Cervigram image segmentation based on reconstructive sparse representations

Shaoting Zhang, Junzhou Huang, Wei Wang, Sharon Xiaolei Huang, Dimitris Metaxas

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

6 Citations (Scopus)

Abstract

We proposed an approach based on reconstructive sparse representations to segment tissues in optical images of the uterine cervix. Because of large variations in image appearance caused by the changing of the illumination and specular reflection, the color and texture features in optical images often overlap with each other and are not linearly separable. By leveraging sparse representations the data can be transformed to higher dimensions with sparse constraints and become more separated. K-SVD algorithm is employed to find sparse representations and corresponding dictionaries. The data can be reconstructed from its sparse representations and positive and/or negative dictionaries. Classification can be achieved based on comparing the reconstructive errors. In the experiments we applied our method to automatically segment the biomarker AcetoWhite (AW) regions in an archive of 60,000 images of the uterine cervix. Compared with other general methods, our approach showed lower space and time complexity and higher sensitivity.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2010
Subtitle of host publicationImage Processing
EditionPART 1
DOIs
StatePublished - Dec 1 2010
EventMedical Imaging 2010: Image Processing - San Diego, CA, United States
Duration: Feb 14 2010Feb 16 2010

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
NumberPART 1
Volume7623
ISSN (Print)1605-7422

Other

OtherMedical Imaging 2010: Image Processing
CountryUnited States
CitySan Diego, CA
Period2/14/102/16/10

Fingerprint

Glossaries
Image segmentation
Cervix Uteri
dictionaries
Biomarkers
Singular value decomposition
Lighting
Color
Textures
Tissue
specular reflection
biomarkers
textures
Experiments
illumination
color
sensitivity

All Science Journal Classification (ASJC) codes

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

Cite this

Zhang, S., Huang, J., Wang, W., Huang, S. X., & Metaxas, D. (2010). Cervigram image segmentation based on reconstructive sparse representations. In Medical Imaging 2010: Image Processing (PART 1 ed.). [762313] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 7623, No. PART 1). https://doi.org/10.1117/12.845461
Zhang, Shaoting ; Huang, Junzhou ; Wang, Wei ; Huang, Sharon Xiaolei ; Metaxas, Dimitris. / Cervigram image segmentation based on reconstructive sparse representations. Medical Imaging 2010: Image Processing. PART 1. ed. 2010. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; PART 1).
@inproceedings{8fecd1f8db3049f0899e38dafa814782,
title = "Cervigram image segmentation based on reconstructive sparse representations",
abstract = "We proposed an approach based on reconstructive sparse representations to segment tissues in optical images of the uterine cervix. Because of large variations in image appearance caused by the changing of the illumination and specular reflection, the color and texture features in optical images often overlap with each other and are not linearly separable. By leveraging sparse representations the data can be transformed to higher dimensions with sparse constraints and become more separated. K-SVD algorithm is employed to find sparse representations and corresponding dictionaries. The data can be reconstructed from its sparse representations and positive and/or negative dictionaries. Classification can be achieved based on comparing the reconstructive errors. In the experiments we applied our method to automatically segment the biomarker AcetoWhite (AW) regions in an archive of 60,000 images of the uterine cervix. Compared with other general methods, our approach showed lower space and time complexity and higher sensitivity.",
author = "Shaoting Zhang and Junzhou Huang and Wei Wang and Huang, {Sharon Xiaolei} and Dimitris Metaxas",
year = "2010",
month = "12",
day = "1",
doi = "10.1117/12.845461",
language = "English (US)",
isbn = "9780819480248",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
number = "PART 1",
booktitle = "Medical Imaging 2010",
edition = "PART 1",

}

Zhang, S, Huang, J, Wang, W, Huang, SX & Metaxas, D 2010, Cervigram image segmentation based on reconstructive sparse representations. in Medical Imaging 2010: Image Processing. PART 1 edn, 762313, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, no. PART 1, vol. 7623, Medical Imaging 2010: Image Processing, San Diego, CA, United States, 2/14/10. https://doi.org/10.1117/12.845461

Cervigram image segmentation based on reconstructive sparse representations. / Zhang, Shaoting; Huang, Junzhou; Wang, Wei; Huang, Sharon Xiaolei; Metaxas, Dimitris.

Medical Imaging 2010: Image Processing. PART 1. ed. 2010. 762313 (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 7623, No. PART 1).

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

TY - GEN

T1 - Cervigram image segmentation based on reconstructive sparse representations

AU - Zhang, Shaoting

AU - Huang, Junzhou

AU - Wang, Wei

AU - Huang, Sharon Xiaolei

AU - Metaxas, Dimitris

PY - 2010/12/1

Y1 - 2010/12/1

N2 - We proposed an approach based on reconstructive sparse representations to segment tissues in optical images of the uterine cervix. Because of large variations in image appearance caused by the changing of the illumination and specular reflection, the color and texture features in optical images often overlap with each other and are not linearly separable. By leveraging sparse representations the data can be transformed to higher dimensions with sparse constraints and become more separated. K-SVD algorithm is employed to find sparse representations and corresponding dictionaries. The data can be reconstructed from its sparse representations and positive and/or negative dictionaries. Classification can be achieved based on comparing the reconstructive errors. In the experiments we applied our method to automatically segment the biomarker AcetoWhite (AW) regions in an archive of 60,000 images of the uterine cervix. Compared with other general methods, our approach showed lower space and time complexity and higher sensitivity.

AB - We proposed an approach based on reconstructive sparse representations to segment tissues in optical images of the uterine cervix. Because of large variations in image appearance caused by the changing of the illumination and specular reflection, the color and texture features in optical images often overlap with each other and are not linearly separable. By leveraging sparse representations the data can be transformed to higher dimensions with sparse constraints and become more separated. K-SVD algorithm is employed to find sparse representations and corresponding dictionaries. The data can be reconstructed from its sparse representations and positive and/or negative dictionaries. Classification can be achieved based on comparing the reconstructive errors. In the experiments we applied our method to automatically segment the biomarker AcetoWhite (AW) regions in an archive of 60,000 images of the uterine cervix. Compared with other general methods, our approach showed lower space and time complexity and higher sensitivity.

UR - http://www.scopus.com/inward/record.url?scp=79751475431&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79751475431&partnerID=8YFLogxK

U2 - 10.1117/12.845461

DO - 10.1117/12.845461

M3 - Conference contribution

SN - 9780819480248

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

BT - Medical Imaging 2010

ER -

Zhang S, Huang J, Wang W, Huang SX, Metaxas D. Cervigram image segmentation based on reconstructive sparse representations. In Medical Imaging 2010: Image Processing. PART 1 ed. 2010. 762313. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; PART 1). https://doi.org/10.1117/12.845461