Breast cancer identification based on artificial intelligent system

Abstract

Worldwide, breast cancer causes a high mortality rate. Early diagnosis is important for treatment, but high-density breast tissues are difficult to analyze. Computer-assisted identification systems were introduced to classify by fine-needle aspirates FNA with features that better represent the images to be classified as a major challenge. This work is fully automated, and it does not require any manual intervention from the user. In this analysis, various texture definitions for the portrayal of breast tissue density on mammograms are examined in addition to contrasting them with other techniques. We have created an algorithm that can be divided into three classes: fatty, fatty-glandular, and dense-glandular. The suggested system works in a spatial-related domain and it results in extreme immunity to noise and background area, with a high rate of precision.