Evaluation of quantitative parametric analysis for characterization of breast lesions in contrast-enhanced MR mammography

Eur Radiol. 2006 Dec;16(12):2834-41. doi: 10.1007/s00330-006-0348-5. Epub 2006 Jul 29.

Abstract

Our aim was to evaluate quantitative parametric analysis for characterization of breast lesions in contrast-enhanced magnetic resonance (MR) mammography. In 62 patients, contrast-enhanced MR mammography revealed 75 suspicious lesions, of which 18 were benign and 57 were malignant. The quantitative parametric analysis delineates signal intensity changes of contrast-enhanced lesions on a pixel-by-pixel basis. The initial rate of enhancement is coded by color intensity: a slow rate is coded to dark; a fast rate, bright. The postinitial enhancement change is coded by color hue: blue for increasing signal intensity, green for plateau, and red for decrease in signal intensity. Malignant lesions showed a significantly higher number of bright-red (P = 0.004) and medium-red (P < 0.001) pixels than benign lesions. Benign lesions showed significantly more blue pixels than did malignant lesions (P = 0.010). Of the 75 lesions, 72 (96%) showed heterogeneous distribution of pixel color hue. Quantitative parametric analysis of contrast kinetics in lesions can replace the subjective manual region of interest (ROI) method and makes a step toward standardization of MR mammography. It allows quantitative evaluation of different contrast kinetics parameters in contrast-enhanced breast lesions.

MeSH terms

  • Adult
  • Aged
  • Breast Neoplasms / diagnosis*
  • Breast Neoplasms / pathology
  • Chi-Square Distribution
  • Color
  • Contrast Media / pharmacokinetics*
  • Diagnosis, Differential
  • Female
  • Gadolinium DTPA / pharmacokinetics*
  • Humans
  • Image Processing, Computer-Assisted
  • Magnetic Resonance Imaging / methods*
  • Middle Aged
  • Software
  • Statistics, Nonparametric

Substances

  • Contrast Media
  • Gadolinium DTPA