Jeffrey A. Belair MD, Presenter: Nothing to Disclose

Margarita Louise Zuley MD, Abstract Co-Author: Research Grant, Hologic, Inc

Marie Adele Ganott MD, Abstract Co-Author: Nothing to Disclose

Amy Elizabeth Kelly MD, Abstract Co-Author: Nothing to Disclose

Dilip Digambar Shinde MD, Abstract Co-Author: Nothing to Disclose

Ratan Shah MD, Abstract Co-Author: Nothing to Disclose

Victor J. Catullo MD, Abstract Co-Author: Nothing to Disclose

Michael Donald Vijay Mishra MD, Abstract Co-Author: Nothing to Disclose

David Gur, Abstract Co-Author: Nothing to Disclose

To compare the ability of non-contrast cone-beam CT (CBCT) to tomosynthesis in the identification and classification of non-calcified breast lesions.

IRB approval was obtained. 30 consecutively accrued biopsy proven malignancies, 10 randomly selected cases with a biopsy proven benign lesion and 20 normal cases that were imaged with both non-contrast CBCT and tomosynthesis were obtained from our research database. Six MQSA qualified breast imagers participated in a fully crossed MRMC retrospective reader study that included evaluation of each case twice, once with CBCT and once with tomosynthesis. Readers gave a probability of malignancy (POM) score on a 0-100 scale, a 7 point BI-RADS classification, and rated overall interpretation confidence for each case on a 5 point scale. Specificity and sensitivity for the detection of malignant lesions was calculated for both modalities. Additionally, average POM scores for all malignant and all benign cases were compared.

Based on preliminary data, the sensitivity and specificity of tomosynthesis for BI-RADS assessment of malignancy was 0.87 (95% confidence interval [CI] = 0.80-0.92) and 0.81 (CI = 0.72-0.87), respectively. The sensitivity and specificity of non-contrast CBCT for BI-RADS assessment was 0.70 (CI = 0.60-0.79) and 0.67 (CI = 0.57-0.77), respectively. Each of the readers had a greater percentage correct for both malignant and benign cases using tomosynthesis versus CBCT. The average POM score using tomosynthesis for all malignant cases was 63.3 (including misinterpretations) and for all benign cases was 8.9. The average POM score using CBCT for all malignant cases was 42.7 and for all benign cases was 15.3. Overall, reader confidence in their diagnoses was higher for both malignant and benign cases using tomosynthesis.

Tomosynthesis is superior to non-contrast CBCT for the identification and classification of non-calcified breast lesions, both malignant and benign. Additionally, readers were more confident characterizing lesions using tomosynthesis compared to CBCT.

Currently, non-contrast CBCT does not appear to have a role in the assessment of non-calcified breast lesions, although advances in technology and reader familiarity may allow for this in the future.

Belair, J, Zuley, M, Ganott, M, Kelly, A, Shinde, D, Shah, R, Catullo, V, Mishra, M, Gur, D, Non-contrast Cone-Beam CT vs Tomosynthesis: Identification and Classification of Benign and Malignant Breast Lesions.  Radiological Society of North America 2012 Scientific Assembly and Annual Meeting, November 25 - November 30, 2012 ,Chicago IL. http://archive.rsna.org/2012/12022690.html