Kernel-based anatomically-aided diffuse optical tomography reconstruction

Reheman Baikejiang, Wei Zhang, Dianwen Zhu, Andrew M. Hernandez, Shadi Aminololama-Shakeri, Guobao Wang, Jin Yi Qi, John M Boone, Chang Qing Li

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Image reconstruction in diffuse optical tomography (DOT) is a challenging task because its inverse problem is nonlinear, ill-posed, and ill-conditioned. Anatomical guidance from imaging modalities with high spatial resolution can substantially improve the quality of reconstructed DOT images. In this paper, inspired by the kernel methods in machine learning, we proposed a kernel method to introduce anatomical guidance into the DOT image reconstruction. In this kernel method, the optical absorption coefficient at each finite element node is represented as a function of a set of features obtained from anatomical images such as computed tomography (CT) images. Compared with Laplacian approaches that include structural priors, the proposed method does not require image segmentation. The proposed kernel method is validated with numerical simulations of 3D DOT reconstruction using synthetic CT data. 5% Gaussian noise was added to both the numerical DOT measurements and the simulated CT images. The proposed method was also validated by an agar phantom experiment with the anatomical guidance from a cone beam CT scan. The effects of voxel size and number of nearest neighbors in the kernel method on the reconstructed DOT images were studied. The results indicate that the spatial resolution and the accuracy of the reconstructed DOT images have been improved substantially after applying the anatomical guidance with the proposed kernel method comparing to the case without guidance. Furthermore, we demonstrated that the kernel method was able to utilize clinical breast CT images as anatomical guidance without segmentation. In addition, we found that the proposed kernel method was robust to the false positive guidance in the anatomical image.

Original languageEnglish (US)
Article number055002
JournalBiomedical Physics and Engineering Express
Volume3
Issue number5
DOIs
StatePublished - Sep 13 2017

Fingerprint

Optical Tomography
Tomography
Computer-Assisted Image Processing
Cone-Beam Computed Tomography
Agar
Breast

Keywords

  • Anatomical guidance
  • Diffuse optical tomography
  • Kernel method
  • Multi-modality imaging

ASJC Scopus subject areas

  • Nursing(all)

Cite this

Kernel-based anatomically-aided diffuse optical tomography reconstruction. / Baikejiang, Reheman; Zhang, Wei; Zhu, Dianwen; Hernandez, Andrew M.; Aminololama-Shakeri, Shadi; Wang, Guobao; Qi, Jin Yi; Boone, John M; Li, Chang Qing.

In: Biomedical Physics and Engineering Express, Vol. 3, No. 5, 055002, 13.09.2017.

Research output: Contribution to journalArticle

Baikejiang, Reheman ; Zhang, Wei ; Zhu, Dianwen ; Hernandez, Andrew M. ; Aminololama-Shakeri, Shadi ; Wang, Guobao ; Qi, Jin Yi ; Boone, John M ; Li, Chang Qing. / Kernel-based anatomically-aided diffuse optical tomography reconstruction. In: Biomedical Physics and Engineering Express. 2017 ; Vol. 3, No. 5.
@article{103bd0858eb24e0286ebdb6bbed3e826,
title = "Kernel-based anatomically-aided diffuse optical tomography reconstruction",
abstract = "Image reconstruction in diffuse optical tomography (DOT) is a challenging task because its inverse problem is nonlinear, ill-posed, and ill-conditioned. Anatomical guidance from imaging modalities with high spatial resolution can substantially improve the quality of reconstructed DOT images. In this paper, inspired by the kernel methods in machine learning, we proposed a kernel method to introduce anatomical guidance into the DOT image reconstruction. In this kernel method, the optical absorption coefficient at each finite element node is represented as a function of a set of features obtained from anatomical images such as computed tomography (CT) images. Compared with Laplacian approaches that include structural priors, the proposed method does not require image segmentation. The proposed kernel method is validated with numerical simulations of 3D DOT reconstruction using synthetic CT data. 5{\%} Gaussian noise was added to both the numerical DOT measurements and the simulated CT images. The proposed method was also validated by an agar phantom experiment with the anatomical guidance from a cone beam CT scan. The effects of voxel size and number of nearest neighbors in the kernel method on the reconstructed DOT images were studied. The results indicate that the spatial resolution and the accuracy of the reconstructed DOT images have been improved substantially after applying the anatomical guidance with the proposed kernel method comparing to the case without guidance. Furthermore, we demonstrated that the kernel method was able to utilize clinical breast CT images as anatomical guidance without segmentation. In addition, we found that the proposed kernel method was robust to the false positive guidance in the anatomical image.",
keywords = "Anatomical guidance, Diffuse optical tomography, Kernel method, Multi-modality imaging",
author = "Reheman Baikejiang and Wei Zhang and Dianwen Zhu and Hernandez, {Andrew M.} and Shadi Aminololama-Shakeri and Guobao Wang and Qi, {Jin Yi} and Boone, {John M} and Li, {Chang Qing}",
year = "2017",
month = "9",
day = "13",
doi = "10.1088/2057-1976/aa87bb",
language = "English (US)",
volume = "3",
journal = "Biomedical Physics and Engineering Express",
issn = "2057-1976",
publisher = "IOP Publishing Ltd.",
number = "5",

