Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features

Xulei Qin, Guolan Lu, Ioannis Sechopoulos, Baowei Fei

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

12 Citations (Scopus)

Abstract

Digital breast tomosynthesis (DBT) is a pseudo-three-dimensional x-ray imaging modality proposed to decrease the effect of tissue superposition present in mammography, potentially resulting in an increase in clinical performance for the detection and diagnosis of breast cancer. Tissue classification in DBT images can be useful in risk assessment, computer-aided detection and radiation dosimetry, among other aspects. However, classifying breast tissue in DBT is a challenging problem because DBT images include complicated structures, image noise, and out-of-plane artifacts due to limited angular tomographic sampling. In this project, we propose an automatic method to classify fatty and glandular tissue in DBT images. First, the DBT images are pre-processed to enhance the tissue structures and to decrease image noise and artifacts. Second, a global smooth filter based on L0 gradient minimization is applied to eliminate detailed structures and enhance large-scale ones. Third, the similar structure regions are extracted and labeled by fuzzy C-means (FCM) classification. At the same time, the texture features are also calculated. Finally, each region is classified into different tissue types based on both intensity and texture features. The proposed method is validated using five patient DBT images using manual segmentation as the gold standard. The Dice scores and the confusion matrix are utilized to evaluate the classified results. The evaluation results demonstrated the feasibility of the proposed method for classifying breast glandular and fat tissue on DBT images.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2014
Subtitle of host publicationImage Processing
PublisherSPIE
ISBN (Print)9780819498274
DOIs
StatePublished - Jan 1 2014
Externally publishedYes
EventMedical Imaging 2014: Image Processing - San Diego, CA, United States
Duration: Feb 16 2014Feb 18 2014

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume9034
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2014: Image Processing
CountryUnited States
CitySan Diego, CA
Period2/16/142/18/14

Fingerprint

Mammography
breast
Breast
textures
Textures
Tissue
gradients
optimization
Artifacts
classifying
artifacts
Radiometry
Oils and fats
Risk assessment
Dosimetry
risk assessment
Fats
confusion
fats
Sampling

Keywords

  • Breast tissue classification
  • Digital breast tomosynthesis
  • Global gradient minimization
  • Texture features

ASJC Scopus subject areas

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

Cite this

Qin, X., Lu, G., Sechopoulos, I., & Fei, B. (2014). Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features. In Medical Imaging 2014: Image Processing [90341V] (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 9034). SPIE. https://doi.org/10.1117/12.2043828

Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features. / Qin, Xulei; Lu, Guolan; Sechopoulos, Ioannis; Fei, Baowei.

Medical Imaging 2014: Image Processing. SPIE, 2014. 90341V (Progress in Biomedical Optics and Imaging - Proceedings of SPIE; Vol. 9034).

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

Qin, X, Lu, G, Sechopoulos, I & Fei, B 2014, Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features. in Medical Imaging 2014: Image Processing., 90341V, Progress in Biomedical Optics and Imaging - Proceedings of SPIE, vol. 9034, SPIE, Medical Imaging 2014: Image Processing, San Diego, CA, United States, 2/16/14. https://doi.org/10.1117/12.2043828
Qin X, Lu G, Sechopoulos I, Fei B. Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features. In Medical Imaging 2014: Image Processing. SPIE. 2014. 90341V. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE). https://doi.org/10.1117/12.2043828
Qin, Xulei ; Lu, Guolan ; Sechopoulos, Ioannis ; Fei, Baowei. / Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features. Medical Imaging 2014: Image Processing. SPIE, 2014. (Progress in Biomedical Optics and Imaging - Proceedings of SPIE).
@inproceedings{0b2db99157624b3f88cb6180e4d14eb0,
title = "Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features",
abstract = "Digital breast tomosynthesis (DBT) is a pseudo-three-dimensional x-ray imaging modality proposed to decrease the effect of tissue superposition present in mammography, potentially resulting in an increase in clinical performance for the detection and diagnosis of breast cancer. Tissue classification in DBT images can be useful in risk assessment, computer-aided detection and radiation dosimetry, among other aspects. However, classifying breast tissue in DBT is a challenging problem because DBT images include complicated structures, image noise, and out-of-plane artifacts due to limited angular tomographic sampling. In this project, we propose an automatic method to classify fatty and glandular tissue in DBT images. First, the DBT images are pre-processed to enhance the tissue structures and to decrease image noise and artifacts. Second, a global smooth filter based on L0 gradient minimization is applied to eliminate detailed structures and enhance large-scale ones. Third, the similar structure regions are extracted and labeled by fuzzy C-means (FCM) classification. At the same time, the texture features are also calculated. Finally, each region is classified into different tissue types based on both intensity and texture features. The proposed method is validated using five patient DBT images using manual segmentation as the gold standard. The Dice scores and the confusion matrix are utilized to evaluate the classified results. The evaluation results demonstrated the feasibility of the proposed method for classifying breast glandular and fat tissue on DBT images.",
keywords = "Breast tissue classification, Digital breast tomosynthesis, Global gradient minimization, Texture features",
author = "Xulei Qin and Guolan Lu and Ioannis Sechopoulos and Baowei Fei",
year = "2014",
month = "1",
day = "1",
doi = "10.1117/12.2043828",
language = "English (US)",
isbn = "9780819498274",
series = "Progress in Biomedical Optics and Imaging - Proceedings of SPIE",
publisher = "SPIE",
booktitle = "Medical Imaging 2014",

