Synthetic Medical Symptoms Classification

A machine learning project that analyzes synthetic medical symptoms data and builds a classification model to predict medical diagnoses based on patient symptoms, vital signs, and laboratory values.

📋 Overview

This project performs comprehensive exploratory data analysis (EDA) on a synthetic medical symptoms dataset and implements a Random Forest classifier to predict medical diagnoses. The model uses various features including patient demographics, symptoms, vital signs, and laboratory test results to classify patients into different diagnostic categories.

🎯 Objectives

• Perform exploratory data analysis on medical symptoms data
• Visualize data distributions, correlations, and patterns
• Build a machine learning model to predict medical diagnoses
• Evaluate model performance using various metrics
• Identify important features for diagnosis prediction

📊 Dataset

The dataset (synthetic_medical_symptoms_dataset.csv) contains 3,000 patient records with the following information:

Features

Demographics:

• age: Patient age
• gender: Patient gender (Male/Female)

Symptoms (0-3 severity scale):

• fever: Fever severity
• cough: Cough severity
• fatigue: Fatigue level
• headache: Headache severity
• muscle_pain: Muscle pain level
• nausea: Nausea severity
• vomiting: Vomiting severity
• diarrhea: Diarrhea severity
• skin_rash: Skin rash presence/severity
• loss_smell: Loss of smell severity
• loss_taste: Loss of taste severity

Vital Signs:

• systolic_bp: Systolic blood pressure (mmHg)
• diastolic_bp: Diastolic blood pressure (mmHg)
• heart_rate: Heart rate (bpm)
• temperature_c: Body temperature (°C)
• oxygen_saturation: Blood oxygen saturation (%)

Laboratory Values:

• wbc_count: White blood cell count (×10³/µL)
• hemoglobin: Hemoglobin level (g/dL)
• platelet_count: Platelet count (×10³/µL)
• crp_level: C-reactive protein level (mg/L)
• glucose_level: Blood glucose level (mg/dL)

Target Variable:

• diagnosis: Medical diagnosis (COVID-19, Influenza, Dengue, Malaria, Pneumonia)

🛠️ Requirements

• Python 3.7 or higher
• Jupyter Notebook or JupyterLab

Installation

1. Clone the repository:

   git clone https://github.com/Denixzertyux/SyntheticMedicalSymptoms.git
   cd SyntheticMedicalSymptoms

2. Create a virtual environment (recommended):

   python -m venv venv
   
   # On Windows:
   venv\Scripts\activate
   
   # On macOS/Linux:
   source venv/bin/activate

3. Install dependencies:

   pip install -r requirements.txt

📁 Project Structure

SyntheticMedicalSymptoms/
│
├── synthetic_medical_symptoms_dataset.csv  # Dataset file (3,000 patient records)
├── medical_symptoms_analysis.ipynb          # Jupyter notebook with complete analysis
├── requirements.txt                         # Python package dependencies
├── LICENSE                                  # MIT License
├── .gitignore                               # Git ignore file
└── README.md                                # Project documentation

🚀 Getting Started

Running the Analysis

1. Start Jupyter Notebook:

   jupyter notebook

2. Open the analysis notebook:

   • Navigate to medical_symptoms_analysis.ipynb in the Jupyter interface
   • Or directly open it:

     jupyter notebook medical_symptoms_analysis.ipynb

3. Run the cells sequentially:

   • Cell 0: Import required libraries and set up environment
   • Cell 1: Load the dataset
   • Cell 2: Data preprocessing and cleaning
   • Cell 3: Exploratory data analysis with visualizations
   • Cell 4: Model building and evaluation

📈 Analysis Workflow

1. Data Loading and Preprocessing

• Loads the CSV dataset
• Converts numeric columns to appropriate data types
• Handles missing values (if any)
• Final dataset shape: 3,000 rows × 24 columns

2. Exploratory Data Analysis

The notebook generates several visualizations:

• Histograms: Distribution of all numeric features
• Count Plot: Distribution of diagnoses
• Correlation Heatmap: Relationships between numeric features
• Pair Plot: Pairwise relationships for key features
• Box Plot: Age distribution across different diagnoses

3. Model Building

• Algorithm: Random Forest Classifier
• Features: All numeric features (excluding gender and diagnosis)
• Train/Test Split: 80/20 with stratification
• Hyperparameters:
  • n_estimators: 100
  • random_state: 42

4. Model Evaluation

The model evaluation includes:

• Accuracy Score: Overall classification accuracy
• Confusion Matrix: Visual representation of classification performance
• Feature Importance: Permutation importance analysis to identify key predictors
• ROC Curve: Receiver Operating Characteristic curve for model performance

📊 Model Performance

The Random Forest classifier achieves an accuracy of approximately 24.17% on the test set. Note that this is a multi-class classification problem with 5 diagnostic categories, so performance should be evaluated in context of:

• Class distribution balance
• Baseline accuracy (random guess: ~20% for 5 classes)
• Confusion matrix patterns
• Feature importance insights

🔍 Key Insights

The analysis provides insights into:

• Feature distributions and their relationships
• Which symptoms and vital signs are most predictive
• Patterns in diagnosis distribution
• Correlations between different medical measurements

⚠️ Important Notes

1. Synthetic Data: This dataset is synthetic and should not be used for real medical decision-making.

2. Model Limitations:

   • The model is trained on synthetic data
   • Performance metrics are for demonstration purposes only
   • Not suitable for clinical use

3. Data Privacy: This is synthetic data, so no real patient information is included.

🔮 Future Improvements

Potential enhancements to the project:

• Hyperparameter tuning for better model performance
• Feature engineering and selection
• Cross-validation for more robust evaluation
• Additional classification algorithms (SVM, XGBoost, Neural Networks)
• Class imbalance handling techniques
• Multi-class ROC curves for all diagnoses
• Model interpretation using SHAP values

🤝 Contributing

Contributions, issues, and feature requests are welcome! Feel free to check the issues page.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

• Thanks to the open-source community for the excellent libraries used in this project
• Synthetic dataset created for educational purposes

📧 Contact

For questions or suggestions, please open an issue on GitHub.

---

⚠️ Disclaimer: This project uses synthetic data and is intended for educational and research purposes only. It should NOT be used for actual medical diagnosis or treatment decisions. Always consult with qualified healthcare professionals for medical advice.