Design and Evaluation of a Multi‑Domain Adaptive Multi‑Factor

Authentication Framework for Holistic Cybersecurity Reinforcement

S. Kannadhasan

, Pilli Lalitha Kumari

, Yuvarani R.

, S. Venkatesh

, Suniesh P.

and N. Shirisha

Department of Electronics and Communication Engineering, Study World College of Engineering, Coimbatore - 641 105,

Tamil Nadu, India

Department of Computer Science and Engineering Visakha Institute of Engineering & Technology, 88th Division, Narava

Visakhapatnam - 530027 Andhra Pradesh, India

Department of Management Studies, Nandha Engineering College, Vaikkalmedu, Erode, Tamil Nadu, India

Department of Computer Science and Engineering, J. J. College of Engineering and Technology, Tiruchirappalli, Tamil

Nadu, India

Department of CSE, New Prince Shri Bhavani College of Engineering and Technology, Chennai, Tamil Nadu, India

Department of CSE, MLR Institute of Technology, Hyderabad, Telangana, India

Keywords: Multi‑Factor Authentication, Cybersecurity, Adaptive Security, Biometric Spoofing Prevention,

Quantum‑Resistant Authentication.

Abstract: In the era of advanced cyber-attacks and the growing digital inter-connectivity the traditional authentication

methods do not guarantee a high security level. This research introduces an advanced multi-factor

authentication (MFA) framework featuring resiliency, user-centricity, and adaptiveness, for modern cyber

security systems hardening. The framework leverages biometric anti-spoofing methods, light-weight IoT-

friendly protocols, and AI-powered behavioral analytics to mitigates usability limitations, phishing risks, and

post-quantum attacks. Then a hybrid solution which combines MFAaaS with the ability to adapt to regulatory

compliance is proposed to address the weaknesses of the current MFAaaS systems. We also conduct real-

world simulations and case studies to demonstrate the effectiveness of the proposed model in enterprise,

healthcare and mobile environments. The results show that the context-based and adaptive MFA system

effectively improves security posture without sacrificing access and usability. This study not only closes the

void between academic investigation and practice, but also produces a road map for everywhere-available

strong authentication in the future cyber world.

1 INTRODUCTION

With the rise of digitalisation, increasing dependence

on online services and interconnected systems have

made strong cyber security more important than ever

before. There are increasing threats from cyber-attack

on system vulnerabilities as well as human

vulnerabilities, such as phishing, identity theft and

social engineering. Legacy authentication methods –

which are based primarily on just a single factor,

including passwords – have not been able to

adequately protect against multifaceted threats such

as these. Multi-Factor Authentication (MFA) has

been developed as a strong application to authenticate

the user by using some knowledge factors

(password), possession factors (ID devices) and

inherence factors (biometric or behavior).

But despite its advantages, traditional MFA

solutions have significant drawbacks such as being

inconvenient for users, not being scalable to IoT type

of environments, not being able to respond to new

emerging threats such as SIM swaps and deepfake

hacks, and becoming vulnerable to quantum compute

attacks. There are also more significant barriers to

establish the use of MFA in practice, including lack

of standardised regulation, lack of user

understanding, and integration issues in legacy

systems.

In this paper, we present the next-generation

MFA framework, which not only overcomes the

shortcomings of MFA solutions but also

Kannadhasan, S., Kumari, P. L., R., Y., Venkatesh, S., P., S. and Shirisha, N.

Design and Evaluation of a Multi-Domain Adaptive Multi-Factor Authentication Framework for Holistic Cybersecurity Reinforcement.

DOI: 10.5220/0013859400004919

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies (ICRDICCT‘25 2025) - Volume 1, pages

151-157

ISBN: 978-989-758-777-1

151

revolutionizes the way of digital identity verification.

Key to this is a focus on intelligent security models

driven by artificial intelligence (AI), next-gen

quantum-resistant cryptographic algorithms, and

privacy preserving decentralized models. This model

hopes to provide a scalable and well-prepared MFA

approach, and to form a stronger defence framework

across wide range of digital environments by seating

at its heart usability and resilience.

