Interobserver and intermethod variability in data interpretation of breast strain elastography in suspicious breast...

Area of Science:

• Medical Imaging
• Radiology
• Oncology

Background:

• Strain elastography (SE) is operator-dependent, leading to interobserver variability in image acquisition and interpretation.
• Variability in SE assessment can impact diagnostic accuracy for breast lesions.

Purpose of the Study:

• To analyze interobserver and intermethod variability of SE in breast lesion interpretation.
• To evaluate the diagnostic performance of combining elasticity scores and strain ratio with B-mode ultrasonography (US).

Main Methods:

• Retrospective study of 70 breast lesions evaluated with B-mode US and SE.
• Two radiologists assessed B-mode US findings, elasticity scores, and strain ratio values.
• Comparison of BI-RADS assessments and biopsy decisions with histopathology, analyzing interobserver agreement for combined US+SE and US+SE+strain ratio.

Main Results:

• Very good interobserver agreement for elasticity scores (κ=0.865) and BI-RADS assessments (κ=0.848–0.902).
• High diagnostic performance with AUCs ranging from 0.851 to 0.916 for combined US+SE and US+SE+strain ratio.
• No significant differences in diagnostic performance between observers for any combination.

Conclusions:

• Combining B-mode US, elasticity scoring, and semiquantitative strain ratio calculation improves diagnostic accuracy for breast lesions.
• Strain ratio calculation helps reduce interpretational variety, particularly beneficial for less experienced practitioners.
• The combined approach enhances the evaluation and further assessment of breast lesions.