Microwave imaging of the breast with incorporated structural information

Amir Golnabi, Paul M. Meaney, Shireen D. Geimer, Keith D. Paulsen

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

3 Citations (Scopus)

Abstract

Microwave imaging for biomedical applications, especially for early detection of breast cancer and effective treatment monitoring, has attracted increasing interest in last several decades. This fact is due to the high contrast between the dielectric properties of the normal and malignant breast tissues at microwave frequencies ranging from high megahertz to low gigahertz. The available range of dielectric properties for different soft tissue can provide considerable functional information about tissue health. Nonetheless, one of the limiting weaknesses of microwave imaging is, unlike that for conventional modalities such as X-ray CT or MRI, it cannot inherently provide high-resolution images. The conventional modalities can produce highly resolved anatomical information but often cannot provide the functional information required for diagnoses. We have developed a soft prior regularization strategy that can incorporate the prior anatomical information from X-ray CT, MR or other sources, and use it in a way to exploit the resolution of these images while also retaining the functional nature of the microwave images. The anatomical information is first used to create an imaging zone mesh, which segments separate internal substructures, and an associated weighting matrix that numerically groups the values of closely related nodes within the mesh. This information is subsequently used as a regularizing term for the Gauss-Newton reconstruction algorithm. This approach exploits existing technology in a systematic way without making potentially biased assumptions about the properties of visible structures. In this paper we continue our initial investigation on this matter with a series of breast-shaped simulation and phantom experiments.

Original languageEnglish
Title of host publicationMedical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging
Volume7626
DOIs
StatePublished - 15 Jun 2010
EventMedical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging - San Diego, CA, United States
Duration: 14 Feb 201016 Feb 2010

Other

OtherMedical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging
CountryUnited States
CitySan Diego, CA
Period14/02/1016/02/10

Fingerprint

Microwaves
breast
Breast
Tissue
Imaging techniques
microwaves
Dielectric properties
X Ray Computed Tomography
X rays
Microwave frequencies
Image resolution
Magnetic resonance imaging
dielectric properties
mesh
Early Detection of Cancer
Health
Monitoring
Breast Neoplasms
microwave frequencies
retaining

Keywords

  • Breast cancer
  • Inverse problem
  • Microwave imaging
  • Soft prior regularization
  • Tikhonov regularization

Cite this

Golnabi, A., Meaney, P. M., Geimer, S. D., & Paulsen, K. D. (2010). Microwave imaging of the breast with incorporated structural information. In Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging (Vol. 7626). [76260P] https://doi.org/10.1117/12.843855
Golnabi, Amir ; Meaney, Paul M. ; Geimer, Shireen D. ; Paulsen, Keith D. / Microwave imaging of the breast with incorporated structural information. Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 7626 2010.
@inproceedings{b383cd9b5895412abfd177086692af45,
title = "Microwave imaging of the breast with incorporated structural information",
abstract = "Microwave imaging for biomedical applications, especially for early detection of breast cancer and effective treatment monitoring, has attracted increasing interest in last several decades. This fact is due to the high contrast between the dielectric properties of the normal and malignant breast tissues at microwave frequencies ranging from high megahertz to low gigahertz. The available range of dielectric properties for different soft tissue can provide considerable functional information about tissue health. Nonetheless, one of the limiting weaknesses of microwave imaging is, unlike that for conventional modalities such as X-ray CT or MRI, it cannot inherently provide high-resolution images. The conventional modalities can produce highly resolved anatomical information but often cannot provide the functional information required for diagnoses. We have developed a soft prior regularization strategy that can incorporate the prior anatomical information from X-ray CT, MR or other sources, and use it in a way to exploit the resolution of these images while also retaining the functional nature of the microwave images. The anatomical information is first used to create an imaging zone mesh, which segments separate internal substructures, and an associated weighting matrix that numerically groups the values of closely related nodes within the mesh. This information is subsequently used as a regularizing term for the Gauss-Newton reconstruction algorithm. This approach exploits existing technology in a systematic way without making potentially biased assumptions about the properties of visible structures. In this paper we continue our initial investigation on this matter with a series of breast-shaped simulation and phantom experiments.",
keywords = "Breast cancer, Inverse problem, Microwave imaging, Soft prior regularization, Tikhonov regularization",
author = "Amir Golnabi and Meaney, {Paul M.} and Geimer, {Shireen D.} and Paulsen, {Keith D.}",
year = "2010",
month = "6",
day = "15",
doi = "10.1117/12.843855",
language = "English",
isbn = "9780819480279",
volume = "7626",
booktitle = "Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging",

