results. This
review depicts the latest researches on learning-based optical tomography of DOT, FMT, BLT,
and PAT. According to the most recent studies, learning-based methods applied in the ¯eld of
optical tomography are categorized as kernel-based methods and deep learning methods. In this
review, the former are regarded as a sort of conventional learning-based methods and the latter
are subdivided into model-based methods, post-processing methods, and end-to-end methods.
Algorithm as well as data acquisition strategy are discussed in this revi ew. The evaluations of
these methods are summarized to illustrate the performance of deep learning-based
reconstruction.
Keywords: Optical imaging; tomography; inverse problem; machine learning; deep learning.
*
Corresponding author.
This is an Open Access article published by World Scienti¯c Publishing Company. It is distributed under the terms of the Creative
Commons Attribution 4.0 (CC BY) License which permits use, distribution and reproduction in any medium, provided the original
work is properly cited.
Journal of Innovative Optical Health Sciences
Vol. 12, No. 6 (2019) 1930011 (14 pages)
#
.
c
The Author(s)
DOI: 10.1142/S1793545819300118
1930011-1
J. Innov. Opt. Health Sci. 2019.12. Downloaded from www.worldscientific.com
by 27.17.81.156 on 12/09/19. Re-use and distribution is strictly not permitted, except for Open Access articles.
1. Introduction
Di®use optical tomography (DOT), °uorescence
molecular tomography (FMT), together with bio-
luminescence tomography (BLT) are noninvasive
modalities for biomedical imaging.
1
–
5
Photons
propagating through biological tissues can be col-
lected in vivo by these modalities, indicating the
spatial optical properties like absorption coe±cient
and scattering coe±cient which can be obtained by
all the three methods, or the internal °uorophore
distribution which can be obtained by FMT and
BLT methods.
6
–
8
The optical coe±cients, as well as
the intensity and lifetime of °uorescence or biolu-
minescence, are generally in°uenced by oxygen
saturation, hemoglobin concentrations and other
situations of the tissue. Therefore, these modalities
of optical tomography are promising to be adopted
in tracing tumors and drug development, and have
been paid more and more attention.
1,9
The inverse problems of optical tomography are
ill-posed and sometimes nonlinear.
10
Therefore, it is
a challenge to obtain accurate and high-resolution
reconstructions by solving the inverse problems. A
mass of methods have been proposed to improve the
quantitative quality by developing novel e±cient
algorithms, and to reduce the time consumption by
developing accelerated algorithms or by rebuilding
the ¯nal results through limited-view and sparse
measurements.
11,12
Multi-modality technologies are
also utilized to obtain high-quality reconstructions
by introducing anatomical guidance from compu-
tational tomography (CT) or other modalities.
13,14
Coupled-physics biomedical imaging technologies,
such as photoacoustic imaging (PAI) or tomogra-
phy (PAT), are also developed to obtain the high-
quality reconstructions by bene¯cially combining
the high-contrast of optical imaging with the high-
spatial resolution of ultrasound imaging.
15
–
19
Regularization and optimization strategies with
iterative algorithms are generally employed to solve
the ill-posed and nonlinear inverse problems.
11,12,20
Learning-based methods have also been introduced
to solve inverse problems. One of the most recently
presented conventional learning-based methods is
kernel method, which can help make best of ana-
tomical information for reconstruction of multi-
modality imaging.
21
However, conventional princi-
ples of reconstruction for optical tomography have
been changed by the development of deep learning
technique.
22
Deep learning, as a new-raised part of
machine learning, has shown huge advantages in
computer vision, particularly in image classi¯cation
and other areas of pattern recognition.
23
Deep
neural networks can learn from the coupled samples
to generate a model mapping from the input to the
output. By introducing deep learning technologies,
the inverse problems in biomedical imaging can be
converted into data-driven problems.
Actually, deep learning technologies have been
widely adopted in solving biomedical problems.
Researchers have been utilizing deep neural net-
works in image segmentation and histopathological
image analysis, as well as in image reconstruction
for CT and magnetic resonance imaging (MRI),
etc.
24
–
27
Meanwhile, novel algorithms which are
based on deep learning have been consistently
employed into reconstruction issues of optical im-
aging and tomography. In order to obtain a com-
prehensive understanding of learning-based
methods, several state-of-the-art algorithms for
reconstructions of DOT, FMT, and BLT, together
with PAT or PAI will be reviewed here. Considering
the great potentiality in reconstruction of optical
tomography, this review will dramatically place
emphasis on deep learning-based algorithms. Ker-
nel-based methods, which are derived from machine
learning technology, are also novel approaches for
the reconstruction of optical tomography and will
be reviewed here as a contrast. All the deep learn-
ing-based algorithms mentioned here will be cate-
gorized into model-based methods, post-processing
methods, and end-to-end methods.
The rest of this review is organized as follows.
In Sec. 2, methodologies of kernel-based methods
for DOT and FMT are depicted. In Sec. 3, deep
learning-based methods including model-based
methods, post-processing methods, and end-to-end
methods are discussed, respectively. In Sec. 4, there
is an overall summary of the methods presented in
this review.
