
Certificate of Completion
THIS ACKNOWLEDGES THAT
HAS COMPLETED THE SUMMER 2026 DATA SCIENCE BOOT CAMP
Tanmoy Kumar Debnath
Roman Holowinsky, PhD
July 20, 2026
DIRECTOR
DATE

TEAM
Early Diabetes Risk Screening Using Interpretable Machine Learning
Tanmoy Kumar Debnath, Shirin Alimirzaei

Diabetes is a major public health concern, and early identification of high-risk individuals can help people take precautionary actions before the condition becomes severe. Depending on clinical, demographic, and lifestyle factors such as age, BMI, blood pressure, cholesterol history, physical activity, smoking status, diet behavior, and general health condition, the risk of diabetes can vary from person to person. In this project, the goal is to explore these factors and develop an interpretable machine learning framework for early diabetes risk screening.
In this project, raw NHANES data will be collected from multiple CDC files, including demographics, examination, laboratory, and questionnaire modules. These files will be merged using participant ID, special missing-value codes will be cleaned, and a diabetes target will be constructed using self-reported diagnosis and clinical laboratory thresholds. Risk factors will be engineered from BMI, blood pressure, waist circumference, smoking status, physical activity, and socioeconomic variables. Finally, an early diabetes screening model will be developed.
The goal of this project is to build a practical and explainable early diabetes screening framework that can identify individuals at higher risk and highlight the most important risk factors, so that lifestyle changes, regular monitoring, and further medical consultation can be encouraged early.
