Format

Send to

Choose Destination
Eur Radiol. 2018 Aug;28(8):3165-3175. doi: 10.1007/s00330-017-5287-9. Epub 2018 Mar 19.

Breast-density assessment with hand-held ultrasound: A novel biomarker to assess breast cancer risk and to tailor screening?

Author information

1
Computer-assisted Applications in Medicine, ETH Zurich, Sternwartstrasse 7, 8092, Zürich, Switzerland.
2
Institute of Diagnostic and Interventional Radiology, University Hospital Zurich, Rämistrasse 100, 8091, Zürich, Switzerland.
3
Department of Radiology, Kantonsspital Baden, Im Ergel 1, 5404, Baden, Switzerland.
4
Institute of Diagnostic and Interventional Radiology, University Hospital Zurich, Rämistrasse 100, 8091, Zürich, Switzerland. Marga.Rominger@usz.ch.

Abstract

OBJECTIVES:

To assess feasibility and diagnostic accuracy of a novel hand-held ultrasound (US) method for breast density assessment that measures the speed of sound (SoS), in comparison to the ACR mammographic (MG) categories.

METHODS:

ACR-MG density (a=fatty to d=extremely dense) and SoS-US were assessed in the retromamillary, inner and outer segments of 106 women by two radiographers. A conventional US system was used for SoS-US. A reflector served as timing reference for US signals transmitted through the breasts. Four blinded readers assessed average SoS (m/s), ΔSoS (segment-variation SoS; m/s) and the ACR-MG density. The highest SoS and ΔSoS values of the three segments were used for MG-ACR whole breast comparison.

RESULTS:

SoS-US breasts were examined in <2 min. Mean SoS values of densities a-d were 1,421 m/s (SD 14), 1,432 m/s (SD 17), 1,448 m/s (SD 20) and 1,500 m/s (SD 31), with significant differences between all groups (p<0.001). The SoS-US comfort scores and inter-reader agreement were significantly better than those for MG (1.05 vs. 2.05 and 0.982 vs. 0.774; respectively). A strong segment correlation between SoS and ACR-MG breast density was evident (rs=0.622, p=<0.001) and increased for full breast classification (rs=0.746, p=<0.001). SoS-US allowed diagnosis of dense breasts (ACR c and d) with sensitivity 86.2 %, specificity 85.2 % and AUC 0.887.

CONCLUSIONS:

Using hand-held SoS-US, radiographers measured breast density without discomfort, readers evaluated measurements with high inter-reader agreement, and SoS-US correlated significantly with ACR-MG breast-density categories.

KEY POINTS:

• The novel speed-of-sound ultrasound correlated significantly with mammographic ACR breast density categories. • Radiographers measured breast density without women discomfort or radiation. • SoS-US can be implemented on a standard US machine. • SoS-US shows potential for a quantifiable, cost-effective assessment of breast density.

KEYWORDS:

Breast; Breast density; Breast neoplasms; Mammography; Ultrasonography

PMID:
29556766
DOI:
10.1007/s00330-017-5287-9
[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Springer
Loading ...
Support Center