Page 1 of 12

British Journal of Healthcare and Medical Research - Vol. 9, No. 3

Publication Date: June, 25, 2022

DOI:10.14738/jbemi.93.12535. Alseddiqi, M., AlMannaei, B., Najam, O., Atawi, K., & Al-Mofleh, A. (2022). The Prospective Benefits of Using Machine Learning for

the Prediction of Breast Cancer. British Journal of Healthcare and Medical Research, 9(3). 216-227.

Services for Science and Education – United Kingdom

The Prospective Benefits of Using Machine Learning for the

Prediction of Breast Cancer

Mohamed Alseddiqi

Clinical Engineering Directorate, king Hamad university hospital

Building -2435, Road 2835, Block 228 P.O. Box 24343

Busaiteen Kingdom of Bahrain

Budoor AlMannaei

Clinical Engineering Directorate, king Hamad university hospital

Building -2435, Road 2835, Block 228 P.O. Box 24343

Busaiteen Kingdom of Bahrain

Osama Najam

Clinical Engineering Directorate, king Hamad university hospital

Building -2435, Road 2835, Block 228 P.O. Box 24343

Busaiteen Kingdom of Bahrain

Khamis Atawi

Clinical Engineering Directorate, king Hamad university hospital

Building -2435, Road 2835, Block 228 P.O. Box 24343

Busaiteen Kingdom of Bahrain

Anwar AL-Mofleh

Clinical Engineering Directorate, king Hamad university hospital

Building -2435, Road 2835, Block 228 P.O. Box 24343

Busaiteen Kingdom of Bahrain

ABSTRACT

Improving the percentage of patients diagnosed with early-stage cancer is a vital

priority of the World Health Organization. Cancer is one of the most unsafe diseases

for humans, yet no enduring cure has been developed. Breast cancer is one of the

most common types of cancer in the Middle East region. Early diagnosis and

treatment of breast cancer can significantly improve the lives of millions of women

across the globe. Due to the advancement in technology, artificial intelligence and

machine learning have been used successfully to discover several dangerous

diseases, and serving in early analysis and treatment. Thus, the integration of

artificial intelligence and machine learning in the scientific field supports

enhancing morbity and mortality rates. This research is a systematic review on

breast cancer discovery and action using genetic sequencing or histopathological

imaging with the help of deep learning and machine learning.

Keywords: Breast cancer; breast cancer diagnosis; Artificial intelligence; Machine

Learning

Page 2 of 12

217

Alseddiqi, M., AlMannaei, B., Najam, O., Atawi, K., & Al-Mofleh, A. (2022). The Prospective Benefits of Using Machine Learning for the Prediction of

Breast Cancer. British Journal of Healthcare and Medical Research, 9(3). 216-227.

URL: http://dx.doi.org/10.14738/jbemi.93.12535

INTRODUCTION

Cancer is the main health difficulty in both developed and developing countries. The expected

number of new patients with cancer each year is predicted to increase from 10 million in 2000

to 16 million by 2020[1]. This amount continues to rise every year by 3-4%, and nearly 60% of

these cases will take place in developing countries where healthcare facilities and patient care

is limited. In the Eastern Mediterranean Region (EMR) cancer incidence is predicted to rise by

1.8-fold during the next decade [2]. Breast cancer (BC) is one of the most widespread

malignancies affecting women worldwide [3]. Around 5-10% of BC cases can possess breast

cancer susceptibility genes that influence to increased risk of malignancy [4-6]. In the Kingdom

of Bahrain there is a very high prevalence rate of BC as it accounts for 37.2% of all cancers in

females, and 20% of all new cancer diagnoses [7]. The age-standardized rate (ASR) per 100,000

women of 53.4 in 2010[8], which is the highest in all of Gulf Cooperation Council (GCC), and one

of the highest in the world [9, 10]. Figure 1(a, b) shows the number of new cases for all ages in

Bahrain [11, 12].

