Skip to Main Content
Ashealth care grows more complex, storage and accessibility of medical information is not only valuable but also necessary. The long-term goal for electronic health records (EHR) is to make patient data securely available to health care providers such as hospitals, emergency personnel and research purpose, when and where the information is needed. This work is focused on medical time-series signals and in particular on Electroencephalograms (EEG). This work presents technique that allows the embedding and retrieval of sensitive numerical data, such as the patient's social security number or birth date, within the medical signal and allows the effective hiding of the sensitive metadata within the signal itself. This methodology builds upon data hiding notions, and presents the following desirable characteristics: (a) it does not distort important EEG characteristics, which are essential for proper medical diagnosis, (b) it allows not only the embedding but also the efficient retrieval of the embedded data, (c)it provide fault tolerance by employingrobust watermarking.