🩺 Diabetes Prediction – Classical and Quantum Approach

This project explores various techniques for diabetes prediction, combining classical Machine Learning approaches with innovative Quantum Machine Learning (QML) methods. The goal is to compare the performance and effectiveness of different models in classifying the presence or absence of diabetes using real-world datasets.

📁 Project Structure

.
├── src/                        # Source code
│   ├── Dataset_Preprocessing.ipynb  # Dataset preprocessing and analysis
│   ├── ML_classification.ipynb      # Classical classification models (ML)
│   └── VQC.ipynb                    # Quantum classifier (Variational Quantum Circuit)
│
├── dataset/                   # Datasets used
│   ├── diabetes.csv                               # Main dataset
│   ├── final_diabetes_dataset_3.csv               # Preprocessed dataset with 3 features
│   ├── final_diabetes_dataset_4.csv               # Preprocessed dataset with 4 features
│   └── final_diabetes_dataset_5.csv               # Preprocessed dataset with 5 features
│
└── README.md                  

🧠 Objective

The main objective is to predict diabetes based on medical parameters (such as glucose, blood pressure, BMI, etc.), testing different classification approaches:

• Traditional Machine Learning models (e.g., Decision Tree)
• Quantum Classifier (VQC – Variational Quantum Circuit)

📊 Dataset

The main dataset (diabetes.csv) is based on the well-known Pima Indians Diabetes Dataset. Preprocessed versions are also available, using different normalization and feature selection techniques.

Key features:

• Number of pregnancies
• Glucose
• Blood pressure
• Skin thickness
• Insulin
• BMI
• Diabetes pedigree
• Age
• Diabetes diagnosis (0 = no, 1 = yes)

⚙️ Dependencies

To run the notebooks, make sure the following packages are installed:

pip install numpy 
pip install pandas 
pip install scikit-learn 
pip install matplotlib 
pip install seaborn 
pip install qiskit

🚀 Execution

1. Preprocessing: Run Dataset_Preprocessing.ipynb to analyze and prepare the data.
2. Classical Classification: Run ML_classification.ipynb to test various ML models.
3. Quantum Classification: Run VQC.ipynb to test a VQC model using Qiskit.

📈 Results

This project compares:

• Accuracy and metrics of classical models
• Performance of the quantum model on datasets with different numbers of features (3, 4, and 5)

🔬 Technologies Used

• Python 3.x
• Scikit-learn
• Qiskit
• Pandas, NumPy, Matplotlib, Seaborn
• Jupyter Notebook

📚 References

• UCI ML Repository – Pima Indians Diabetes Dataset
• Qiskit Documentation

👤 Authors

Project developed by Rocco Pio Vardaro and Antonio Pio Francica as part of a study/experimentation on classical and quantum technologies applied to predictive medicine.