Identifying The Breast Cancer-related Determinants Using Traditional and Machine Learning Methods: A Case Study in Bangladesh

Faculty Mentor

Jing Kersey

Location

Russell Union 2047

Type of Research

On-going

Session Format

Oral Presentation

College

Jiann-Ping Hsu College of Public Health

Department

Department of Biostatistics, Epidemiology and Environmental Health Sciences (BEES)

Abstract

A malignant disease, breast cancer is influencing different aspects of life, such as physical, mental, and community integration in Bangladesh. Early diagnosis and balanced lifestyle are essential for risk mitigation and prevent it. This study focused on detecting significant factors which are related with breast cancer in Bangladesh to support prevention planning. This study applied a case-control study, including a total of 240 women in Bangladesh. Statistical and machine learning techniques, were applied to determine the factors of with breast cancer. The significant factors are age, education, fatty food consumption, pregnancy history, breastfeeding duration, obesity, sunlight exposure, exercise, and family history. For machine learning methods, with an accuracy of 0.9577 and 0.9436 for logistic regression (LR), and random forest (RF), respectively, the above factors classify cases and controls. For these two methods, area under curve (AUC) were 0.9817 and 0.9875. The odds ratio presents that the risk of this cancer grows with age, the women who are over 40 years having a 30.0 (CI: 3.97-226.5, p=0.001) times greater risk compared to those who are under 20. Fatty food intake raises this risk by 7.327 (CI: 3.942-13.619, p< 0.001) times, while pregnancy after age 30 increases it by 2.303 times. Other factors include shorter breastfeeding duration (less than one year), obesity, limited sunlight exposure, lack of exercise, and a family history of breast cancer, which heightens the risk of breast cancer. RF analysis identified age, breastfeeding duration, and pregnancy history as the most significant contributors to breast cancer risk among all analyzed factors. Public health policies should be initiated at the government, society, and family levels to promote early pregnancy, longer breastfeeding, regular exercise, healthy diets, and routine screening for women over 40 or with a family history to decrease the risk of breast cancer.

Program Description

.

Start Date

4-23-2026 10:15 AM

End Date

4-23-2026 10:30 AM

This document is currently not available here.

Share

COinS
 
Apr 23rd, 10:15 AM Apr 23rd, 10:30 AM

Identifying The Breast Cancer-related Determinants Using Traditional and Machine Learning Methods: A Case Study in Bangladesh

Russell Union 2047

A malignant disease, breast cancer is influencing different aspects of life, such as physical, mental, and community integration in Bangladesh. Early diagnosis and balanced lifestyle are essential for risk mitigation and prevent it. This study focused on detecting significant factors which are related with breast cancer in Bangladesh to support prevention planning. This study applied a case-control study, including a total of 240 women in Bangladesh. Statistical and machine learning techniques, were applied to determine the factors of with breast cancer. The significant factors are age, education, fatty food consumption, pregnancy history, breastfeeding duration, obesity, sunlight exposure, exercise, and family history. For machine learning methods, with an accuracy of 0.9577 and 0.9436 for logistic regression (LR), and random forest (RF), respectively, the above factors classify cases and controls. For these two methods, area under curve (AUC) were 0.9817 and 0.9875. The odds ratio presents that the risk of this cancer grows with age, the women who are over 40 years having a 30.0 (CI: 3.97-226.5, p=0.001) times greater risk compared to those who are under 20. Fatty food intake raises this risk by 7.327 (CI: 3.942-13.619, p< 0.001) times, while pregnancy after age 30 increases it by 2.303 times. Other factors include shorter breastfeeding duration (less than one year), obesity, limited sunlight exposure, lack of exercise, and a family history of breast cancer, which heightens the risk of breast cancer. RF analysis identified age, breastfeeding duration, and pregnancy history as the most significant contributors to breast cancer risk among all analyzed factors. Public health policies should be initiated at the government, society, and family levels to promote early pregnancy, longer breastfeeding, regular exercise, healthy diets, and routine screening for women over 40 or with a family history to decrease the risk of breast cancer.