ExpressionHash: Securing Telecare Medical Information Systems Using BioHashing

The COVID-19 outbreak and its medical distancing phenomenon have effectively turned the global healthcare challenge into an opportunity for Telecare Medical Information Systems. Such systems employ the latest mobile and digital technologies and provide several advantages like minimal physical contact between patient and healthcare provider, easy mobility, easy access, consistent patient engagement, and cost-effectiveness. Any leakage or unauthorized access to users’ medical data can have serious consequences for any medical information system. The majority of such systems thus rely on biometrics for authenticated access but biometric systems are also prone to a variety of attacks like spoofing, replay, Masquerade, and stealing of stored templates. In this article, we propose a new cancelable biometric approach which has tentatively been named as “Expression Hash” for Telecare Medical Information Systems. The idea is to hash the expression templates with a set of pseudo-random keys which would provide a unique code (expression hash). This code can then be serving as a template for verification. Different expressions would result in different sets of expression hash codes, which could be used in different applications and for different roles of each individual. The templates are stored on the server-side and the processing is also performed on the server-side. The proposed technique is a multi-factor authentication system and provides advantages like enhanced privacy and security without the need for multiple biometric devices. In the case of compromise, the existing code can be revoked and can be directly replaced by a new set of expression hash code. The well-known JAFFE (The Japanese Female Facial Expression) dataset has been for empirical testing and the results advocate for the efficacy of the proposed approach.     

生物特征模板保护

对当前国内外生物特征模板保护技术发展的现状进行综述和探索,对该方向的研究内容进行详细的梳理和分类.首先阐述传统生物特征识别系统存在的本质缺陷和易于遭受到的攻击的形式,进而从理论上引出了生物特征模板保护的必要性及其难点所在.然后以模板保护算法的具体操作方式为分类标准,详细阐述了当前在这个领域出现的比较有代表性的算法,诸如Biohashing和模糊保险箱(Fuzzy Vault)等.通过实验验证了Biohashing算法的优势和缺陷,并且在FVC2002 DB2数据库上对提出的改进Fuzzy Vault算法进