Model and Expansion Based Methods of Detection of Small Masses in Radiographs of Dense Breasts
Abstract
This report describes progress made in during the second year of study. Our goal is to detect masses in dense mammograms having a diameter less than 1 cm. The idea of this project is to detect subtle masses by tuning the central frequency and width of a basis function used in an overcomplete expansion. By modeling the shape of a mass through this flexibility we hope to detect small and subtle masses in dense breasts and improve the chances of early detection in screening mammography. This phase of our study showed that fractional splines functions are a powerful tool for the representation of masses in mammograms and are well matched to the mass detection problem. We showed that by using a continuously varying parameter of order, that accurate approximations of mass shapes could be obtained through overcomplete expansions of a fractional spline wavelet transform. In the context of addressing the problem of finding the best scale, using a library of bases computed by wavelet packets was an efficient method in finding the best scale. Finally, we ported this algorithm using the Matlab compiler to allow integration into our Mammography Computer Aided Diagnosis (CAD) workstation for real-time interactive screening of mammography cases during the third year.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2001
- Accession Number
- ADA398954
Entities
People
- Andrew F. Laine
Organizations
- Columbia University