Classification of mammographic microcalcification clusters

Author

Aghdasi, F. ; Ward, R.K. ; Palcic, B.

Author_Institution

Dept. of Electr. Eng., British Columbia Univ., Vancouver, BC, Canada

fYear

1993

fDate

14-17 Sep 1993

Firstpage

1196

Abstract

Microcalcifications are non-specific indicators of breast carcinoma in conventional mammographic examinations. To assist in improved diagnostic accuracy the authors evaluated over 100 numerical features extracted from microcalcifications in digitized mammographic images. These features quantify the number, size, shape, roughness, and configuration of clusters of microcalcifications. Preliminary results indicate that a small subset of features can achieve acceptable classification accuracies

Keywords

diagnostic radiography; medical image processing; breast carcinoma; classification accuracy; diagnostic accuracy; digitized mammographic images; features subset; mammographic microcalcification clusters classification; medical diagnostic imaging; nonspecific indicators; numerical features; Breast cancer; Feature extraction; Graphics; Histograms; Image segmentation; Optical films; Optical variables control; Pathology; Physics; Shape;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical and Computer Engineering, 1993. Canadian Conference on

Conference_Location

Vancouver, BC

Print_ISBN

0-7803-2416-1

Type

conf

DOI

10.1109/CCECE.1993.332470

Filename

332470

Link To Document

https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=2173516