2. Kernel-Based Methods
Kernel-based method is one series of pattern anal-
ysis algorithms in machine learning, including the
most famous methods such as support vector ma-
chine (SVM), Gaussian process and so on, owing
the name to the utilization of kernel functions.
28,29
By mapping to high-dimension spaces described by
kernel functions, nonlinear problems can be
L. Zhang & G. Zhang
1930011-2
J. Innov. Opt. Health Sci. 2019.12. Downloaded from www.worldscientific.com
by 27.17.81.156 on 12/09/19. Re-use and distribution is strictly not permitted, except for Open Access articles.
changed into linear problems. In Wang's research, a
kernel-based method was introduced to positron
emission computed tomography (PET) to help the
image reconstruction.
21
Features obtained from
prior information are modeled and incorporated
into the forward function. Inspired by the well
performance of kernel-based method in PET re-
construction, similar approaches have been pro-
posed to improve reconstructions in optical
tomography in the last couple years.
Reconstructions in both DOT and FMT su®er
from the ill-posed and nonlinear inverse problems.
Anatomical guidance, as prior information, is of
great importance in improving the reconstruction
quality.
30,31
However, introducing anatomical in-
formation from other imaging modalities with high
spatial resolution calls for image segmentation and
registration ¯rst. Utilizing a kernel-based method,
optical absorption coe±cient at each ¯nite element
node is represented as a function of a set of features
obtained from anatomical images without any
preprocessing.
2.1. Kernel-based methods for DOT
As mentioned in Baikejiang et al.'s work, a Gauss-
ian kernel-based method was presented to involve
anatomical information in the forward model for the
DOT image reconstruction.
28,32
According to relative literatures, the theory of
Gaussian kernel-based reconstruction method for
DOT can be described as follows.
As DOT reconstruction is usually implemented
on ¯nite element method (FEM), the kernel func-
tion can be formulized as
k
m;n
¼
exp
jjf
m
f
n
jj
2
2
; f
n
2 knn of f
m
;
0; otherwise;
8
<
:
ð1Þ
where f
m
and f
n
are feature vectors corresponding
to ¯nite element nodes m and n extracted from
anatomical image, respectively. Feature vector f
m
consists of all the values of the neighbor voxels
surrounding the ¯nite element node m. knn denotes
the k closest neighbors of f
m
and can be acquired by
K nearest neighbors (KNN) search algorithm.
26
Ignoring the optical scattering coe±cient, the
absorption coe±cient vector can be written as
a
¼ K; ð2Þ
where K is the kernel matrix de¯ned by the kernel
function k
m;n
. Vector is the nth value of the vector
to be reconstructed.
In general, the objective function of inverse
problem for DOT can be written as
arg min F ð
a
Þ¼
1
2
jjy W ð
a
Þjj
2
; ð3Þ
where W is the system matrix representing the
forward model. By combining Eqs. (2) and (3), the
inverse problem for DOT can be described as
arg min F ðÞ¼
1
2
jjy W ðKÞjj
2
: ð4Þ
Then, the inverse problem of DOT can be trans-
ferred into a minimized problem of vector and the
prior information has been involved. Vector is
then obtained by conventional iterative method
such as Tikhonov regularization, and the absorption
coe±cient can be obtained using Eq. (2).
33
In the numerical simulation of Baikejiang's work,
it was shown that the voxel number of each corre-
sponding node and the number of nearest neighbors
had impact on the performance of kernel-based
method signi¯cantly. Di®erent values of k (16, 32,
or 64) and di®erent voxel numbers (3 3 3,
5 5 5, 7 7 7, or 9 9 9) were taken into
account to optimize the kernel-based method. Vol-
ume ratio (VR), Dice similarity coe±cient (Dice),
contrast-to-noise ratio (CNR), and mean square
error (MSE) are introduced as evaluation metrics.
34
Reconstruction results indicated that the kernel-
based method outperformed other cases whether
the value of k or the voxel number increased.
Then other three numerical experiments were
carried out to validate the performance of kernel-
based method. CT contrast e®ect in kernel-based
method for CT-guided DOT reconstruction, e®ect
of the false positive target in the kernel-based
method, and clinical breast CT image as anatomical
guidance were studied and compared to the
Laplacian-type soft prior method, respectively.
35
Numerical reconstruction results indicated that the
kernel-based method was robust to CT contrast and
the false positive targets in the guided CT image.
Although in some cases the reconstructions of ker-
nel-based method were not as good as that of soft
prior method, it still outperformed the method
without prior information and did not need seg-
mentation for the CT guidance.
Brief review on learning-based methods for optical tomography
1930011-3
J. Innov. Opt. Health Sci. 2019.12. Downloaded from www.worldscientific.com
by 27.17.81.156 on 12/09/19. Re-use and distribution is strictly not permitted, except for Open Access articles.
剩余13页未读,继续阅读
weixin_38557530
- 粉丝: 5
- 资源: 897
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功
评论0