Published March 2012 | Version v1
Journal article

One-to-one comparison between digital spot compression view and digital breast tomosynthesis

  • 1. University of Genoa, Institute of Anatomy, Department of Experimental Medicine, Genoa (Italy)
  • 2. University of Genoa, Radiology Department, Genoa (Italy)
  • 3. IRCCS Azienda Ospedaliera Universitaria San Martino - IST-Istituto Nazionale per la Ricerca sul Cancro, Department of Medical Physics, Genova (Italy)
  • 4. IRCCS Azienda Ospedaliera Universitaria San Martino - IST-Istituto Nazionale per la Ricerca sul Cancro, UOC Senologia Diagnostica, Genova (Italy)

Description

To assess if digital breast tomosynthesis (DBT) is at least equal to digital spot compression view (DSCV). Following institutional approval and written informed consent, both DBT and DSCV were obtained in women with a screening abnormality. The diagnostic accuracy of DBT and DSCV was evaluated by two radiologists of varying experience (Reader1 and Reader2). 52 consecutive recalled women without calcification (mean age: 51 ± 12 years) underwent DSCV and DBT. Overall sensitivity was equal for both techniques (100% [95% CI, 91-100%] for DBT and 100% [95% CI, 91-100%] for DSCV). Overall specificity was higher for DBT (100% [95%CI, 91-100%]) than for DSCV (94% [95% CI, 91-100%]). Specificity for DSCV was higher for Reader1 (95% [95% CI, 91-100%]). Reader2 had lower values of specificity (92% [95% CI, 90-92%]). On DSCV, three and two false positives were recorded by Reader2 and Reader1, respectively. Overall, the area under the curve (AUC) was greater for DBT (AUC = 1) than for DSCV (AUC = 0.963). The mean difference between the two techniques was not significantly different (P = 0.43). In this dataset, diagnostic accuracy of digital breast tomosynthesis is at least equal to that of digital spot compression. (orig.)

Availability note (English)

Available from: http://dx.doi.org/10.1007/s00330-011-2305-1

Additional details

Identifiers

Publishing Information

Journal Title
European Radiology
Journal Volume
22
Journal Issue
3
Journal Page Range
p. 539-544
ISSN
0938-7994
CODEN
EURAE3