abhinavy1/AnonyMed_Medical_Data_Anonymization_System

AnonyMed-Medical-Data-Anonymization-System

AnonyMed is a comprehensive tool for anonymizing, managing, and analyzing medical data while preserving patient privacy. This system implements various privacy-preserving techniques including k-anonymity and pseudonymization to help healthcare organizations handle sensitive patient information securely.

📌 Overview

The medical field requires special attention to privacy due to the sensitive nature of health data. AnonyMed provides tools to:

• —Anonymize individual patient records
• —Process uploaded CSV files containing patient information
• —Perform privacy analysis and risk scoring
• —Visualize anonymized data patterns
• —Conduct cohort analysis on anonymized datasets
• —Export privacy-preserving datasets for research

🔑 Core Features

🛡️ Data Anonymization

• —Consistent pseudonymization of patient identifiers
• —Age bracketing to prevent exact age identification
• —Location generalization for addresses
• —Medical condition categorization
• —Automated PHI (Protected Health Information) detection in free text
• —Secure encryption of original data with access controls

🔍 Privacy Analysis

• —K-anonymity verification and reporting
• —Re-identification risk scoring
• —Privacy dashboard with visualizations
• —Data utility preservation metrics

📊 Data Visualization & Analysis

• —Distribution visualizations for demographic data
• —Cohort analysis capabilities
• —Patient clustering with t-SNE visualization
• —Privacy risk distribution charts

🛠️ Implementation Notes

📦 Dependencies

The system requires several Python libraries:

• —pandas, numpy — Data manipulation
• —matplotlib, seaborn — Visualization
• —scikit-learn — Clustering and dimensionality reduction
• —gradio — User interface
• —cryptography — Secure data storage
• —spaCy (optional) — PHI detection using NLP

💾 Data Storage

• —Anonymized data is stored in CSV format
• —Original identifiers are encrypted and stored separately
• —Encryption keys are securely managed

🧩 Privacy Mechanisms

1. 1.Pseudonymization: Replaces identifiers with consistent pseudonyms
2. 2.Generalization: Reduces precision of values (e.g., age ranges, location regions)
3. 3.Categorization: Groups medical conditions into broader categories
4. 4.K-anonymity: Ensures each record shares attributes with at least k-1 other records
5. 5.Access Control: Encryption and purpose logging for authorized data recovery

⚠️ Limitations

1. 1.Basic NLP capabilities — spaCy may miss uncommon/contextual PHI
2. 2.Limited differential privacy — Formal DP techniques not implemented
3. 3.Simplistic risk scoring — Only k-anonymity based
4. 4.Limited attribute inference protection
5. 5.Simplified cohort analysis — Not a full statistical framework

🔐 Security Considerations

• —Additional security measures should be implemented in production
• —Authorized Access should include proper authentication & authorization
• —Key management should use secure, dedicated services
• —Logging and audit trails should be comprehensive
• —Regular security audits are strongly recommended