DICOM PII De-Identification and Reversible Data Hiding

In the medical field, sharing medical images is crucial for remote diagnosis, and DICOM images are the most commonly used format. However, because they contain patients' Personally Identifiable Information (PII), strict protection is required to prevent unauthorized access or disclosure. This study adopts pseudonymization techniques to replace patients' identifiable information with virtual data and integrates encryption methods to enhance data confidentiality, ensuring that even if unauthorized access occurs, the data remains difficult to interpret. Furthermore, this study utilizes reversible data hiding techniques to embed the processed sensitive information into the images, allowing the images to be fully restored when needed without compromising their diagnostic utility. This method achieves the dual objectives of secure sharing and privacy protection.