Semi-quantitative strain ratio in the differential diagnosis of breast masses: Measurements using one region-of-interest

Jung Hyun Yoon, Mi Kyung Song, Eun Kyung Kim

Research output: Contribution to journalArticlepeer-review

8 Citations (Scopus)


The purpose of this study was to evaluate the diagnostic performances of semi-quantitative strain ratio measured by using one region-of-interest (ROI) on breast US elastography images. Two hundred one breast masses of 165 women (mean age: 47.2 y) were included. Ultrasonography (US) and elastography images of the masses were obtained and prospectively analyzed according to elasticity pattern, strain ratio, and final Breast Imaging-Reporting and Data System (BI-RADS) assessments. Of the 201 breast masses, 127 (63.2%) were benign and 74 (36.8%) were malignant. Elastography pattern and strain ratio (cut-off of 1.44) had significantly higher specificity than gray-scale US, 39.4% and 61.4% versus 29.1%, respectively (all p < 0.05). Area under the receiver operating characteristics curve (Az) was highest for gray-scale US (0.646), without statistical significances, than for elastography pattern (0.596, p = 0.159) or strain ratio (0.625, p = 0.610). Semi-quantitative strain ratio measured from one ROI has comparable diagnostic performances to gray-scale US, which may contribute to more accurate differential diagnosis of breast masses seen on US.

Original languageEnglish
Pages (from-to)1800-1806
Number of pages7
JournalUltrasound in Medicine and Biology
Issue number8
Publication statusPublished - 2016 Aug 1

Bibliographical note

Funding Information:
This study has been supported by a research fund from Samsung Medison .

Publisher Copyright:
© 2016 World Federation for Ultrasound in Medicine & Biology.

All Science Journal Classification (ASJC) codes

  • Biophysics
  • Radiological and Ultrasound Technology
  • Acoustics and Ultrasonics


Dive into the research topics of 'Semi-quantitative strain ratio in the differential diagnosis of breast masses: Measurements using one region-of-interest'. Together they form a unique fingerprint.

Cite this