Personalized Mobile eHealth Services for Secure User Access Through a Multi Feature Biometric Framework

Personalized Mobile eHealth Services for Secure User Access Through a Multi Feature Biometric Framework

Georgios C. Manikis (Institute of Computer Science, Foundation for Research and Technology Hellas, Heraklion, Greece), Marios Spanakis (Institute of Computer Science, Foundation for Research and Technology Hellas, Heraklion, Greece) and Emmanouil G. Spanakis (Institute of Computer Science, Foundation for Research and Technology Hellas, Heraklion, Greece)
Copyright: © 2019 |Pages: 12
DOI: 10.4018/IJRQEH.2019010104

Abstract

Humans have various features that differentiates one person from another which can be used to identify an individual for security purposes. These biometrics can authenticate or verify a person's identity and can be sorted in two classes, physiological and behavioural. In this article, the authors present their results of experimentation on publicly available facial images and the efficiency of a prototype version of SpeechXRays, a multi-modal biometric system that uses audio-visual characteristics for user authentication in eHealth platforms. Using the privacy and security mechanism provided, based on audio and video biometrics, medical personnel are able to be verified and subsequently identified for two different eHealth applications. These verified persons are then able to access control, identification, workforce management or patient record storage. In this work, the authors argue how a biometric identification system can greatly benefit healthcare, due to the increased accuracy of identification procedures.
Article Preview
Top

Introduction

Biometric authentication is the process of verification of a person’s identity using a physical trait or a behavioural characteristic in order to accept the identity of the person and verify him/her as an authorized user (Jain et al., 2007; Li & Jain, 2009). From the technical perspective, biometric systems mainly rely on models derived from pattern recognition, where several characteristics from a person (e.g. voice, facial expression, fingerprint, etc.) are first transformed into a feature vector and then processed to deny or reject the verification and identification of a user. A major prerequisite in this process is the so-called training phase of the model composed of a pipeline in which: (i) captured biometric characteristics from specific users are stored in a database, and (ii) used for training the model based on that known content. Once training is performed accurately, the biometric system can be applied for verification and identification.

Verification process addresses biometric authentication of a specific user who claims an identity and desires to be recognized by the system (i.e. John Doe uses his magnetic ID card and his fingerprints, and requests access to building A). The system performs a one-to-one comparison between the biometrics of the user requesting access and his/ her corresponding characteristics retrieved from the database. A pattern recognition model estimates the level of similarity or matching score between the characteristics and allows access, in case this similarity metric shows a value above a predefined level of security. User identification is a more computationally complex process in which the biometric system searches all the available information stored in the database when a user requests access without providing any credentials. The system is responsible for answering questions such as “who is this person?”, “is he-she enrolled in the database?” and conducts a one-to-many comparison to verify that the user is registered to the security protocol and has been granted access to the requested entrance. Biometric recognition systems extract the features from voice, face and compare them with templates stored in databases for verification of a person. If the system uses only a single trait it is characterized as a unimodal biometric system, whereas if two or more biometric traits are used to identify and authenticate a person, the system is called multi-modal (Larcher et al., 2012; Ross et al., 2006).

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 9: 4 Issues (2020): 1 Released, 3 Forthcoming
Volume 8: 4 Issues (2019)
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing