Visualized Forward Secure Group Certificate Digital Signature Scheme and Its Application on Electronic Medical Record System

• Main Authors: Yao-ChangYu, 于燿彰
• Other Authors: Ting-Wei Hou
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/07076719226834664594

博士 === 國立成功大學 === 工程科學系 === 102 === Integrity and trustworthiness are important to electronic medical records. Digital signature is a significant method to ensure the integrity and trustworthiness of electronic medical records. The key(s) used to generate/verifying a medical staff’s digital signatures is generally stored in a smart card, which is called a medical personnel certificate. There are three situations that would force a certificate to be revoked: certificate expiration, loss of the certificate, and retirement of the certificate holder. An important issue is that after a certificate is revoked, the digital documents signed by the revoked certificate are longer verifiable. This research presents two new visualized forward secure digital signature algorithms, which are not based on RSA algorithm, for electronic medical records. A digital signature generated by applying either of the proposed algorithms, has the capability as forward secure, which means even a medical staff’s certificate is revoked, the medical records signed by the revoked private key still can be verified. The idea of group certificate helps to reduce the number of certificates and the time of searching certificates. Also, in order to extend the trust from a visualized traditional seal or a handwritten signature to a digital signature, an approach is proposed to embed the digital signature into a seal image or a handwritten signature image of the holder instead of a human unrecognizable form. Performance comparisons with existing methods, and security analysis are performed and a prototype is implemented to demonstrate the feasibility of the proposed approach. The peak signal to noise ratio (PSNR) of the digital signature visualization process is 60~70db, which is very good in human eyes’ recognition level (60~80dB).