Figure 1a: Number of new cases in 2020, males all ages

Figure 1b: Number of new cases in 2020, females all ages

The incidence of BC correlates strongly with age, with the highest incidence rates observed in

older women and only 6% occurring in women under 40 years old. The mean age at diagnosis

in Bahrain was 50.7 years [13]. However, BC is usually more aggressive and advanced in the

younger age groups [14]. Unfortunately, despite recent advances in the management of BC the

ASR has decreased from 58.2 per 100,000 in 2000 to 53.4 per 100,000 in 2010, [15].

Furthermore, in Bahrain, the 5-year survival for BC was 63% compared to 80-90% in most

developed countries [16]. Survival in BC is thought to be dependent on many factors such as

the histology, tumor size, grade, lymph node status, hormone receptor status (HRS), estrogen

(ER), progesterone (PR) and human epidermal growth factor receptor (HER)-2 over- expression, as well as the stage at presentation.

Many types of breast cancer affect the human body; some of the most common breast cancers

are invasive ductal carcinoma and invasive lobular carcinoma. Both types of breast cancer grow

outside the ducts and lobular, and spread into other parts of the breast tissue. Invasive cancer

cells can also be metastasized to other organs and tissues of the body through blood vessels and

lymph vessels [17]. There are three types of tissues in women’s breasts: as shown in Figure2.

Fibrous tissue grips the breast tissue in place, Glandular tissue is the segment of the breast that

produces milk called the lobes, Epithelial tissue and fibrous tissue together are called fibro- glandular tissue. Fatty tissue fills the cavity between the fibrous tissue, lobes, and ducts. The

main functionality of fatty tissue is to determine the breast structure and size. Based on results,

Page 3 of 12

218

British Journal of Healthcare and Medical Research (BJHMR) Vol 9, Issue 3, June - 2022

Services for Science and Education – United Kingdom

the percentage of fibrous tissue and glandular tissue can determine the breast density collated

with the number of fatty tissues in a woman’s breast. Breast density is categorized into four

divisions based on mammography diagnostics:

1. The breasts are almost entirely fatty (about 10% of women).

2. There are scattered areas of fibro-glandular density (about 40% of women).

3. The breasts are evenly dense throughout (about 40% of women).

4. The breasts are extremely dense (about 10% of women).

Figure 2: The anatomy of the breast

Breast cancer diagnostics at an early stage can help to reduce mortality rates while increasing

survival rates. An important role in medical practice is to determine if a breast biopsy is

required through the utilization of imaging techniques such as mammography, ultrasound

imaging (US), magnetic resonance imaging (MRI), and Positron Emission Tomography (PET)

[18-20].

Due to obstacles faced by physicians during diagnosis, breast cancer detection from imaging

can be misinterpreted. Mammography, for example, can be compromised in detecting breast

cancer without the as it can cause many limitations in the diagnostic procedures, leading to

false clinical diagnostics. Enhancing the quality of imaging diagnostics of breast cancer by

applying principles and algorithms of artificial intelligence (AI) in mammography can improve

mammography detection quality and reduce physicians' workload by minimizing the second

call for diagnostic and improving waiting periods for the results.

Researchers found that AI models could predict breast cancer from scans with a similar level of

accuracy to expert physicians. Compared to human interpretation, AI showed an absolute

reduction in the error of cases where the cancer was incorrectly identified and cases where the

cancer was missed. [21]. Early diagnosis of breast cancer with AI gives the patients more

chances of enchased treatment and a greater rate of survival, even with the late-stage cancer

diagnosis. In specific, treatments can be modified and altered well to treat a patient through

their diagnosis. [22-23].

AI has become more popular in the last six years as it is improving rapidly in many fields; it is

noticeable that AI is ordinary in scientific and engineering communities [24]. There are many

implementations of AI that benefit the healthcare system. For example, “Drug development is