Accession Number : ADA303517

Title :   Computer-Aided Diagnosis and Automated Screening of Digital Mammograms.

Descriptive Note : Annual rept. 1 Sep 94-31 Aug 95,

Corporate Author : UNIVERSITY OF SOUTH FLORIDA TAMPA

Personal Author(s) : Woods, Kevin S.

PDF Url : ADA303517

Report Date : 29 SEP 1995

Pagination or Media Count : 39

Abstract : The primary objective of the proposed research is to develop computer techniques, methods and rigorous testing to provide reliable results which can be used by radiologists to increase diagnostic accuracy and decrease the rate of undetected cancer. First, "basic' level computer science algorithms and concepts have been developed and refined. These research topics include development of a novel algorithm for generating ROC curves for artificial neural networks, a method for multiple classifiers. These algorithms are fundamental building blocks to be used in the construction of the final system. The second general area of research is more application-specific. The topic of this research has been to develop reliable routines for segmenting mammographic abnormalities using texture features, as opposed to contrast or pixel intensity features.

Descriptors :   *DIAGNOSIS(MEDICINE), *COMPUTER AIDED DIAGNOSIS, *SCREENS(DISPLAYS), *BREAST CANCER, *MAMMOGRAPHY, ALGORITHMS, METHODOLOGY, NEURAL NETS, AUTOMATION, COMPUTERS, INTENSITY, ACCURACY, MODULAR CONSTRUCTION, RELIABILITY, COMPUTER APPLICATIONS, TEXTURE, ABNORMALITIES.

Subject Categories : Medicine and Medical Research
      Medical Facilities, Equipment and Supplies
      Cybernetics

Distribution Statement : APPROVED FOR PUBLIC RELEASE