Computer-aided classification of breast masses using speckle features of automated breast ultrasound images

Woo Kyung Moon, Chung Ming Lo, Jung Min Chang, Chiun Sheng Huang, Jeon Hor Chen, Ruey Feng Chang

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

Purpose: To develop an ultrasound computer-aided diagnosis (CAD) system using speckle features of automated breast ultrasound (ABUS) images. Methods: The ABUS images of 147 pathologically proven breast masses (76 benign and 71 malignant cases) were used. For each mass, a volume of interest (VOI) was cropped to define the tumor area, and the average number of speckle pixels within a VOI was calculated. In addition, first-order and second-order statistical analyses of the speckle pixels were used to quantify the information of gray-level distributions and the spatial relations among the pixels. Receiver operating characteristic curve analysis was used to evaluate the performance. Results: The proposed CAD system based on speckle patterns achieved an accuracy of 84.4 (124147), a sensitivity of 83.1 (5971), a specificity of 85.5 (6576), and an Az of 0.91. The performance indices of the speckle features were comparable to the performance indices of the morphological features, which include shape and ellipse-fitting features (p-value > 0.05). Furthermore, combining speckle and morphological features yielded an Az that was significantly better than the Az of the morphological features alone (0.96 vs 0.91, p-value 0.0154). Conclusions: The results suggest that the proposed speckle features, while combined with morphological features, are promising for the classification of breast masses detected using ABUS.

Original languageEnglish
Pages (from-to)6465-6473
Number of pages9
JournalMedical Physics
Volume39
Issue number10
DOIs
Publication statusPublished - Oct 2012
Externally publishedYes

Fingerprint

Breast
ROC Curve
Neoplasms

Keywords

  • automated breast ultrasound
  • breast cancer
  • computer-assisted diagnosis
  • speckle

ASJC Scopus subject areas

  • Biophysics
  • Radiology Nuclear Medicine and imaging

Cite this

Moon, W. K., Lo, C. M., Chang, J. M., Huang, C. S., Chen, J. H., & Chang, R. F. (2012). Computer-aided classification of breast masses using speckle features of automated breast ultrasound images. Medical Physics, 39(10), 6465-6473. https://doi.org/10.1118/1.4754801

Computer-aided classification of breast masses using speckle features of automated breast ultrasound images. / Moon, Woo Kyung; Lo, Chung Ming; Chang, Jung Min; Huang, Chiun Sheng; Chen, Jeon Hor; Chang, Ruey Feng.

In: Medical Physics, Vol. 39, No. 10, 10.2012, p. 6465-6473.

Research output: Contribution to journalArticle

Moon, WK, Lo, CM, Chang, JM, Huang, CS, Chen, JH & Chang, RF 2012, 'Computer-aided classification of breast masses using speckle features of automated breast ultrasound images', Medical Physics, vol. 39, no. 10, pp. 6465-6473. https://doi.org/10.1118/1.4754801
Moon, Woo Kyung ; Lo, Chung Ming ; Chang, Jung Min ; Huang, Chiun Sheng ; Chen, Jeon Hor ; Chang, Ruey Feng. / Computer-aided classification of breast masses using speckle features of automated breast ultrasound images. In: Medical Physics. 2012 ; Vol. 39, No. 10. pp. 6465-6473.
@article{549f1f8e6f174379b63e6806a8f1f430,
title = "Computer-aided classification of breast masses using speckle features of automated breast ultrasound images",
abstract = "Purpose: To develop an ultrasound computer-aided diagnosis (CAD) system using speckle features of automated breast ultrasound (ABUS) images. Methods: The ABUS images of 147 pathologically proven breast masses (76 benign and 71 malignant cases) were used. For each mass, a volume of interest (VOI) was cropped to define the tumor area, and the average number of speckle pixels within a VOI was calculated. In addition, first-order and second-order statistical analyses of the speckle pixels were used to quantify the information of gray-level distributions and the spatial relations among the pixels. Receiver operating characteristic curve analysis was used to evaluate the performance. Results: The proposed CAD system based on speckle patterns achieved an accuracy of 84.4 (124147), a sensitivity of 83.1 (5971), a specificity of 85.5 (6576), and an Az of 0.91. The performance indices of the speckle features were comparable to the performance indices of the morphological features, which include shape and ellipse-fitting features (p-value > 0.05). Furthermore, combining speckle and morphological features yielded an Az that was significantly better than the Az of the morphological features alone (0.96 vs 0.91, p-value 0.0154). Conclusions: The results suggest that the proposed speckle features, while combined with morphological features, are promising for the classification of breast masses detected using ABUS.",
keywords = "automated breast ultrasound, breast cancer, computer-assisted diagnosis, speckle",
author = "Moon, {Woo Kyung} and Lo, {Chung Ming} and Chang, {Jung Min} and Huang, {Chiun Sheng} and Chen, {Jeon Hor} and Chang, {Ruey Feng}",
year = "2012",
month = "10",
doi = "10.1118/1.4754801",
language = "English",
volume = "39",
pages = "6465--6473",
journal = "Medical Physics",
issn = "0094-2405",
publisher = "AAPM - American Association of Physicists in Medicine",
number = "10",

