K Skerl1, S Vinnicombe2, E Giannotti3, K Thomson2, A Evans2. 1. Medical Research Institute, Ninewells Hospital & Medical School, Mailbox 4, Dundee DD1 9SY, UK. Electronic address: k.skerl@dundee.ac.uk. 2. Medical Research Institute, Ninewells Hospital & Medical School, Mailbox 4, Dundee DD1 9SY, UK. 3. Medical Research Institute, Ninewells Hospital & Medical School, Mailbox 4, Dundee DD1 9SY, UK; Department of Biomedical Sciences, Experimental and Clinical "Mario Serio" University of Florence, Viale Morgagni 50, 50134 Firenze (FI), Italy.
Abstract
AIM: To evaluate the influence of the region of interest (ROI) size and lesion diameter on the diagnostic performance of 2D shear wave elastography (SWE) of solid breast lesions. MATERIALS AND METHODS: A study group of 206 consecutive patients (age range 21-92 years) with 210 solid breast lesions (70 benign, 140 malignant) who underwent core biopsy or surgical excision was evaluated. Lesions were divided into small (diameter <15 mm, n=112) and large lesions (diameter ≥15 mm, n=98). An ROI with a diameter of 1, 2, and 3 mm was positioned over the stiffest part of the lesion. The maximum elasticity (Emax), mean elasticity (Emean) and standard deviation (SD) for each ROI size were compared to the pathological outcome. Statistical analysis was undertaken using the chi-square test and receiver operating characteristic (ROC) analysis. RESULTS: The ROI size used has a significant impact on the performance of Emean and SD but not on Emax. Youden's indices show a correlation with the ROI size and lesion size: generally, the benign/malignant threshold is lower with increasing ROI size but higher with increasing lesion size. CONCLUSIONS: No single SWE parameter has superior performance. Lesion size and ROI size influence diagnostic performance.
AIM: To evaluate the influence of the region of interest (ROI) size and lesion diameter on the diagnostic performance of 2D shear wave elastography (SWE) of solid breast lesions. MATERIALS AND METHODS: A study group of 206 consecutive patients (age range 21-92 years) with 210 solid breast lesions (70 benign, 140 malignant) who underwent core biopsy or surgical excision was evaluated. Lesions were divided into small (diameter <15 mm, n=112) and large lesions (diameter ≥15 mm, n=98). An ROI with a diameter of 1, 2, and 3 mm was positioned over the stiffest part of the lesion. The maximum elasticity (Emax), mean elasticity (Emean) and standard deviation (SD) for each ROI size were compared to the pathological outcome. Statistical analysis was undertaken using the chi-square test and receiver operating characteristic (ROC) analysis. RESULTS: The ROI size used has a significant impact on the performance of Emean and SD but not on Emax. Youden's indices show a correlation with the ROI size and lesion size: generally, the benign/malignant threshold is lower with increasing ROI size but higher with increasing lesion size. CONCLUSIONS: No single SWE parameter has superior performance. Lesion size and ROI size influence diagnostic performance.
Authors: Max Denis; Adriana Gregory; Mahdi Bayat; Robert T Fazzio; Dana H Whaley; Karthik Ghosh; Sejal Shah; Mostafa Fatemi; Azra Alizad Journal: PLoS One Date: 2016-10-24 Impact factor: 3.240
Authors: Soong June Bae; Jong Tae Park; Ah Young Park; Ji Hyun Youk; Jong Won Lim; Hak Woo Lee; Hak Min Lee; Sung Gwe Ahn; Eun Ju Son; Joon Jeong Journal: J Breast Cancer Date: 2018-06-20 Impact factor: 3.588