View Proposal

---

Proposer

Eric Nimako Aidoo

Title

Modelling risk factors for Breast cancer: a comparison of Bayesian and frequentist approaches

Goal

Description

Breast cancer remains one of the most common cancers among women, affecting millions of individuals each year. Timely identification of breast cancer is crucial for improving patient outcomes and reducing mortality rates. Traditionally, breast cancer diagnosis has relied on manual evaluation by medical professionals, including physical examinations, imaging techniques, and biopsy analysis. These methods may not fully account for the inherent complexity and uncertainty within patient data. Bayesian modelling presents a compelling alternative by integrating prior clinical knowledge with observed data and explicitly quantifying uncertainty in risk predictions. This study aims to utilise Bayesian modeling techniques to predict breast cancer risk levels and examine the relationships between these risk levels and potential risk factors to support early detection and targeted treatment strategies. The performance of the Bayesian framework will be compared to the conventional frequentist approach of risk classification.

Resources

Background

Url

Difficulty Level

Moderate

Ethical Approval

None

Number Of Students

2

Supervisor

Eric Nimako Aidoo

Keywords

bayesian models, breast cancer risk, risk level classification, risk factors.

Degrees

Master of Science in Data Science
Bachelor of Science in Statistical Data Science
BSc Data Sciences