}

TY - JOUR

T1 - Computer-aided classification of breast masses using speckle features of automated breast ultrasound images

AU - Moon, Woo Kyung

AU - Lo, Chung Ming

AU - Chang, Jung Min

AU - Huang, Chiun Sheng

AU - Chen, Jeon Hor

AU - Chang, Ruey Feng

PY - 2012/10

Y1 - 2012/10

N2 - Purpose: To develop an ultrasound computer-aided diagnosis (CAD) system using speckle features of automated breast ultrasound (ABUS) images. Methods: The ABUS images of 147 pathologically proven breast masses (76 benign and 71 malignant cases) were used. For each mass, a volume of interest (VOI) was cropped to define the tumor area, and the average number of speckle pixels within a VOI was calculated. In addition, first-order and second-order statistical analyses of the speckle pixels were used to quantify the information of gray-level distributions and the spatial relations among the pixels. Receiver operating characteristic curve analysis was used to evaluate the performance. Results: The proposed CAD system based on speckle patterns achieved an accuracy of 84.4 (124147), a sensitivity of 83.1 (5971), a specificity of 85.5 (6576), and an Az of 0.91. The performance indices of the speckle features were comparable to the performance indices of the morphological features, which include shape and ellipse-fitting features (p-value > 0.05). Furthermore, combining speckle and morphological features yielded an Az that was significantly better than the Az of the morphological features alone (0.96 vs 0.91, p-value 0.0154). Conclusions: The results suggest that the proposed speckle features, while combined with morphological features, are promising for the classification of breast masses detected using ABUS.

AB - Purpose: To develop an ultrasound computer-aided diagnosis (CAD) system using speckle features of automated breast ultrasound (ABUS) images. Methods: The ABUS images of 147 pathologically proven breast masses (76 benign and 71 malignant cases) were used. For each mass, a volume of interest (VOI) was cropped to define the tumor area, and the average number of speckle pixels within a VOI was calculated. In addition, first-order and second-order statistical analyses of the speckle pixels were used to quantify the information of gray-level distributions and the spatial relations among the pixels. Receiver operating characteristic curve analysis was used to evaluate the performance. Results: The proposed CAD system based on speckle patterns achieved an accuracy of 84.4 (124147), a sensitivity of 83.1 (5971), a specificity of 85.5 (6576), and an Az of 0.91. The performance indices of the speckle features were comparable to the performance indices of the morphological features, which include shape and ellipse-fitting features (p-value > 0.05). Furthermore, combining speckle and morphological features yielded an Az that was significantly better than the Az of the morphological features alone (0.96 vs 0.91, p-value 0.0154). Conclusions: The results suggest that the proposed speckle features, while combined with morphological features, are promising for the classification of breast masses detected using ABUS.

KW - automated breast ultrasound

KW - breast cancer

KW - computer-assisted diagnosis

KW - speckle

UR - http://www.scopus.com/inward/record.url?scp=84867295558&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867295558&partnerID=8YFLogxK

U2 - 10.1118/1.4754801

DO - 10.1118/1.4754801

M3 - Article

VL - 39

SP - 6465

EP - 6473

JO - Medical Physics

JF - Medical Physics

SN - 0094-2405

IS - 10

ER -