Diagnostic value of MR elastography in addition to contrast-enhanced MR imaging of the breast-initial clinical results

Assessment of breast cancer tumour size using six different methods

OBJECTIVES

Tumour size estimates using mammography (MG), conventional ultrasound (US), compound imaging (CI) and real-time elastography (RTE) were compared with histopathological specimen sizes.

METHODS

The largest diameters of 97 malignant breast lesions were measured. Two US and CI measurements were made: US1/CI1 (hypoechoic nucleus only) and US2/CI2 (hypoechoic nucleus plus hyperechoic halo). Measurements were compared with histopathological tumour sizes using linear regression and Bland-Altman plots.

RESULTS

Size prediction was best with ultrasound (US/CI/RTE: R (2) 0.31-0.36); mammography was poorer (R(2) = 0.19). The most accurate method was US2, while US1 and CI1 were poorest. Bland-Altman plots showed better size estimation with US2, CI2 and RTE, with low variation, while mammography showed greatest variability. Smaller tumours were better assessed than larger ones. CI2 and US2 performed best for ductal tumours and RTE for lobular cancers. Tumour size prediction accuracy did not correlate significantly with breast density, but on MG tumours were more difficult to detect in high-density tissue.

CONCLUSIONS

The size of ductal tumours is best predicted with US2 and CI2, while for lobular cancers RTE is best. Hyperechoic tumour surroundings should be included in US and CI measurements and RTE used as an additional technique in the clinical staging process.

Axial-shear strain imaging for differentiating benign and malignant breast masses

Axial strain imaging has been utilized for the characterization of breast masses for over a decade; however, another important feature namely the shear strain distribution around breast masses has only recently been used. In this article, we examine the feasibility of utilizing in vivo axial-shear strain imaging for differentiating benign from malignant breast masses. Radio-frequency data was acquired using a VFX 13-5 linear array transducer on 41 patients using a Siemens SONOLINE Antares real-time clinical scanner at the University of Wisconsin Breast Cancer Center. Free-hand palpation using deformations of up to 10% was utilized to generate axial strain and axial-shear strain images using a two-dimensional cross-correlation algorithm from the radio-frequency data loops. Axial-shear strain areas normalized to the lesion size, applied strain and lesion strain contrast was utilized as a feature for differentiating benign from malignant masses. The normalized axial-shear strain area feature estimated on eight patients with malignant tumors and 33 patients with fibroadenomas was utilized to demonstrate its potential for lesion differentiation. Biopsy results were considered the diagnostic standard for comparison. Our results indicate that the normalized axial-shear strain area is significantly larger for malignant tumors compared with benign masses such as fibroadenomas. Axial-shear strain pixel values greater than a specified threshold, including only those with correlation coefficient values greater than 0.75, were overlaid on the corresponding B-mode image to aid in diagnosis. A scatter plot of the normalized area feature demonstrates the feasibility of developing a linear classifier to differentiate benign from malignant masses. The area under the receiver operator characteristic curve utilizing the normalized axial-shear strain area feature was 0.996, demonstrating the potential of this feature to noninvasively differentiate between benign and malignant breast masses.