2 PROBLEM STATEMENT

Description Although Multi-Factor Authentication

(MFA) has been heralded as key to the improvement

of cybersecurity, current MFA deployments are

typically flawed by severe limitations that drastically

reduce its security and usability. Existing

technologies find it hard to achieve a good tradeoff

between security and user convenience, and

commonly present usability friction and low

compliance. Furthermore, the aforementioned MFA

solutions are not effective in IoT ecosystem and are

not adaptable to the resource-limited environment

and challenging for preventing advanced threats

including phishing, biometric spoofing, and SIM-

swapping attacks. The advent of quantum computing

will further amplify the demand for more secure

means of authentication, as traditional cryptographic

approaches may be rendered useless. Moreover,

uneven implementation of user control requirements

and the complexity of MFA integration into legacy

and hybrid cloud environments limit its general

applicability. The urgency of MFA mechanisms, push

us in the direction of deploying, sensing and

evaluating/ managing a next generation MFA

framework, which would be adaptable, user centric,

cost sensitive, capable of withstanding,

current/intelligence threat.

3 LITERATURE SURVEY

Authentication methods have evolved at the very

heart of cyber security and fundamental changes have

been driven in the past couple of years. Traditional

password-based scheme has been widely criticized

due to its vulnerability to the brute force attack,

credential spilling, phishing (Bhargava & Wu, 2021).

Therefore, the multi-factor authentication (MFA) is

offered as a more secure platform where the users are

asked for more than one type of evidence of identity

(eg a password, and a token, or biometric

characteristic) (Abid & Ahmad, 2023).

Yet MFA remains far from universally adopted,

primarily due to its usability difficulties. Researches

have shown that MFA systems are usually perceived

as inconvenient by users which influence the

compliance and user satisfaction negatively (Alotaibi

& Furnell, 2021; Ferreira & Antunes, 2021). This

problem is even more prominent in fields that access

must be simple such as the health sector and mobile

service (Saleh & AlZain, 2024).

Recent developments have been toward to

improve MFA using biometric authentication.

Though biometric identifiers enhance security, they

also raise novel risks including spoofing and deepfake

attacks (Barni & Dini, 2022; Jang & Kim, 2024). To

mitigate these challenges, liveness detection and

hybrid biometric approaches enhancing resiliency

have been proposed.

Phishing and SIM-swapping continue to pose a

threat to MFA systems using SMS based onetime

passwords (Ahn et al., 2022; Tan & Lim, 2021). AI-

based context sensitive authorization techniques

have been proposed to mitigate these attacks by

monitoring the risk during runtime (Liu et al., 2022;

Hussain & Sohail, 2022). However, the problem of

algorithmic fairness and interpretability in AI-driven

security models remains challenging (Gao & Liu,

2022).

In the context of IoT deployment, MFA approach

is not practical, as the latter has the limitations

regarding the computational power. In an attempt to

close this gap, several researchers has presented

extremely lightweight secure protocols, specifically

design for IoT devices (Banik et al., 2023; Bukhari &

Salah, 2022). However, the issue of interoperability

and standardization among different platforms is still

open.

Cloud based MFA-as-a-Service (MFAaaS) is

emerging, but is associated with potential issues

related to centralization and single points of failure

(Ghorbani and Salah, 2021). There is a new line of

investigation in the potential use of blockchain based

decentralized authentication systems, which are more

secure and transparent (Bukhari & Salah, 2022).

The arrival of quantum computing also constitutes

an existential threat to existing authentication

mechanisms. While there is limited but developing

practice along the similar approach, quantum-

resistant methods for MFA is aimed to be examined

to act as a backup for authenticating system (Hong &

Kim, 2023). Also, incorporation of compliance-aware

MFA in the current business and government

methodologies is under exploration in order to adhere

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

152

to data protection standards such as the GDPR and

HIPAA (Farooq & Alshamrani, 2023; Malik & Rizvi,

2022).

Together, these results clearly point to MFA as a

fundamental tool in cybersecurity, but one whose

current limits demand of the next generation a

balanced ability to accommodate usability,

adaptability, scalability, and future resilience. The

purpose of this work is to fill the gaps by presenting

an overall architectural blueprint that combines

cutting edge technologies of biometrics, Ai, IoTC,

and decentralization and quantum security.

4 METHODOLOGY

In this project, we take a layered, modular approach

to develop and evaluate at implementation and design

time a next-generation MFA framework that focuses

on adaptability, user-intuitiveness, and resilience.

Four major steps are included in the methodology:

system design, prototype implementation, integration

of adaptive strategies, and performance

measurement using real-world examples.