}

Golnabi, A, Meaney, PM, Geimer, SD & Paulsen, KD 2010, Microwave imaging of the breast with incorporated structural information. in Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. vol. 7626, 76260P, Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging, San Diego, CA, United States, 14/02/10. https://doi.org/10.1117/12.843855

Microwave imaging of the breast with incorporated structural information. / Golnabi, Amir; Meaney, Paul M.; Geimer, Shireen D.; Paulsen, Keith D.

Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 7626 2010. 76260P.

Research output: Chapter in Book/Report/Conference proceedingConference contributionResearchpeer-review

TY - GEN

T1 - Microwave imaging of the breast with incorporated structural information

AU - Golnabi, Amir

AU - Meaney, Paul M.

AU - Geimer, Shireen D.

AU - Paulsen, Keith D.

PY - 2010/6/15

Y1 - 2010/6/15

N2 - Microwave imaging for biomedical applications, especially for early detection of breast cancer and effective treatment monitoring, has attracted increasing interest in last several decades. This fact is due to the high contrast between the dielectric properties of the normal and malignant breast tissues at microwave frequencies ranging from high megahertz to low gigahertz. The available range of dielectric properties for different soft tissue can provide considerable functional information about tissue health. Nonetheless, one of the limiting weaknesses of microwave imaging is, unlike that for conventional modalities such as X-ray CT or MRI, it cannot inherently provide high-resolution images. The conventional modalities can produce highly resolved anatomical information but often cannot provide the functional information required for diagnoses. We have developed a soft prior regularization strategy that can incorporate the prior anatomical information from X-ray CT, MR or other sources, and use it in a way to exploit the resolution of these images while also retaining the functional nature of the microwave images. The anatomical information is first used to create an imaging zone mesh, which segments separate internal substructures, and an associated weighting matrix that numerically groups the values of closely related nodes within the mesh. This information is subsequently used as a regularizing term for the Gauss-Newton reconstruction algorithm. This approach exploits existing technology in a systematic way without making potentially biased assumptions about the properties of visible structures. In this paper we continue our initial investigation on this matter with a series of breast-shaped simulation and phantom experiments.

AB - Microwave imaging for biomedical applications, especially for early detection of breast cancer and effective treatment monitoring, has attracted increasing interest in last several decades. This fact is due to the high contrast between the dielectric properties of the normal and malignant breast tissues at microwave frequencies ranging from high megahertz to low gigahertz. The available range of dielectric properties for different soft tissue can provide considerable functional information about tissue health. Nonetheless, one of the limiting weaknesses of microwave imaging is, unlike that for conventional modalities such as X-ray CT or MRI, it cannot inherently provide high-resolution images. The conventional modalities can produce highly resolved anatomical information but often cannot provide the functional information required for diagnoses. We have developed a soft prior regularization strategy that can incorporate the prior anatomical information from X-ray CT, MR or other sources, and use it in a way to exploit the resolution of these images while also retaining the functional nature of the microwave images. The anatomical information is first used to create an imaging zone mesh, which segments separate internal substructures, and an associated weighting matrix that numerically groups the values of closely related nodes within the mesh. This information is subsequently used as a regularizing term for the Gauss-Newton reconstruction algorithm. This approach exploits existing technology in a systematic way without making potentially biased assumptions about the properties of visible structures. In this paper we continue our initial investigation on this matter with a series of breast-shaped simulation and phantom experiments.

KW - Breast cancer

KW - Inverse problem

KW - Microwave imaging

KW - Soft prior regularization

KW - Tikhonov regularization

UR - http://www.scopus.com/inward/record.url?scp=77953320910&partnerID=8YFLogxK

U2 - 10.1117/12.843855

DO - 10.1117/12.843855

M3 - Conference contribution

SN - 9780819480279

VL - 7626

BT - Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging

ER -

Golnabi A, Meaney PM, Geimer SD, Paulsen KD. Microwave imaging of the breast with incorporated structural information. In Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 7626. 2010. 76260P https://doi.org/10.1117/12.843855