Early Detection of Malignancy in Human Breasts Using Deep Learning

Breast cancer is the fitful growth of cells that originate in the breast tissue. Breast

cancer is any form of malignancy which develops from breast cells. Cancer is a disease in

which malignancy cells in the body grow, change and multiply out of control. It is the most

common cancer disease in women worldwide, especially those who are in the age group of 40

to 55 years and one of the leading causes of death from cancer in women. Breast cancer is

reported as the second most prevalent cause of the cancerous death in the report of World

Health Organization]. Breast cancer is"the second most common cancer worldwide. There

were over 2 million new cases reported in 2018.

The recent statistics of a study by American Cancer Society gives an alarming message

that about 3,30,080 new cases of breast cancer are expected to be diagnosed in women in the U.S.

and about 2,550 new cases of invasive breast cancer are expected to be diagnosed in men in 2018.

These new cases of breast cancer in women are further classified as 2,66,120 new cases of invasive

and 63,960 new cases of non-invasive breast cancer. A man's lifetime risk of breast cancer is about

1 in 1,000 whereas 1 in 8 for women which is about 12.4% chances of having breast cancer over

the course of the lifetime . The worldwide cancer statistics report says breast cancer

is the third most common cancer in the world in terms of number of new cases .

In India, there are around 2.5 million diagnosed with cancer and will lead to 5,56,600

deaths per year. By 2020, India is reported to have over 17.5 lakhs diagnosed patients with

breast cancer and near 8.8 lakhs deaths caused by this disease . As reported in literature,

most prevailing cancer in India is breast cancer and its explanation is a 27% count in all types

of cancer in women.

Early detection is an exhausting effort and a nerve center on cancer detection. Imaging modalities and

advancements in technology disported a far-reaching role in early detection [4]. Mammography

has been extensively recognized as a diagnostic and shielding weapon for an early detection of

breast cancer and even up to ten years before detection and diagnosis stage. Mammography can

curtail the number of mortality native to breast cancer, particularly among women in an age group

of years 40-70. Mammography implies the assimilation of anomalies such as calcifications and masses to detect breast cancer. It makes image analysis as a thrust area of research for an early detection of breast cancer.

The enactment of a Computer Aided Diagnosis (CAD) system can revamp the early cancer

detection rate and could also share the experts workload [6]. A recent report manifested that the

enactment of CAD refined the process of breast cancer detection by 7.62% and this automated or

semi-automated gizmo has captivated great considerations [6]. Many Computer Aided Diagnosis

systems prospected to assist radiologists to curtail the diagnostic errors effectively by using

mammogram and MRI imaging modality of breast mass or calcification [33] [26]. Computer

Aided Diagnosis systems substantiated a probably compelling tool for detecting the masses or

micro calcifications in recent considerations