Figure 1: Adaptive multi-factor authentication workflow.

The first step is to design a new architecture that

combines traditional (passwords and tokens) and

advanced (biometric and behavioral) authentication

in conjunction with AI-based context analyzing. The

figure 1 shows the Adaptive Multi-Factor

Authentication Workflow. In order to achieve cross-

platform compatibility, the architecture of the

CVADL system is built considering a service-

oriented architecture (SOA) to ensure successful

integration with different environments, such as

clouds, mobile devices, IoT endpoints, and legacy

infrastructures. The thing we''re most focused on is

interoperability and modularity - so you can upgrade

or swap out individual pieces without the whole

stack coming tumbling down.

Table 1: Comparison of authentication methods used in the

proposed framework.

Authentic

ation

Facto

Securit

y Level

Usabilit

Implementa

tion

Complexity

Spoofing

Resistanc

Password Low High Low Low

Biometric

(Face/Fin

gerprint)

High

Mediu

Medium Medium

OTP via

App

Mediu

High Medium Low

Behavior

Profiling

High High High High

AI-Based

Contextu

al Auth

High High High High

A functional prototype of the MFA system is

created in the second phase and to ensure

reproducibility and cost considerations open sources

tools and libraries are used. Facial recognition with

liveness detection, keystroke dynamics, and a secure

mobile authenticator application form the prototype.

They are applied in conjunction with conventional

authentication techniques (e.g., PIN codes, OTP), so

that we are able to compare user behavior with work

carried out by the system. Identity records are

managed by the blockchain, which solves for

centralization risk, and keys are exchanged with PKI.

In order to accommodate resource constrained

devices such as IoT sensors, a lightweight validation

protocol is designed with help of EC cryptography

and efficient hashing functions.

Design and Evaluation of a Multi-Domain Adaptive Multi-Factor Authentication Framework for Holistic Cybersecurity Reinforcement

153

The third layer brings AI as an adaptive

intelligence to the system, reasoning through

machine learning model trained within behavioral

and context data. These models are trained based on

the user’s interaction with the system (eg, how often

they use a certain device, where they log in from, the

times at which they access the system) to identify

(and request) additional layers of authentication on

the fly. We incorporate a feedback loop for improving

model accuracy over time by utilizing federated

learning principles without compromising privacy.

Quantum-safe encryption algorithms (e.g. lattice-

based cryptography) are also integrated in the data

exchange protocol for post-quantum threat scenarios.

The third phase is a full-fledged evaluation of the

prototype in three use cases: on the enterprise

networks, healthcare systems and mobile banking

apps. Assessment includes penetration testing, user

acceptance testing and benchmarking. Performance

metrics considered are false acceptance rate (FAR),

false rejection rate (FRR), authentication time, and

user satisfaction ratio. Surveys and interviews are

employed to collect qualitative feedback on usability

and trust. Furthermore, the system is stress-tsted

under simualted attack vectors like Phishing,

Biometric Spoofing and AI driven Impersonation to

justify the strength of the system.

All experiments are conducted ethically and

training and evaluation data are anonymized to

ensure user privacy. The results of each phase are

either liable to re-enter the system, to improve its

design and efficiency, leading to a continuous

improvement cycle. The proposed framework is a

new production-ready model that can deliver a

flexible, secure and scalable MFA solution that can

help mitigate against the evolving threat landscape of

today’s digital environments.

5 RESULTS AND DISCUSSION

Experimental results on our proposed next-

generation Multi-Factor Authentication (MFA) are

promising from four perspectives security

effectiveness, usability, adaptability and robustness.

Through extensive simulation and real-world

deployment in three different domains enterprise

networks, mobile financial applications, and health-

care applications the system showed significantly

better performance over the state-of-the-art MFA

systems.

One of the most important discoveries concerns

the precision and strength of the authentication. AI-

powered behavioral analytics and biometric liveness

detection technology helped lower the false

acceptance rate (FAR) to 0.18% and the false

rejection rate (FRR) to 1.37% – down from the

average 1.6% FAR and 4.5% FRR of baseline

technology. This enhancement reinforces the idea that

verifying users through integrated adaptive

authentication layers and AI-driven decision-making

does not simply enhance security, but helps to refine

the trustworthiness of authentication. Facial

recognition with 3D depth mapping and fingerprint

validation as biometric modalities demonstrated

great resistance against spoofing attacks. In

controlled penetration testing with fake biometric

data and deepfake tampering, the system accurately

detected and rejected 96% of spoofing attempts,

demonstrating its efficacy against identity theft.

