This poster is published under an
open license. Please read the
disclaimer for further details.
Keywords:
Breast, Ultrasound, Biopsy, Education, Education and training
Authors:
P. Valkovic Zujic, T. Grubesic, D. Dodig, D. Miletic; Rijeka/HR
DOI:
10.1594/ecr2015/C-0496
Methods and materials
Three hundred sixty-seven ultrasound - guided core needle biopsies were performed between October 2012 and April 2014,
at the Department of Radiology,
Clinical Hospital Center Rijeka,
Croatia.
One hundred fifty-nine lesions were documented prior to biopsy as hard-copy images in a jpeg.
format.
One hundred thirty-four lesions that corresponded to a solid mass visible in two different planes were included in the study.
Dedicated breast radiologist and radiology resident with no prior experience in breast imaging,
independently reviewed all US-images; each separate lesion was described using the features from the sonographic BI-RADS lexicon (4th edition).
In our study,
we evaluated a total of seven mass descriptors.
The shape of the mass was described as oval,
round or irregular (Fig.
1-3).
The lesion orientation was described as parallel to the skin surface (horizontal orientation) or not parallel (vertical orientation) (Fig.
4,
5).
The mass margins were described as circumscribed,
microlobulated or spiculated (Fig.
6-8).
Transition zone was described as clear or if echogenic hallo was present (Fig.
9).
The echo pattern of the mass was described as anechoic,
hypoechoic,
hyperechoic,
isoechoic or heterogeneous (Fig.
10-14).
Posterior acoustic features were described as absent,
enhanced,
shadowing,
or a combined pattern of the latter two (Fig.
15-18).
Surrounding parenchyma was evaluated as normal or distorted (Fig.
19,
20).
The radiologists were blinded to the histologic diagnosis during the evaluation.
To determine the interobserver variability,
Cohen’s kappa statistic was calculated for all mass descriptors.
After result analysis we report the appropriate PABAK (Prevalence and Bias Adjusted Kappa) value.