Regular Articles

Partial dependence of breast tumor malignancy on ultrasound image features derived from boosted trees

[+] Author Affiliations
Wei Yang

Shanghai Jiao Tong University, Department of Biomedical Engineering, Shanghai 200240, China and Southern Medical University, School of Biomedical Engineering, Guangzhou 510515, China

Su Zhang

Shanghai Jiao Tong University, Department of Biomedical Engineering, Shanghai 200240, China

Wenying Li

Shanghai Jiao Tong University, Shanghai Sixth People’s Hospital, Shanghai 200231, China

Yaqing Chen

Shanghai Jiao Tong University, Shanghai Sixth People’s Hospital, Shanghai 200231, China

Hongtao Lu

Shanghai Jiao Tong University, Department of Computer Science, Shanghai 200240, China

Wufan Chen

Southern Medical University, School of Biomedical Engineering, Guangzhou 510515, China

Yazhu Chen

Shanghai Jiao Tong University, Department of Biomedical Engineering, Shanghai 200240, China

J. Electron. Imaging. 19(2), 023004 (April 12, 2010). doi:10.1117/1.3385763
History: Received June 09, 2009; Revised February 22, 2010; Accepted March 01, 2010; Published April 12, 2010; Online April 12, 2010
Text Size: A A A

Various computerized features extracted from breast ultrasound images are useful in assessing the malignancy of breast tumors. However, the underlying relationship between the computerized features and tumor malignancy may not be linear in nature. We use the decision tree ensemble trained by the cost-sensitive boosting algorithm to approximate the target function for malignancy assessment and to reflect this relationship qualitatively. Partial dependence plots are employed to explore and visualize the effect of features on the output of the decision tree ensemble. In the experiments, 31 image features are extracted to quantify the sonographic characteristics of breast tumors. Patient age is used as an external feature because of its high clinical importance. The area under the receiver-operating characteristic curve of the tree ensembles can reach 0.95 with sensitivity of 0.95 (6164) at the associated specificity 0.74 (77104). The partial dependence plots of the four most important features are demonstrated to show the influence of the features on malignancy, and they are in accord with the empirical observations. The results can provide visual and qualitative references on the computerized image features for physicians, and can be useful for enhancing the interpretability of computer-aided diagnosis systems for breast ultrasound.

Figures in this Article
© 2010 SPIE and IS&T

Citation

Wei Yang ; Su Zhang ; Wenying Li ; Yaqing Chen ; Hongtao Lu, et al.
"Partial dependence of breast tumor malignancy on ultrasound image features derived from boosted trees", J. Electron. Imaging. 19(2), 023004 (April 12, 2010). ; http://dx.doi.org/10.1117/1.3385763


Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Related Book Chapters

Topic Collections

PubMed Articles
Advertisement
  • Don't have an account?
  • Subscribe to the SPIE Digital Library
  • Create a FREE account to sign up for Digital Library content alerts and gain access to institutional subscriptions remotely.
Access This Article
Sign in or Create a personal account to Buy this article ($20 for members, $25 for non-members).
Access This Proceeding
Sign in or Create a personal account to Buy this article ($15 for members, $18 for non-members).
Access This Chapter

Access to SPIE eBooks is limited to subscribing institutions and is not available as part of a personal subscription. Print or electronic versions of individual SPIE books may be purchased via SPIE.org.