Table 2: MFA performance metrics in tested deployment

environments.

Environment

FAR

(%)

FRR

(%)

Avg.

Auth

Time

(ms)

User

Satisfacti

on (%)

Enterprise 0.20 1.30 920 88

Mobile

Banking

0.25 1.50 980 85

Healthcare

Access

0.18 1.27 890 87

Figure 2: FAR vs FRR in different environments.

There were substantial efficiency improvements in

the implementation of lightweight cryptographic

protocols for IOT-constrained devices as well. For

low-power situations such as in smart health

monitoring systems and embedded sensors in

industrial setup, the proprietary elliptic curve-based

authentication protocol 115-millisecond average

processing time that is sufficiently within the

acceptable latency bounds due to the real-time data

validation required. The figure 2 shows the FAR vs

FRR in Different Environments.This proves the

scalability and usability of the framework and its

applicability for extending strong authentication to

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

154

environments ignored by ordinary MFA solutions

because of the computational overhead.

In terms of usability, user satisfaction feedback

obtained in surveys and focus groups reported a

remarkable enhancement in perceived convenience

and trust. A majority of users (85%) had a positive

experience across the three deployment

environments, the simplified authentication process

and minimal interference at access were specifically

mentioned. But it was the flexibility of the

authentication layers that was important here. By

judiciously evaluating contextual components

(geographical position, identity of devices, usage

patterns), the system could be responsive to request

additional components solely whenever these

exceeded certain anomaly thresholds. This was to

minimise user inconvenience, while maintaining a

high-level of security. By contrast, with static MFA,

the full post-logon authentication steps for every

reduced by factor 1 ÷ 0.57 = 1.75 (43%)—a

substantial saving in operational efficiency.

Table 3: Resistance of the MFA framework to cyber

attacks.

Attack

Type

Detection

Rate (%)

Mitigation Strategy

Phishing 97.5

Contextual triggers +

ehavioral auth

SIM-

Swapping

94.3 Push-notification over SMS

Biometric

Spoofing

96.0

Liveness detection + 3D

facial mapping

Credential

Replay

99.2

AI anomaly detection +

timestamp checks

One of the challenge tasks in this study was to

test the robustness of the framework with new cyber

threats. The system beautifully withstood various

attack scenarios (phishing, SIM swap and credential

replay) under simulation pressure. Phishing

simulations to fake login portals and attempts to steal

credentials were thwarted by targeted popups that

require re-authentication via biometric measures or

device means. Moreover, the MFA-as-a-Service

(MFAaaS) over blockchain introduced in this design

also contributed to immutability and auditability in

authentication log, hence unauthorized access in

compromised network was avoided. This distributed

model ensured that there are no single points of

failure as can exist in a traditional MFA service.

Figure 3: Detection Rates Against Attack Types.

Table 4: User perception of MFA usability and security.

Feedback

Criteria

Positive

Response (%)

Neutral

(%)

Negative

(%)

Ease of Use 86 9 5

Perceived

Security

91 7 2

Willingness

to Adopt

89 8 3

At the level of preparation for post-quantum

security, the deployment of lattice-based

cryptographic algorithms provided a crucial degree of

prospective security. The figure 3 shows the

Detection Rates Against Attack Types. Till quantum

attacks pose an actual threat, quantum-resistant key

exchange was included in the prototype for a

benchmark against which to compare future versions,

and to support scaling. This compromise in

processing overhead overhead was acceptable for

enterprise-class solutions, but more apparent on

mobile clients. But that was softened by selective

encryption, with only quantum-safe protocols used

for sensitive authentication information.

Figure 4: User satisfaction metrics.

Another criterion of this study was adherence to

regulation. It was tested against a series of GDPR,

Design and Evaluation of a Multi-Domain Adaptive Multi-Factor Authentication Framework for Holistic Cybersecurity Reinforcement

155

HIPAA, and ISO/IEC 27001 compliance criteria. It

satisfied data minimization, encryption, access

control and auditability requirements across all the

considered environments. This policy engine

provided extensible of a policy-based engine which

allowed administrators to dictate the authentication

policies based on international/region-based

compliance requirements thereby adding another

level of flexibility within the framework.

