Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images

I. Reiser, S. P. Joseph, R. M. Nishikawa, M. L. Giger, John M Boone, Karen K Lindfors, A. Edwards, N. Packard, R. H. Moore, D. B. Kopans

Research output: Chapter in Book/Report/Conference proceedingConference contribution

12 Citations (Scopus)

Abstract

Recently, tomosynthesis (DBT) and CT (BCT) have been developed for breast imaging. Since each modality produces a fundamentally different representation of the breast volume, our goal was to investigate whether a 3D segmentation algorithm for breast masses could be applied to both DBT and breast BCT images. A secondary goal of this study was to investigate a simplified method for comparing manual outlines to a computer segmentation. The seeded mass lesion segmentation algorithm is based on maximizing the radial gradient index (RGI) along a constrained region contour. In DBT, the constraint function was a prolate spherical Gaussian, with a larger FWHM along the depth direction where the resolution is low, while it was a spherical Gaussian for BCT. For DBT, manual lesion outlines were obtained in the in-focus plane of the lesion, which was used to compute the overlap ratio with the computer segmentation. For BCT, lesions were manually outlined in three orthogonal planes, and the average overlap ratio from the three planes was computed. In DBT, 81% of all lesions were segmented at an overlap ratio of 0.4 or higher, based on manual outlines in one slice through the lesion center. In BCT, 93% of all segmentations achieved an average overlap ratio of 0.4, based on the manual outlines in three orthogonal planes. Our results indicate mass lesions in both BCT and DBT images can be segmented with the proposed 3D segmentation algorithm, by selecting an appropriate set of parameters and after images have undergone specific pre-processing.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2010: Computer-Aided Diagnosis
PublisherSPIE
Volume7624
ISBN (Electronic)9780819480255
DOIs
StatePublished - 2010
EventMedical Imaging 2010: Computer-Aided Diagnosis - San Diego, United States
Duration: Feb 16 2010Feb 18 2010

Other

OtherMedical Imaging 2010: Computer-Aided Diagnosis
CountryUnited States
CitySan Diego
Period2/16/102/18/10

Fingerprint

breast
lesions
Breast
evaluation
Phosmet
Full width at half maximum
Imaging techniques
Processing
preprocessing
gradients
Direction compound

Keywords

  • 3D segmentation
  • breast imaging
  • computer-aided diagnosis
  • CT
  • tomosynthesis

ASJC Scopus subject areas

  • Atomic and Molecular Physics, and Optics
  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Reiser, I., Joseph, S. P., Nishikawa, R. M., Giger, M. L., Boone, J. M., Lindfors, K. K., ... Kopans, D. B. (2010). Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images. In Medical Imaging 2010: Computer-Aided Diagnosis (Vol. 7624). [76242N] SPIE. https://doi.org/10.1117/12.844484

Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images. / Reiser, I.; Joseph, S. P.; Nishikawa, R. M.; Giger, M. L.; Boone, John M; Lindfors, Karen K; Edwards, A.; Packard, N.; Moore, R. H.; Kopans, D. B.

Medical Imaging 2010: Computer-Aided Diagnosis. Vol. 7624 SPIE, 2010. 76242N.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Reiser, I, Joseph, SP, Nishikawa, RM, Giger, ML, Boone, JM, Lindfors, KK, Edwards, A, Packard, N, Moore, RH & Kopans, DB 2010, Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images. in Medical Imaging 2010: Computer-Aided Diagnosis. vol. 7624, 76242N, SPIE, Medical Imaging 2010: Computer-Aided Diagnosis, San Diego, United States, 2/16/10. https://doi.org/10.1117/12.844484
Reiser I, Joseph SP, Nishikawa RM, Giger ML, Boone JM, Lindfors KK et al. Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images. In Medical Imaging 2010: Computer-Aided Diagnosis. Vol. 7624. SPIE. 2010. 76242N https://doi.org/10.1117/12.844484
Reiser, I. ; Joseph, S. P. ; Nishikawa, R. M. ; Giger, M. L. ; Boone, John M ; Lindfors, Karen K ; Edwards, A. ; Packard, N. ; Moore, R. H. ; Kopans, D. B. / Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images. Medical Imaging 2010: Computer-Aided Diagnosis. Vol. 7624 SPIE, 2010.
@inproceedings{60de8216a3ca4abb951a33e3ffcd911d,
title = "Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images",
abstract = "Recently, tomosynthesis (DBT) and CT (BCT) have been developed for breast imaging. Since each modality produces a fundamentally different representation of the breast volume, our goal was to investigate whether a 3D segmentation algorithm for breast masses could be applied to both DBT and breast BCT images. A secondary goal of this study was to investigate a simplified method for comparing manual outlines to a computer segmentation. The seeded mass lesion segmentation algorithm is based on maximizing the radial gradient index (RGI) along a constrained region contour. In DBT, the constraint function was a prolate spherical Gaussian, with a larger FWHM along the depth direction where the resolution is low, while it was a spherical Gaussian for BCT. For DBT, manual lesion outlines were obtained in the in-focus plane of the lesion, which was used to compute the overlap ratio with the computer segmentation. For BCT, lesions were manually outlined in three orthogonal planes, and the average overlap ratio from the three planes was computed. In DBT, 81{\%} of all lesions were segmented at an overlap ratio of 0.4 or higher, based on manual outlines in one slice through the lesion center. In BCT, 93{\%} of all segmentations achieved an average overlap ratio of 0.4, based on the manual outlines in three orthogonal planes. Our results indicate mass lesions in both BCT and DBT images can be segmented with the proposed 3D segmentation algorithm, by selecting an appropriate set of parameters and after images have undergone specific pre-processing.",
keywords = "3D segmentation, breast imaging, computer-aided diagnosis, CT, tomosynthesis",
author = "I. Reiser and Joseph, {S. P.} and Nishikawa, {R. M.} and Giger, {M. L.} and Boone, {John M} and Lindfors, {Karen K} and A. Edwards and N. Packard and Moore, {R. H.} and Kopans, {D. B.}",
year = "2010",
doi = "10.1117/12.844484",
language = "English (US)",
volume = "7624",
booktitle = "Medical Imaging 2010: Computer-Aided Diagnosis",
publisher = "SPIE",

