Predicting Diabetes Risk Using Patient Data

Objective: Develop a predictive model to identify individuals at high risk of developing Type 2 diabetes based on clinical and lifestyle factors.

? Tools & Tech:

Python (Scikit-learn, Pandas, Matplotlib)

Jupyter Notebook

Streamlit (for building a simple web app)

Dataset Includes:

Age, BMI, blood pressure

Glucose levels, insulin levels

Physical activity, family history

Outcome label (diabetic or not)

Workflow:

Data cleaning and normalization

Exploratory data analysis to find correlations

Train/test split and model training (e.g., logistic regression, decision tree)

Evaluate model performance (ROC curve, confusion matrix)

Deploy a basic app where users can input values and get a risk score

Outcome: Model achieved 88% accuracy. Found that high BMI and glucose levels were the strongest predictors. The app can be used in clinics for early screening.