An Intelligent-Based Wavelet Classifier for Accurate Prediction of Breast Cancer

Introduction

Breast cancer is presently a standout the most well-known sicknesses among ladies in the both developed and developing nations. In the real situation, one out of 500 ladies will get breast cancer sooner or later in her life (Ramón González, Moreno, Fernández, Izquierdo, Borrás, & Gispert, 2005). It’s recorded that 23% cancer cases and 14% lead to death. It is evaluated that more than 1.6 million new instances of breast cancer disease happened among ladies worldwide in 2010 (Jemal, Bray, Forman, O’Brien, Ferlay, Center, & Parkin, 2012). In 2011, about 1.7 million individuals were risked having breast malignancy; in that 527 new USA patients of breast cancer were analyzed every day and 110 individuals were die in every day. Early detection stays vital for survival, especially in low and moderate nations where the sicknesses are analyzed in late stages and assets are extremely constrained. One demonstrated method for diminishing mortality from bosom tumor is the screening of asymptotic ladies by mammography.

Figure 1.

Mammogram of the breast

Mammography is the best screening method that utilizations low amount X-ray beams to make a picture of the breast to discover breast tumor. The screened Images are shown in Figure 1. Mammography has been successful in screening asymptomatic ladies. The Cancer Society prescribes that all ladies who completed 40 years should go screening mammography once in a year. Thick breast tissue can look white or light dots on a mammogram. This can make mammograms harder to decipher in younger ladies, who have a tendency to have denser breast shown in Figure 2.

Figure 2.

a) Fatty and b) Dense tissue

Because of the thickness of the breasts radiologists may miss up to 30% of malignant tissue that cause cancer. Fortunately, even well trained radiologists may felt difficulties in diagnosing mammograms. Two capable markers of malignancy that are generally utilized as a part of assessing mammograms .There are masses and micro calcifications. The first method diagnosing breast using masses in computerized mammography is more complex. This done by using feature selection and ensemble methods. Later technique is calcification in which complexity lesser than mass analysis. CAD (Computer Aided Diagnosis) of Micro calcifications in Digital Mammograms was used to support Early Diagnosis of Breast cancer.

Figure 3.

Identification of calcifications area in breast through CAD

In order to support the radiologists in finding accurate cancer tissue, Computer Aided Diagnosis is used. Computer aided design can discover tumors that are invisible to radiologist. After CAD analysis as shown in Figure 3, radiologist will do a visual check of those abnormalities and able to separate cancer tissues from genuine one.

Related Works

The second leading cause of death for women in the world is Breast cancer. It represents about 22% of new cancer cases. As per the World Health Organization report, breast cancer causes 450,000 deaths worldwide every year. Although the incidence of breast cancer all over the world has increased over the past decade, mortality rate has declined among women of all ages. It is due to the widespread adoption of the technique of mammography screening, and the significant improvements made in breast cancer treatment has paved the way for the reduction of mortality rates (Anandakumar & Umamaheswari, 2017). Mammography has been proved as the best method of screening which has reduced the mortality rates by 30-70%. Computer Aided Diagnosis (CAD) systems can aid the diagnosis of the disease by increasing the sensitivity to abnormalities when compared to the double checking that would be performed by the radiologists. These systems have proved to be cost-effective and efficient and also they are available commercially.