Open AccessBI-RADS Breast Tumor Classification Through Image Mining
Author(s) -
Valesca J. S. Da Silva,
Mateus M. R. Da Silva,
Marcelino P. S. Silva,
Joaogueira
Publication year - 2019
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/kdmile.2019.8791
Subject(s) - bi rads , mammography , computer science , artificial intelligence , breast cancer , contextual image classification , breast imaging , pattern recognition (psychology) , image (mathematics) , cancer , medicine
In this article, a computer aided diagnostic system for BI-RADS classification of breast cancer is proposed. The approach involves image processing capabilities to extract features from tumors in mammography and image mining to classify them as BI-RADS 2, BI-RADS 3, BI-RADS 4C or BI-RADS 5. Images from the BCDR repository were used for the experiments. The results showed the efficacy of the proposed method, which classified tumors with considerable accuracy in four BI-RADS categories.