Computerized image analysis for identifying triple-negative breast cancers and differentiating them from other molecular subtypes of breast cancer on dynamic contrast-enhanced MR images: a feasibility study

Abstract

Purpose: To determine the feasibility of using a computer-aided diagnosis (CAD) system to differentiate among triple-negative breast cancer, estrogen receptor (ER)-positive cancer, human epidermal growth factor receptor type 2 (HER2)-positive cancer, and benign fibroadenoma lesions on dynamic contrast material-enhanced (DCE) magnetic resonance (MR) images.

Materials and methods: This is a retrospective study of prospectively acquired breast MR imaging data collected from an institutional review board-approved, HIPAA-compliant study between 2002 and 2007. Written informed consent was obtained from all patients. The authors collected DCE MR images from 65 women with 76 breast lesions who had been recruited into a larger study of breast MR imaging. The women had triple-negative (n = 21), ER-positive (n = 25), HER2-positive (n = 18), or fibroadenoma (n = 12) lesions. All lesions were classified as Breast Imaging Reporting and Data System category 4 or higher on the basis of previous imaging. Images were subject to quantitative feature extraction, feed-forward feature selection by means of linear discriminant analysis, and lesion classification by using a support vector machine classifier. The area under the receiver operating characteristic curve (Az) was calculated for each of five lesion classification tasks involving triple-negative breast cancers.

Results: For each pair-wise lesion type comparison, linear discriminant analysis helped identify the most discriminatory features, which in conjunction with a support vector machine classifier yielded an Az of 0.73 (95% confidence interval [CI]: 0.59, 0.87) for triple-negative cancer versus all non-triple-negative lesions, 0.74 (95% CI: 0.60, 0.88) for triple-negative cancer versus ER- and HER2-positive cancer, 0.77 (95% CI: 0.63, 0.91) for triple-negative versus ER-positive cancer, 0.74 (95% CI: 0.58, 0.89) for triple-negative versus HER2-positive cancer, and 0.97 (95% CI: 0.91, 1.00) for triple-negative cancer versus fibroadenoma.

Conclusion: Triple-negative cancers possess certain characteristic features on DCE MR images that can be captured and quantified with CAD, enabling good discrimination of triple-negative cancers from non-triple-negative cancers, as well as between triple-negative cancers and benign fibroadenomas. Such CAD algorithms may provide added diagnostic benefit in identifying the highly aggressive triple-negative cancer phenotype with DCE MR imaging in high-risk women.

Figure 1a:

Comparison of enhanced appearance and textural features of triple-negative cancer and fibroadenoma. (a) Contrast-enhanced MR image (26/6.5, 30° flip angle, 1.5 T) and (b) static intensity variance feature map at peak enhancement in 44-year-old woman with triple-negative cancer. (c) Contrast-enhanced MR image (26/6.42; flip angle, 30°; 1.5 T) and (d) static intensity variance feature map at peak enhancement in 51-year-old woman with fibroadenoma.

Figure 1b:

Comparison of enhanced appearance and textural features of triple-negative cancer and fibroadenoma. (a) Contrast-enhanced MR image (26/6.5, 30° flip angle, 1.5 T) and (b) static intensity variance feature map at peak enhancement in 44-year-old woman with triple-negative cancer. (c) Contrast-enhanced MR image (26/6.42; flip angle, 30°; 1.5 T) and (d) static intensity variance feature map at peak enhancement in 51-year-old woman with fibroadenoma.

Figure 1c:

Comparison of enhanced appearance and textural features of triple-negative cancer and fibroadenoma. (a) Contrast-enhanced MR image (26/6.5, 30° flip angle, 1.5 T) and (b) static intensity variance feature map at peak enhancement in 44-year-old woman with triple-negative cancer. (c) Contrast-enhanced MR image (26/6.42; flip angle, 30°; 1.5 T) and (d) static intensity variance feature map at peak enhancement in 51-year-old woman with fibroadenoma.

Figure 1d:

Comparison of enhanced appearance and textural features of triple-negative cancer and fibroadenoma. (a) Contrast-enhanced MR image (26/6.5, 30° flip angle, 1.5 T) and (b) static intensity variance feature map at peak enhancement in 44-year-old woman with triple-negative cancer. (c) Contrast-enhanced MR image (26/6.42; flip angle, 30°; 1.5 T) and (d) static intensity variance feature map at peak enhancement in 51-year-old woman with fibroadenoma.

Figure 2a:

Comparison of dynamic enhanced appearance and textural discrimination between triple-negative cancer (TN) in 52-year-old woman and ER-positive cancer in 55-year-old woman. (a) Normalized signal intensity–time curves and (b) normalized mean x-direction Sobel filter map. Note that the two curve shapes are more distinct in the textural kinetic plot. Images associated with this plot can be found in Appendix E1 (online).

Figure 2b:

Comparison of dynamic enhanced appearance and textural discrimination between triple-negative cancer (TN) in 52-year-old woman and ER-positive cancer in 55-year-old woman. (a) Normalized signal intensity–time curves and (b) normalized mean x-direction Sobel filter map. Note that the two curve shapes are more distinct in the textural kinetic plot. Images associated with this plot can be found in Appendix E1 (online).

Figure 3:

Comparison of dynamic enhanced appearance and textural kinetic discrimination between triple-negative cancer (TN) in 38-year-old woman and HER2-positive cancer in 61-year-old woman. Graphs show (a) normalized mean lesion contrast energy and (b) normalized mean lesion image intensity versus normalized time. Note that the time course of relative signal intensities is similar, whereas the time course of relative contrast energy varies widely between the two tumors.

Figure 3b:

Comparison of dynamic enhanced appearance and textural kinetic discrimination between triple-negative cancer (TN) in 38-year-old woman and HER2-positive cancer in 61-year-old woman. Graphs show (a) normalized mean lesion contrast energy and (b) normalized mean lesion image intensity versus normalized time. Note that the time course of relative signal intensities is similar, whereas the time course of relative contrast energy varies widely between the two tumors.