A Machine Learning-Based Diabetes Risk Prediction System
This project is a Streamlit-based Diabetes Prediction System that provides:
- Real-time diabetes prediction
- Interactive analytics dashboard
- Multiple machine learning model comparison
- Feature importance analysis
- Correlation heatmap visualization
- ROC curve analysis
- Confusion matrix evaluation
- Modern responsive user interface
The system uses Machine Learning, Classification Algorithms, and Interactive Visualizations to predict diabetes risk based on patient clinical information.
- Diabetes Prediction
- Interactive Streamlit Dashboard
- Data Visualization & Insights
- Feature Importance Analysis
- Correlation Heatmap
- ROC Curve Analysis
- Confusion Matrix Visualization
- Model Performance Evaluation
- Responsive Modern UI
- Real-Time Risk Assessment
Diabetes-Prediction-System/
│
├── app.py
├── best_diabetes_model.pkl
├── diabetes_scaler.pkl
├── diabetes.csv
├── Logistic_regression.ipynb
├── Decision_tree.ipynb
├── Random_forest.ipynb
├── Naive_bayes.ipynb
├── Support_vector_machine.ipynb
├── Kernel_SVM.ipynb
├── K-nearest_neighbours.ipynb
├── requirements.txt
└── README.md
- Python 3.x
- Streamlit
- Pandas
- NumPy
- Scikit-Learn
- Plotly
- Joblib
- Matplotlib
Repository Link:
https://github.com/AmitSharma9754/Diabetes-Risk-Prediction-System
Clone using Git: git clone https://github.com/AmitSharma9754/Diabetes-Risk-Prediction-System.git cd Diabetes Risk Prediction System
pip install -r requirements.txtstreamlit run app.py| Module | Description |
|---|---|
| Diabetes Prediction | Predicts diabetes risk using machine learning |
| Risk Assessment | Probability-based prediction analysis |
| Data Visualization | Interactive charts and dataset insights |
| Model Accuracy | Accuracy, Precision, Recall and ROC-AUC analysis |
| Feature Importance | Importance ranking of medical features |
| Correlation Heatmap | Relationship between dataset features |
| Confusion Matrix | Classification performance visualization |
| Developer Information | Project and developer details |
The model uses the following clinical features:
- Pregnancies
- Glucose
- Blood Pressure
- Skin Thickness
- Insulin
- BMI
- Diabetes Pedigree Function
- Age
- Outcome
0 → Non-Diabetic
1 → Diabetic
- Logistic Regression
- Decision Tree Classifier
- Random Forest Classifier
- Naive Bayes Classifier
- Support Vector Machine (SVM)
- Kernel SVM
- K-Nearest Neighbors (KNN)
- Best Performing Model(KNN) Selected and Saved as:
best_diabetes_model.pkl
- Diabetes Distribution Analysis
- Glucose Distribution Visualization
- Glucose vs BMI Scatter Plot
- Age Distribution Histogram
- Correlation Heatmap
- Feature Importance Chart
- ROC Curve
- Confusion Matrix
- Risk Assessment Dashboard
Enter patient information:
- Pregnancies
- Glucose
- Blood Pressure
- Skin Thickness
- Insulin
- BMI
- Diabetes Pedigree Function
- Age
Click:
Run Prediction
The application will generate:
- Diabetes Prediction
- Prediction Probability
- Risk Assessment
- Confidence Score
Explore:
- Data Visualization Tab
- Model Accuracy Tab
- Feature Importance Analysis
- Correlation Analysis
- Cloud Deployment
- User Authentication
- PDF Report Generation
- Patient History Tracking
- Deep Learning Models
- Multi-Disease Prediction
- Mobile Application Support
This application is developed strictly for educational and learning purposes only.
The predicted results generated by this system are machine learning estimations and should not be considered medical diagnoses.
Always consult qualified healthcare professionals for medical advice and treatment.
The developer is not responsible for any decisions made based on the predictions generated by this application.
You can contribute by:
- Improving model accuracy
- Enhancing the user interface
- Adding new visualizations
- Optimizing performance
- Adding additional machine learning models
- Fixing bugs
Pull requests are welcome.
Machine Learning Project
Streamlit Web Application
Multiple Classification Model Comparison
Real-Time Diabetes Prediction
Interactive Data Visualization
Medical Dataset Analysis
Model Evaluation Dashboard
Portfolio & Resume Ready Project
Amit Sharma
Email: Amitsharma97545@gmail.com