Quantitative 3D subharmonic imaging for characterizing breast lesions

Anush Sridharan, John R. Eisenbrey, Maria Stanczak, Priscilla Machado, Annina Wilkes, Alexander Sevrukov, Haydee Ojeda-Fournier, Robert F. Mattrey, Kirk Wallace, Flemming Forsberg

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The ability to visualize breast lesion vascularity and quantify the vascular heterogeneity using contrast-enhanced 3-D nonlinear ultrasound imaging was investigated in a clinical population. Patients (n = 236) identified with breast lesions on mammography were scanned using power Doppler imaging, contrast-enhanced 3D HI, and 3D SHI on a modified Logiq 9 scanner (GE Healthcare). Time-intensity curve volumes were developed corresponding to ultrasound contrast agent flow in the lesions after being identified in 4DView (GE Medical Systems). Time-points corresponding to baseline, peak intensity and complete washout of contrast were identified to generate vascular heterogeneity plots of the lesion volume (in the center and periphery as well as the ratio of the two). Vascularity was observed with power Doppler imaging in 93 lesions (69 benign and 24 malignant). The 3D HI showed flow in 8 lesions (5 benign and 3 malignant), whereas 3D SHI visualized flow in 83 lesions (58 benign and 25 malignant). Parametric volumes, that contained a single parametric value for every voxel within the 3D volume, were generated based on perfusion (PER) and area under the curve (AUC). ROC analysis and reverse, step-wise logistical regression were used to assess diagnostic accuracy with biopsy results as the reference. Analysis of vascular heterogeneity in the 3D SHI volumes found benign lesions having a significant difference in vascularity between central and peripheral sections (1.8 ± 0.16 vs. 1.2 ± 0.09 dB, p = 0.0003, respectively), whereas malignant lesions showed no difference (1.7 ± 0.33 vs. 1.3 ± 0.21 dB, p = 0.23), indicative of more vascular coverage. Diagnostic accuracy (i.e., area under the ROC curve) for heterogeneity, PER and AUC ranged from 0.52 to 0.75. The best logistical regression model (heterogeneity ratio, PER central and AUC central) achieved an area of 0.88. In conclusion, 3D SHI is able to detect UCA flow in vascular breast masses. Evaluation of vascular heterogeneity and parametric maps suggests such quantitative parameters might aid in the characterization of breast lesions.

Original languageEnglish (US)
Title of host publication2017 IEEE International Ultrasonics Symposium, IUS 2017
PublisherIEEE Computer Society
ISBN (Electronic)9781538633830
DOIs
StatePublished - Oct 31 2017
Event2017 IEEE International Ultrasonics Symposium, IUS 2017 - Washington, United States
Duration: Sep 6 2017Sep 9 2017

Other

Other2017 IEEE International Ultrasonics Symposium, IUS 2017
CountryUnited States
CityWashington
Period9/6/179/9/17

    Fingerprint

Keywords

  • Breast cancer
  • Microbubbles
  • Subharmonic imaging
  • Ultrasound
  • Ultrasound contrast agents

ASJC Scopus subject areas

  • Acoustics and Ultrasonics

Cite this

Sridharan, A., Eisenbrey, J. R., Stanczak, M., Machado, P., Wilkes, A., Sevrukov, A., Ojeda-Fournier, H., Mattrey, R. F., Wallace, K., & Forsberg, F. (2017). Quantitative 3D subharmonic imaging for characterizing breast lesions. In 2017 IEEE International Ultrasonics Symposium, IUS 2017 [8092623] IEEE Computer Society. https://doi.org/10.1109/ULTSYM.2017.8092623