}

TY - JOUR

T1 - Kernel-based anatomically-aided diffuse optical tomography reconstruction

AU - Baikejiang, Reheman

AU - Zhang, Wei

AU - Zhu, Dianwen

AU - Hernandez, Andrew M.

AU - Aminololama-Shakeri, Shadi

AU - Wang, Guobao

AU - Qi, Jin Yi

AU - Boone, John M

AU - Li, Chang Qing

PY - 2017/9/13

Y1 - 2017/9/13

N2 - Image reconstruction in diffuse optical tomography (DOT) is a challenging task because its inverse problem is nonlinear, ill-posed, and ill-conditioned. Anatomical guidance from imaging modalities with high spatial resolution can substantially improve the quality of reconstructed DOT images. In this paper, inspired by the kernel methods in machine learning, we proposed a kernel method to introduce anatomical guidance into the DOT image reconstruction. In this kernel method, the optical absorption coefficient at each finite element node is represented as a function of a set of features obtained from anatomical images such as computed tomography (CT) images. Compared with Laplacian approaches that include structural priors, the proposed method does not require image segmentation. The proposed kernel method is validated with numerical simulations of 3D DOT reconstruction using synthetic CT data. 5% Gaussian noise was added to both the numerical DOT measurements and the simulated CT images. The proposed method was also validated by an agar phantom experiment with the anatomical guidance from a cone beam CT scan. The effects of voxel size and number of nearest neighbors in the kernel method on the reconstructed DOT images were studied. The results indicate that the spatial resolution and the accuracy of the reconstructed DOT images have been improved substantially after applying the anatomical guidance with the proposed kernel method comparing to the case without guidance. Furthermore, we demonstrated that the kernel method was able to utilize clinical breast CT images as anatomical guidance without segmentation. In addition, we found that the proposed kernel method was robust to the false positive guidance in the anatomical image.

AB - Image reconstruction in diffuse optical tomography (DOT) is a challenging task because its inverse problem is nonlinear, ill-posed, and ill-conditioned. Anatomical guidance from imaging modalities with high spatial resolution can substantially improve the quality of reconstructed DOT images. In this paper, inspired by the kernel methods in machine learning, we proposed a kernel method to introduce anatomical guidance into the DOT image reconstruction. In this kernel method, the optical absorption coefficient at each finite element node is represented as a function of a set of features obtained from anatomical images such as computed tomography (CT) images. Compared with Laplacian approaches that include structural priors, the proposed method does not require image segmentation. The proposed kernel method is validated with numerical simulations of 3D DOT reconstruction using synthetic CT data. 5% Gaussian noise was added to both the numerical DOT measurements and the simulated CT images. The proposed method was also validated by an agar phantom experiment with the anatomical guidance from a cone beam CT scan. The effects of voxel size and number of nearest neighbors in the kernel method on the reconstructed DOT images were studied. The results indicate that the spatial resolution and the accuracy of the reconstructed DOT images have been improved substantially after applying the anatomical guidance with the proposed kernel method comparing to the case without guidance. Furthermore, we demonstrated that the kernel method was able to utilize clinical breast CT images as anatomical guidance without segmentation. In addition, we found that the proposed kernel method was robust to the false positive guidance in the anatomical image.

KW - Anatomical guidance

KW - Diffuse optical tomography

KW - Kernel method

KW - Multi-modality imaging

UR - http://www.scopus.com/inward/record.url?scp=85031126989&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031126989&partnerID=8YFLogxK

U2 - 10.1088/2057-1976/aa87bb

DO - 10.1088/2057-1976/aa87bb

M3 - Article

AN - SCOPUS:85031126989

VL - 3

JO - Biomedical Physics and Engineering Express

JF - Biomedical Physics and Engineering Express

SN - 2057-1976

IS - 5

M1 - 055002

ER -