Optimization of Region of Interest Drawing for Quantitative Analysis: Differentiation Between Benign and Malignant Breast Lesions on Contrast-Enhanced Sonography

J Ultrasound Med. 2015 Nov;34(11):1969-76. doi: 10.7863/ultra.14.10042. Epub 2015 Sep 18.

Abstract

Objectives: This study was performed to evaluate the diagnostic utility of quantitative analysis of benign and malignant breast lesions using contrast-enhanced sonography.

Methods: Contrast-enhanced sonography using the perflubutane-based contrast agent Sonazoid (Daiichi Sankyo, Tokyo, Japan) was performed in 94 pathologically proven palpable breast mass lesions, which could be depicted with B-mode sonography. Quantitative analyses using the time-intensity curve on contrast-enhanced sonography were performed in 5 region of interest (ROI) types (manually traced ROI and circular ROIs of 5, 10, 15, and 20 mm in diameter). The peak signal intensity, initial slope, time to peak, positive enhancement integral, and wash-out ratio were investigated in each ROI.

Results: There were significant differences between benign and malignant lesions in the time to peak (P < .05), initial slope (P < .001), and positive enhancement integral (P < .05) for the manual ROI. Significant differences were found between benign and malignant lesions in the time to peak (P < .05) for the 5-mm ROI; the time to peak (P < .05) and initial slope (P< .05) for the 10-mm ROI; absolute values of the peak signal intensity (P< .05), time to peak (P< .01), and initial slope (P< .005) for the 15-mm ROI; and the time to peak (P < .05) and initial slope (P < .05) for the 20-mm ROI. There were no statistically significant differences in any wash-out ratio values for the 5 ROI types.

Conclusions: Kinetic analysis using contrast-enhanced sonography is useful for differentiation between benign and malignant breast lesions.

Keywords: breast lesions; breast ultrasound; contrast-enhanced sonography; kinetic parameters; quantitative analysis; receiver operating characteristic curve.

MeSH terms

  • Adult
  • Aged
  • Algorithms
  • Breast Neoplasms / diagnostic imaging*
  • Contrast Media
  • Diagnosis, Differential
  • Female
  • Ferric Compounds*
  • Humans
  • Image Enhancement / methods*
  • Image Interpretation, Computer-Assisted / methods*
  • Iron*
  • Male
  • Middle Aged
  • Observer Variation
  • Oxides*
  • Pattern Recognition, Automated / methods*
  • Reproducibility of Results
  • Sensitivity and Specificity
  • Ultrasonography, Mammary / methods*

Substances

  • Contrast Media
  • Ferric Compounds
  • Oxides
  • Sonazoid
  • Iron