Synthesising Abnormal Structures in Mammograms Using Pyramid Decomposition

Authors:
Michael Berks;Chris Rose;Caroline Boggis;Susan Astley
Affiliations:
Imaging Science and Biomedical Engineeering, University of Manchester, Manchester, UK M13 9PT;Imaging Science and Biomedical Engineeering, University of Manchester, Manchester, UK M13 9PT;Nightingale Centre, Wythenshawe Hopsital, Manchester, UK;Imaging Science and Biomedical Engineeering, University of Manchester, Manchester, UK M13 9PT
Venue:
IWDM '08 Proceedings of the 9th international workshop on Digital Mammography
Year:
2008

Citing 2
Cited 0

The steerable pyramid: a flexible architecture for multi-scale derivative computation

ICIP '95 Proceedings of the 1995 International Conference on Image Processing (Vol. 3)-Volume 3 - Volume 3
Automated Registration of Volumes of Interest for a Combined X-Ray Tomosynthesis and Ultrasound Breast Imaging System

IWDM '08 Proceedings of the 9th international workshop on Digital Mammography

Quantified Score

Hi-index	0.00

Visualization

Abstract

The appearance of tissue in mammograms is altered by the presence of a mass. Existing mass synthesis methods have not addressed this structural deformation. We present a method for modifying the fine-scale detail in regions of digital mammograms. Regions are decomposed into sub-bands localised in scale and orientation using a wavelet-like decomposition. New structures are synthesised in the high frequency components of the decomposition, leaving coarse levels of the pyramid unchanged. In contrast to existing methods of synthesising mammographic texture, this ensures the global appearance of the region is maintained. Early results in the form of synthesised regions show the promise of this approach.