}

TY - GEN

T1 - Breast tissue classification in digital tomosynthesis images based on global gradient minimization and texture features

AU - Qin, Xulei

AU - Lu, Guolan

AU - Sechopoulos, Ioannis

AU - Fei, Baowei

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Digital breast tomosynthesis (DBT) is a pseudo-three-dimensional x-ray imaging modality proposed to decrease the effect of tissue superposition present in mammography, potentially resulting in an increase in clinical performance for the detection and diagnosis of breast cancer. Tissue classification in DBT images can be useful in risk assessment, computer-aided detection and radiation dosimetry, among other aspects. However, classifying breast tissue in DBT is a challenging problem because DBT images include complicated structures, image noise, and out-of-plane artifacts due to limited angular tomographic sampling. In this project, we propose an automatic method to classify fatty and glandular tissue in DBT images. First, the DBT images are pre-processed to enhance the tissue structures and to decrease image noise and artifacts. Second, a global smooth filter based on L0 gradient minimization is applied to eliminate detailed structures and enhance large-scale ones. Third, the similar structure regions are extracted and labeled by fuzzy C-means (FCM) classification. At the same time, the texture features are also calculated. Finally, each region is classified into different tissue types based on both intensity and texture features. The proposed method is validated using five patient DBT images using manual segmentation as the gold standard. The Dice scores and the confusion matrix are utilized to evaluate the classified results. The evaluation results demonstrated the feasibility of the proposed method for classifying breast glandular and fat tissue on DBT images.

AB - Digital breast tomosynthesis (DBT) is a pseudo-three-dimensional x-ray imaging modality proposed to decrease the effect of tissue superposition present in mammography, potentially resulting in an increase in clinical performance for the detection and diagnosis of breast cancer. Tissue classification in DBT images can be useful in risk assessment, computer-aided detection and radiation dosimetry, among other aspects. However, classifying breast tissue in DBT is a challenging problem because DBT images include complicated structures, image noise, and out-of-plane artifacts due to limited angular tomographic sampling. In this project, we propose an automatic method to classify fatty and glandular tissue in DBT images. First, the DBT images are pre-processed to enhance the tissue structures and to decrease image noise and artifacts. Second, a global smooth filter based on L0 gradient minimization is applied to eliminate detailed structures and enhance large-scale ones. Third, the similar structure regions are extracted and labeled by fuzzy C-means (FCM) classification. At the same time, the texture features are also calculated. Finally, each region is classified into different tissue types based on both intensity and texture features. The proposed method is validated using five patient DBT images using manual segmentation as the gold standard. The Dice scores and the confusion matrix are utilized to evaluate the classified results. The evaluation results demonstrated the feasibility of the proposed method for classifying breast glandular and fat tissue on DBT images.

KW - Breast tissue classification

KW - Digital breast tomosynthesis

KW - Global gradient minimization

KW - Texture features

UR - http://www.scopus.com/inward/record.url?scp=84902096285&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902096285&partnerID=8YFLogxK

U2 - 10.1117/12.2043828

DO - 10.1117/12.2043828

M3 - Conference contribution

C2 - 25426271

AN - SCOPUS:84902096285

SN - 9780819498274

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

BT - Medical Imaging 2014

PB - SPIE

ER -