The mention of these results represents a major

progression in the development of authentication

technologies. Conventional MFA schemes are

effective to certain extent, due to their rigidness, user

exhaustion and security vulnerability in aggressive

threat landscapes. This work reveals that if we look

at MFA from the perspective of being context-aware,

intelligent and decentralized then a majority of the

stated challenges can be tackled. Additionally, the

paper closes the loop between theory and deployment

by verifying the model over a wide variety of real-

world scenarios.

In summary we believe that the results provide

additional evidence that next generation MFA models

based on adaptive intelligence, biometric robustness,

IoT scale and quantum-resiliency are potentially

transformative when it comes to the security of

contemporary digital ecosystems. The work provides

a roadmap for future development and paves ways for

future improvements including privacy-preserving

federated identity verification, compatibility with

digital identity wallets, and smooth cross-platform

compatibility.

Table 5: Comparative analysis between proposed and

existing MFA models.

Feature

Traditional

MFA

Proposed

Framework

Adaptive

Intelligence

No Yes

Biometric Spoof

Resistance

Low High

IoT

Compatibility

Limited Fully Supported

Quantum Safety Absent Integrated

User-Centric

Design

Minimal Emphasized

Regulatory

Policy

Integration

Partial Dynamic

6 CONCLUSIONS

As digital environments become increasingly

interconnected, and threats continue to advance in

sophistication, protecting user identity has become a

cornerstone of any cyber security strategy. This work

has shown that legacy MFA solutions offer some

level of protection, but it is no longer enough.

Through the development of a new-generation

framework which takes adaptive intelligence as well

as biometric robustness to be fundamental issues,

includes quantum-safe encryption and a decentralized

structure, this paper shows a proactive response that

is compatible with contemporary security needs.

The model was able to overcome several

drawbacks of previous systems such as: usability

issues, susceptibility to novel threats, and inability to

scale down to resource-limited environments like

IoT. Through real-world deployment and through

simulation in enterprise, healthcare, and mobile

scenarios, we have shown the framework's ability to

simultaneously improve security and the user

experience. Using machine learning to analyze the

behavior, and applying blockchain for the purpose of

transparency and fault tolerance, the system realized

an effective balance of dynamic authentications and

operational efficiency.

In addition, this work is prepared for prospective

challenges by incorporating post-quantum

cryptographic techniques, making the long-term

viability of authentication systems resilient to

immediate disruptions due to impending advances in

quantum computation. It is also the ideal solution for

various organizational needs in different

jurisdictions, due to regulatory flexibility and policy-

driven design.

In summary, this article provides a complete and

scalable solution that transforms the role of MFA in

cybersecurity. It highlights the necessity to move

away from static fixed targets toward intelligent,

user-focused systems that can change in the face of

changing threats. This constitutes a blueprint for

future advances in authentication, and valuable

guidance for organizations looking to boost their

security without diluting accessibility or breaking

user trust.

REFERENCES

Abid, A., & Ahmad, M. (2023). Multi-factor authentication

for enhancing cybersecurity: A comparative analysis.

Journal of Cybersecurity Research, 10(2), 45–58.

https://doi.org/10.1016/j.jcsr.2023.100234

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

156

Ahn, C., Lee, S., & Kim, H. (2022). MFA effectiveness

against phishing attacks: An empirical evaluation.

Computers & Security, 118, 102725. https://doi.org

/10.1016/j.cose.2022.102725

Alotaibi, F., & Furnell, S. (2021). Usability challenges in

multi-factor authentication systems. Information and

Computer Security, 29(4), 604– 620. https://doi.org/

10.1108/ICS-04-2021-0067

Anderson, B., & Flores, W. (2021). Strengthening

enterprise cybersecurity through adaptive MFA

policies. International Journal of Information Security

Science, 10(3), 55– 68. https://doi.org/10.1016/j.ijiss

.2021.03.005

Banik, M., Sarkar, S., & Pal, S. (2023). Secure

authentication models for IoT devices using MFA.

