Multi-Biometric Authentication Using Deep Learning Classifier for Securing of Healthcare Data

Abstract

Ensuring the security of healthcare data is becoming an increasingly important problem as modern technology is integrated into existing medical services. As a consequence of the adoption of healthcare data in the health care sector, it is becoming more and more common for a health professional to edit and view a patient's record using a tablet PC. To protect the patient's privacy, a secure authentication system to access patient records must be used. Yet, most Health apps used by consumers do not fall under federal or regional health privacy laws, even when the apps are used to manage a chronic illness. To solve this issue multi-biometric authentication is performed in this work via the use of deep learning classifier. This paper analyzes the performance of combining the use of on-line signature and fingerprint authentication to perform robust user authentication. Signatures are verified using the dynamic time warping (DTW) technique of string matching. The proposed minutiae-based matching algorithm, stores merely a small number of minutiae points, which greatly reduces the storage requirement with the help of phase correlation. Here, matching score level fusion is used by applying weighted sum rule for the biometric fusion process. To improve the authentication performance, deep learning classifier is proposed in this work for multi-biometrics authentication. When a biometric authentication request is submitted, the proposed authentication system uses deep learning to automatically select an appropriate matching image. In the experiment, biometric authentication was performed on healthcare in the UCI database. Multi -Biometric Authentication was used during the authentication stage.

A Systematic Literature Review of the Types of Authentication Safety Practices among Internet Users

“…There were 17 studies about textual password authentication [17], [21], [23], [25 -27], [30], [32 -38], [41 -43] and three studies about graphical password authentication [22], [34], [39]. Seventeen studies discussed multifactor authentication [17], [19], [21], [23 -28], [30], [32 -37], [42] (as shown in Figure .2.). The review found that most scholars had chosen to focus on biometric authentication.…”

“…There were some studies excluded from this review [48 -52] as they did not focus on types of authentication methods as a safety practice for information security. Of the 28 articles reviewed, 13 were based in India [20 -26], [30], [34 -37], [42]; four were from China [32], [33], [38], [44]; two from Saudi Arabia [18], [19] one from Poland [27]; one from the Czech Republic [31]; one from the United Arab Emirates [28]; one from Turkey [43]; one from Ukraine [40]; 1 from Jordan [17]; and one each from Zambia [29]; Philippines [39]; and the United States [41] (as shown in Figure 2).…”

“…However, this study found the scheme can only be applied for a small proportion of users. In addition, digital signature was also implemented in a health care system with fingerprint verification [19]. As proposed, the signature verification process has several steps.…”

A Novel Multimodal Biometric System using Face and FKP

“…The performance of the biometric verification system is estimated by measuring Genuine Acceptance Rate (GAR), False Acceptance Rate (FAR), and False Rejection Rate (FRR) [19]. Formula to calculate the FAR and FRR is given below: FAR= α / β × 100 Where, α = Number of accepted imposter β = Total number of imposter access FRR= γ / μ × 100 Where, γ = Number of rejected clients μ = Total number of client access GAR is defined as a percentage of legitimate users accepted by the biometric system and formula to calculate GAR is as follows:…”

Petri Nets Generating Array Languages

“…Colour Petri nets and timed colour Petri nets were introduced for password generation [11]. Password and authentication issues are addressed by various researchers in finding better solution towards rectification [18][19][20][21]. Motivated by these concepts we have defined a notion of diamond arrays generated by Petri nets.…”