Accession Number : ADA292689
Title : Implementation of Computer Assisted Breast Cancer Diagnosis.
Descriptive Note : Annual rept. 1 Dec 93-30 Nov 94,
Corporate Author : GEORGETOWN UNIV WASHINGTON DC
Personal Author(s) : Lo, Shih-Chung B.
PDF Url : ADA292689
Report Date : 30 DEC 1994
Pagination or Media Count : 14
Abstract : This project aims at the implementation of a computer-aided diagnosis system for the detection of microcalcifications on mammograms based on the algorithms developed by the principal investigator and others. In addition, the proposed research includes: (1) algorithm improvement for the detection of microcalcifications, (2) mammographic image compression and its impact on computer-aided diagnosis (CADx), and (3) computer-aided classification of benign and malignant masses on mammograms. In the past year, we have developed several algorithms and have studied part of the proposed research: (a) development of filtering techniques with wavelet transform to reduce mammographic structures other than microcalcifications, (b) performance of preliminary study in the detection of microcalcifications, (c) performance of mammographic compression studies using split gray values in conjunction with wavelet and full-frame discrete cosine transform (DCT) techniques, (d) evaluation of the impact of the compression with respect to various degrees of data compression, and (f) implementation of CADx system in a DEC Alpha workstation.
Descriptors : *DIAGNOSIS(MEDICINE), *COMPUTER AIDED DIAGNOSIS, *CANCER, COMPRESSION, ALGORITHMS, DETECTION, IMPACT, IMAGES, CLASSIFICATION, DIAGNOSIS(GENERAL), COMPUTER APPLICATIONS, FRAMES, VALUE, FILTERS, SPLITTING, DISCRETE FOURIER TRANSFORMS, GRAY(COLOR).
Subject Categories : Medicine and Medical Research
Distribution Statement : APPROVED FOR PUBLIC RELEASE