Share Email Print
cover

Proceedings Paper

Three-dimensional imaging of breast calcifications
Author(s): Andrew D. A. Maidment; Michael Albert; Emily F. Conant
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

Approximately 50 percent of breast cancers are detected on the basis of calcifications alone. Regrettably, the presence of such calcifications is non-specific; only 30 percent of biopsies based on suspicious calcifications are malignant. We have investigated three methods (LVR) for 3D imaging and analysis of microcalcifications. Our aim is to increase specificity by more accurately distinguishing between calcifications indicative of benign and malignant breast lesions. We have demonstrated that 3D imaging of calcifications is possible using an LVR technique that includes semi-automated segmentation, correlation, and reconstruction of the calcifications. A clinical study of he LVR method is ongoing in which 2D film and digital images are compared to 3D images. The images are evaluated using a rating of 1 to 5, where 1 equals definitely benign, 5 equals definitely malignant, and a score of 3 or higher requires biopsy. To date, 3 radiologists have evaluated the images of 44 patients for which biopsy results were available. The use of 2D and 3D digital images resulted in doubling the diagnostic accuracy from 36 percent to 77 percent. Comparison to other techniques is ongoing. Additionally, a high resolution CT scanner for breast tissue specimens is under construction for comparison of the reconstructed images to a 'gold standard'.

Paper Details

Date Published: 1 March 1998
PDF: 9 pages
Proc. SPIE 3240, 26th AIPR Workshop: Exploiting New Image Sources and Sensors, (1 March 1998); doi: 10.1117/12.300057
Show Author Affiliations
Andrew D. A. Maidment, Thomas Jefferson Univ. (United States)
Michael Albert, Thomas Jefferson Univ. (United States)
Emily F. Conant, Thomas Jefferson Univ. (United States)


Published in SPIE Proceedings Vol. 3240:
26th AIPR Workshop: Exploiting New Image Sources and Sensors
J. Michael Selander, Editor(s)

© SPIE. Terms of Use
Back to Top