Enhancing Prediction Precision and Reliability in Cervical, Lung, and Breast Cancer Diagnosis

Enhancing Prediction Precision and Reliability in Cervical, Lung, and Breast Cancer Diagnosis

Sayani Ghosh, Arnab Dutta, Shivnath Ghosh, Avijit Kumar Chaudhuri
DOI: 10.4018/979-8-3693-3679-3.ch003
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This study covers a wide range of cancers, focusing on cervical, lung, and breast cancer. Developing fast, accurate, and interpretable machine learning models for early diagnosis is critical to reducing the multifactorial mortality associated with these cancer types. Using a two-stage hybrid feature selection method, this study evaluates classification models using specific cervical, lung, and breast cancer data obtained from the UCI Machine Learning Repository. The cervical cancer dataset contains 36 features, the lung cancer dataset contains 16 features, and the breast cancer dataset contains 31 features. In the first stage, a random forest architecture is used for feature selection to identify features 5,7, and 7 that show a strong correlation with their cancer while reducing the difference between them. In Stage 2. Logistic regression (LR), naive bayes (NB), support vector machine (SVM), random forest (RF), and decision making (DT) were used to identify cervical cancer, lung, and breast cancer patients for five selections.
Chapter Preview
Top

1. Introduction

Cancer is a global health problem and causes morbidity and mortality worldwide. An estimated 9,958,133 people died of cancer in 2020 and 19,292,789 new cases were reported (Sung et al., 2021). These different diseases pose a significant burden on global health, especially in low- and middle-income countries. India accompanied 6.9% of the world total (1,324,413 cases) in 2020. In Indian society, women are more likely to get cancer than men (Mishra et al., 2011). Worldwide, cancer is expected to reach 22 million new cases by 2030. The impact of cancer is far-reaching, affecting individuals, families and communities. Due to health disparities, developing countries often face additional challenges in providing adequate screening, treatment, and support. Cancer has brought a heavy burden to the world, breast cancer, lung cancer, and lung cancer are the cause of many diseases and deaths worldwide.

1.1 Cervical Cancer

Cervical cancer is a type of gynecological cancer that includes the onset of cancer. With its rate (6.5%), it ranks fourth (7.7%) in terms of mortality among women worldwide (Sung et al., 2021). Contrary to the trend in the world, the incidence (18.3%) and mortality (18.7%) of cervical cancer in India is higher among all female cervical cancer patients with different types of cancer such as blood cancer, ovarian cancer, blood cancer, lung cancer, and oral cancer. It ranks second among them, cancer, stomach cancer, etc. Cancer is a type of cancer that occurs due to poor growth and proliferation of cells in the cervix (the opening from the uterus to the vagina or birth canal) (Ji et al., 2011), uncontrolled cell growth, and proliferation of cells.

Early symptoms of breast cancer include frequent urination and postmenopausal bleeding. As with uterine cancer, serious symptoms such as abdominal pain, loss of appetite, weight loss, fatigue, swelling in the legs, and bladder, and kidney failure may occur (Schiffman et al., 2007). There are two varieties of cervical cancer: squamous epithelial carcinoma and glandular carcinoma. In normal women, it takes 15 to 20 years for cervical cancer to progress from cervical intraepithelial neoplasia (CIN) to cancer. Most cases of cervical cancer (99%) are associated with the human papillomavirus (HPV) virus, a sexually transmitted virus.

While 70-80% of cervical cancers worldwide are attributed to HPV (mainly HPV-16 and HPV-18 genotypes), HPV is the most common cause of cervical cancer in India. This rate is 88-97% in women with breast cancer. HPV infection can heal on its own within a few months, and approximately 90% of patients can prevent cervical cancer within 2 years thanks to early diagnosis and treatment of precancerous disease. HPV is a group of viruses that frequently infect the genitals of men and women (World Health Organization (WHO), 2021).

The immune system that is resistant to HPV usually blocks the virus in infected women, but in a very small number of women, the virus survives long before cervical cancer cells (Andreaus et al., 2013). It is currently unclear whether the above risk factors cause cervical cancer independently or as a combination of HPV infection (Jemal et al., 2010; Bobdey et al., 2016; Kjellberg et al., 2000; Plummer et al., 2003; Luhn et al., 2013; International Agency for Research on Cancer (IARC) Multicentric Cervical Cancer Study Group et al., 2002; Rajkumar et al., 2020). In a country like India, the unclear relationship between a patient's risk and knowledge of the disease and the many tests required for effective treatment can cause medical procedures to slow down and delay, leading to increased mortality.

In the era of cool computing and artificial intelligence, analytics can work through data mining and machine learning to understand patterns in medical data. Check cancer risks. The power of classification, clustering, and prediction derived from various optimization, statistical, and probabilistic methods is essential for data-driven diagnosis. Identifying women at high risk for breast cancer will help prioritize early cancer diagnosis and treatment. Cancer screening and early diagnosis help reduce the risk of death in cancer patients. Cancer is a global health problem and reported morbidity and mortality rates are the highest among all cancers.

Complete Chapter List

Search this Book:
Reset