IEEE Internet of Things Journal, 10(8), 6754–6765.

https://doi.org/10.1109/JIOT.2023.3245556

Barni, M., & Dini, G. (2022). Advancements in biometric

authentication for MFA frameworks. IEEE

Transactions on Dependable and Secure Computing,

19(5), 2760– 2773. https://doi.org/10.1109/TDSC.20

21.3109346

Bhargava, B., & Wu, J. (2021). Reducing phishing risks

through MFA implementation. ACM Computing

Surveys, 54(9), 1–36. https://doi.org/10.1145/3477121

Bukhari, S. F. A., & Salah, K. (2022). Blockchain-assisted

MFA: Trends and future directions. IEEE Access, 10,

16545– 16561. https://doi.org/10.1109/ACCESS.2022

.3148865

Cai, Z., & He, Y. (2024). Behavioral biometrics in MFA for

mobile security. Journal of Network and Computer

Applications, 224, 103945. https://doi.org/10.1016/j.

jnca.2024.103945

Chatterjee, S., & De, S. (2022). Hybrid multi-factor

authentication for cloud platforms. Future Generation

Computer Systems, 132, 34– 48. https://doi.org/10.1

016/j.future.2022.01.002

Duan, J., Wang, J., & Ren, K. (2021). Analyzing MFA in

defending against credential stuffing attacks. IEEE

Transactions on Information Forensics and Security,

16, 5120– 5133. https://doi.org/10.1109/TIFS.2021.3

113052

Farooq, M., & Alshamrani, A. (2023). MFA in

cybersecurity compliance frameworks: A critical

review. Computers & Security, 128, 102755.

https://doi.org/10.1016/j.cose.2023.102755

Ferreira, A., & Antunes, L. (2021). Improving user

experience in MFA: Balancing security and usability.

Information Security Journal: A Global Perspective,

30(3), 105–116.

https://doi.org/10.1080/19393555.2021.1922203

Gao, J., & Liu, Y. (2022). Comparative study on MFA

technologies for mobile banking security. IEEE

Transactions on Mobile Computing, 21(6), 2022–2036.

https://doi.org/10.1109/TMC.2021.3119257

Ghorbani, A. A., & Salah, K. (2021). MFA as a service:

Opportunities and challenges. IEEE Internet

Computing, 25(1), 70– 77. https://doi.org/10.1109/M

IC.2020.3039137

Hong, S., & Kim, H. (2023). Quantum-resistant MFA

systems: Next-generation cybersecurity. IEEE

Transactions on Quantum Engineering, 4, 1–13.

https://doi.org/10.1109/TQE.2023.3230214

Hussain, R., & Sohail, H. (2022). Machine learning

approaches for adaptive MFA. Applied Sciences, 12(9),

4562. https://doi.org/10.3390/app12094562

Jang, H., & Kim, S. (2024). MFA against deepfake-based

identity threats. IEEE Transactions on Emerging

Topics in Computing, 12(1), 156– 167. https://doi.o

rg/10.1109/TETC.2023.3295401

Li, L., & Xu, Y. (2021). Authentication resilience: The role

of MFA in cyber resilience frameworks. International

Journal of Information Management, 58, 102322.

https://doi.org/10.1016/j.ijinfomgt.2020.102322

Liu, Q., Zhang, X., & Li, W. (2022). Smart MFA systems:

Integrating AI in authentication frameworks. IEEE

Systems Journal, 16(3), 3957– 3968. https://doi.org/

10.1109/JSYST.2021.3051234

Malik, H., & Rizvi, S. (2022). Policy implications of MFA

adoption in critical infrastructure. Government

Information Quarterly, 39(2), 101670. https://doi.or

g/10.1016/j.giq.2021.101670

Park, S., & Kim, Y. (2023). Biometric MFA: Challenges

and future prospects. Sensors, 23(2), 512.

https://doi.org/10.3390/s23020512

Saleh, M., & AlZain, M. (2024). MFA adoption barriers in

healthcare cybersecurity frameworks. Health and

Technology, 14(2), 431– 442. https://doi.org/10.

1007/s12553-024-00729-x

Tan, K., & Lim, J. (2021). A novel OTP generation scheme

for secure MFA systems. Security and Communication

Networks, 2021, 5567993. https://doi.org/10.1155/2

021/5567993

Zhao, Z., & Zhang, X. (2023). The evolution of MFA

technologies in response to cyber threats. Computer

Standards & Interfaces, 86, 103673. https://doi.org/

10.1016/j.csi.2022.103673

Design and Evaluation of a Multi-Domain Adaptive Multi-Factor Authentication Framework for Holistic Cybersecurity Reinforcement

157