Applying an Unsupervised Machine Learning Approach to Detect Dietary Habits of Breast Cancer Patients in Bangladesh

Abstract

Purpose: The study aims to examine the association between food habits/lifestyle and breast cancer using an unsupervised machine approach. 

Method: The dataset was collected from the hospitals of eight divisional cities in Bangladesh using a semi-structured questionnaire. Descriptive statistical tools and an unsupervised machine learning approach- Factor Analysis were used to analyze the data.

Results: The highest numbers of breast cancer patients were observed in the Sylhet division, followed by the Dhaka and Khulna divisions. It is noted that, overall, left breast cancer patients outnumber right breast cancer patients. We found that betel nuts, beverages, beef/mutton, etc. are high commonalities, which indicates that these food habits are highly associated with breast cancer. Moreover, most of the patients can’t bear the cost of treatment.

 Conclusions: It is concluded that most breast cancer patients are used to taking betel nuts and beverages that may cause this disease. Therefore, we should avoid unhealthy and junk foods.