An Apparent Diffusion Coefficient Histogram Method Versus a Traditional 2-Dimensional Measurement Method for Identifying Non-Puerperal Mastitis From Breast Cancer at 3.0 T

J Comput Assist Tomogr. 2018 Sep/Oct;42(5):776-783. doi: 10.1097/RCT.0000000000000758.

Abstract

Purpose: This study aimed to investigate the utility of a volumetric apparent diffusion coefficient (ADC) histogram method for distinguishing non-puerperal mastitis (NPM) from breast cancer (BC) and to compare this method with a traditional 2-dimensional measurement method.

Materials and methods: Pretreatment diffusion-weighted imaging data at 3.0 T were obtained for 80 patients (NPM, n = 27; BC, n = 53) and were retrospectively assessed. Two readers measured ADC values according to 2 distinct region-of-interest (ROI) protocols. The first protocol included the generation of ADC histograms for each lesion, and various parameters were examined. In the second protocol, 3 freehand (TF) ROIs for local lesions were generated to obtain a mean ADC value (defined as ADC-ROITF). All of the ADC values were compared by an independent-samples t test or the Mann-Whitney U test. Receiver operating characteristic curves and a leave-one-out cross-validation method were also used to determine diagnostic deficiencies of the significant parameters.

Results: The ADC values for NPM were characterized by significantly higher mean, 5th to 95th percentiles, and maximum and mode ADCs compared with the corresponding ADCs for BC (all P < 0.05). However, the minimum, skewness, and kurtosis ADC values, as well as ADC-ROITF, did not significantly differ between the NPM and BC cases.

Conclusions: Thus, the generation of volumetric ADC histograms seems to be a superior method to the traditional 2-dimensional method that was examined, and it also seems to represent a promising image analysis method for distinguishing NPM from BC.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Aged
  • Algorithms
  • Breast / diagnostic imaging
  • Breast Neoplasms / diagnostic imaging*
  • Diagnosis, Differential
  • Diffusion Magnetic Resonance Imaging / methods*
  • Female
  • Humans
  • Image Interpretation, Computer-Assisted / methods*
  • Mastitis / diagnostic imaging*
  • Middle Aged
  • Reproducibility of Results
  • Retrospective Studies
  • Sensitivity and Specificity
  • Young Adult