}

TY - GEN

T1 - Evaluation of a 3D lesion segmentation algorithm on DBT and breast CT images

AU - Reiser, I.

AU - Joseph, S. P.

AU - Nishikawa, R. M.

AU - Giger, M. L.

AU - Boone, John M

AU - Lindfors, Karen K

AU - Edwards, A.

AU - Packard, N.

AU - Moore, R. H.

AU - Kopans, D. B.

PY - 2010

Y1 - 2010

N2 - Recently, tomosynthesis (DBT) and CT (BCT) have been developed for breast imaging. Since each modality produces a fundamentally different representation of the breast volume, our goal was to investigate whether a 3D segmentation algorithm for breast masses could be applied to both DBT and breast BCT images. A secondary goal of this study was to investigate a simplified method for comparing manual outlines to a computer segmentation. The seeded mass lesion segmentation algorithm is based on maximizing the radial gradient index (RGI) along a constrained region contour. In DBT, the constraint function was a prolate spherical Gaussian, with a larger FWHM along the depth direction where the resolution is low, while it was a spherical Gaussian for BCT. For DBT, manual lesion outlines were obtained in the in-focus plane of the lesion, which was used to compute the overlap ratio with the computer segmentation. For BCT, lesions were manually outlined in three orthogonal planes, and the average overlap ratio from the three planes was computed. In DBT, 81% of all lesions were segmented at an overlap ratio of 0.4 or higher, based on manual outlines in one slice through the lesion center. In BCT, 93% of all segmentations achieved an average overlap ratio of 0.4, based on the manual outlines in three orthogonal planes. Our results indicate mass lesions in both BCT and DBT images can be segmented with the proposed 3D segmentation algorithm, by selecting an appropriate set of parameters and after images have undergone specific pre-processing.

AB - Recently, tomosynthesis (DBT) and CT (BCT) have been developed for breast imaging. Since each modality produces a fundamentally different representation of the breast volume, our goal was to investigate whether a 3D segmentation algorithm for breast masses could be applied to both DBT and breast BCT images. A secondary goal of this study was to investigate a simplified method for comparing manual outlines to a computer segmentation. The seeded mass lesion segmentation algorithm is based on maximizing the radial gradient index (RGI) along a constrained region contour. In DBT, the constraint function was a prolate spherical Gaussian, with a larger FWHM along the depth direction where the resolution is low, while it was a spherical Gaussian for BCT. For DBT, manual lesion outlines were obtained in the in-focus plane of the lesion, which was used to compute the overlap ratio with the computer segmentation. For BCT, lesions were manually outlined in three orthogonal planes, and the average overlap ratio from the three planes was computed. In DBT, 81% of all lesions were segmented at an overlap ratio of 0.4 or higher, based on manual outlines in one slice through the lesion center. In BCT, 93% of all segmentations achieved an average overlap ratio of 0.4, based on the manual outlines in three orthogonal planes. Our results indicate mass lesions in both BCT and DBT images can be segmented with the proposed 3D segmentation algorithm, by selecting an appropriate set of parameters and after images have undergone specific pre-processing.

KW - 3D segmentation

KW - breast imaging

KW - computer-aided diagnosis

KW - CT

KW - tomosynthesis

UR - http://www.scopus.com/inward/record.url?scp=84897022642&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84897022642&partnerID=8YFLogxK

U2 - 10.1117/12.844484

DO - 10.1117/12.844484

M3 - Conference contribution

AN - SCOPUS:84897022642

VL - 7624

BT - Medical Imaging 2010: Computer-Aided Diagnosis

PB - SPIE

ER -