A multi-scale approach to the computer-aided detection of microcalcification clusters in digital mammograms

摘要

A computer-aided detection (CADe) system for the identification ofmicrocalcification clusters in digital mammograms has been developed. It ismainly based on the application of wavelet transforms for image filtering andneural networks for both the feature extraction and the classificationprocedures. This CADe system is easily adaptable to different databases. We report andcompare the FROC curves obtained on the private database we used for developingthe CADe system and on the publicly available MIAS database. The resultsachieved on the two databases show the same trend, thus demonstrating the goodgeneralization capability of the system.