Healthcare devices are the virtual lifelines of today’s healthcare systems and are commonly used to avoid, track, or diagnose diseases. However, these devices have also become an easy target for cyber intrusions due to software and design-related vulnerabilities. Ironically, due to cyber-attacks, devices meant to protect the health of patients’ are now becoming a major health threat. Only one among such breaches is the troubling instance of intruders gaining easy access to insulin pumps [1–5]. An exponential increase in attacks on the entire healthcare sector has been reported in the last few years. Security experts suggest that the high demand and cost of healthcare data on the dark web may be the explanation for this increase [6–9]. Several attacks on healthcare devices are carried out, risking not only the effectiveness of the system but also corrupting the data. For example, any running window system can easily target custom worm and infect more than 100 MRI machines at a time.

The FDA drafted a study that examined the security of healthcare devices in 2012 [2] to increasing episodes of breaches. The FDA issued a set of guidelines on the security of healthcare devices’ for manufactures in 2013 [3]. Researchers, developers, and manufacturers are imminently needed to work on more foolproof mechanisms to resolve the threats to healthcare devices. Attempts to incorporate improved security measures right from the healthcare device design and development process would be a pre-emptive step in this direction [10–15].

In addition to the security of healthcare devices, the credibility of the devices [16–21] is yet another fact that needs to be improved. The security element of the device varies from the safety of the healthcare device’s design. Although safety focuses on the hazardous circumstances that may arise accidentally and intensely, the security of the device from modifications to the data contained in the device may be made. Attacks on the healthcare network and network of healthcare devices threaten both the security of devices and the safety of information [22–25]. Due to the network breach of healthcare devices, research studies performed in this area have found many vulnerabilities and safety issues. Implantable Medical Devices (IMDs) [26], wearable devices [27,28] and surgical robots [29] are some of these devices. There are network flaws in the hospital network configuration [30–33], a third party service provider networks (like, laboratories, pharmacies, etc.) [34,35].

All these events allow the attackers to enter the network and take control. To gain access to the healthcare devices and steal the credentials [36–39] and exploit the vulnerabilities the attackers use the network. It is important to remove the efficacy that is sometimes introduced into healthcare devices due to design or software development to maximize the effectiveness of the healthcare devices [40–43]. The healthcare device credibility is thus an essential research premise and assumes greater significance in the efforts to make the medical devices more secure and reliable [44,45].

Against this context, the current research aims to examine the different security checking mechanisms and methodologies that are currently being used to determine the security of healthcare devices. The analysis then draws on the experts’ views on the security of healthcare devices as inputs for the implementation of the proposed methodology. We will outline the relevant method, security examination mechanisms, and methodology for determining the security of the healthcare devices. As an insight into the methodology, we apply the expert’s opinion about the security of the medical devices. And the debate on the empirical analysis of the method and sensitivity analysis of the empirical discussed.

Healthcare devices are effectively implanted and connected to the body of the patients and collect the body’s sensitive data [46]. The device gathers the patient’s activity of the body and sends it to the experts and laboratories through the network for processing. Network associated, low-power sensor-based devices are more vulnerable to intrusions. Many manufacturers want to supply embedded devices that use less power at a low cost. Therefore, additional safety mechanisms are not favored, which would make the device complex and raise the costs of manufacturing [47–49]. For example, when making on-site medical devices such as X-ray, MRI, and Ultrasound, etc., keeping the device secure and protecting it from hackers is not the priority of the developers.

Developer’s concentrate on the working of the device, as most of them, do not even know about the safety and precautions. In a healthcare device, parts such as hardware and outdated software also invite vulnerabilities in the device that are easy for hackers to exploit [50,51]. Reverse engineering techniques can manipulate on-site devices and easily breach the security and availability of devices. The software and hardware layers are also part of the security needs [52,53]. Integrity, availability, authentication, confidentiality, safety and privacy, unauthorized tampering, are the essential features for developing highly secure healthcare devices. Due to traditional security algorithms that cannot be used because of implantable and sensor devices, maintaining the security of the devices is a tough task challenge.

But in recent years, refreshers have built a new algorithm for the CIA that resolve integrity and security issues. These are not ideal for all forms of cyber-attacks because of certain bars. The on-site healthcare devices (MRI, X-ray, and ultrasound) are also vulnerable to cyber-attacks [26–29]. Various techniques for preserving the security of the devices have been developed. Such approaches have been elucidated in Fig. 1.

Some real-world issues demand unique or multi choice-based solutions that are crucial for real users to conduct without any solid base. To tackle this situation and give an ideal quantitative solution to these issues the adopted MCDM approaches are implemented by various researchers. Specifically adopted AHP approach combined with fuzzy linguistic term set theory is more effective and simple in the comparison of others it is evident from various previous research initiatives [5,9]. If there is more than one option available for evaluation in the technique during the computation process, then this situation influences the calculated results even more strongly. In the context of the proposed article authors adopt a hesitant fuzzy set-based MCDM approach that gives an extra efficiency in results for evaluation. Besides, the TOPSIS approach has been used to assess the security of healthcare devices results obtained. Besides, to get more productive and accurate results, this study adopts the hesitant fuzzy approach. Moreover, for testing the evaluated results adopted the methodology of TOPSIS is the most promised and effective approach available in MCDM approaches. The biggest beneficial advantage of this methodology is that it gives a positive and negative both impact evaluation in the same evaluation and considers it in the calculation.

The authors followed the hesitant fuzzy set approach [15] when decision-makers find the possibility or situation of choosing any other value for numbering besides previously used ones. This type of situation opens a possibility of hesitant value use in evaluation which is prepared and discussed firstly by a study [28] and then modified and more systematically explained by Algarni et al. [14].

In cloud-based security architecture, Torra et al. [28] presented a TOPSIS integrated methodology that yielded successful results. For this article, the methodology adopted can skip and handle ambiguities and other AHP-TOPSIS methodology issues. Besides, by evaluating stock selection on paper, the model often validates its results. Similarly, Torra et al. [28,29] in their study have used the same approach. The authors verified the reliability of this technique by objectively analyzing the security of healthcare devices. Besides, Xu et al. [30] and Kumar et al. [31] have also used the specified technique to generate convincing results for their study in the sense of the energy solution.

In our research, HF-AHP methods were enlisted to estimate the priority of the healthcare device security factors, and then we tested their approach HF-TOPSIS on alternatives for similar factors [32]. A phase by phase methodology, in brief, is discussed below:

Phase 1: The first step in the implemented approach is the hierarchy development of factors.

Phase 2: In Tab. 1, examiners use linguistic terminology to create accurate and beneficial assessment criteria for the decision-makers.

Phase 3: The next step in technique evaluation is the adoption of fuzzy wrappers [29] from Eq. (1).

(1) OWA(a1,a2,…an)=∑j=1n⁡Wjbj

Same as experts evaluate the trapezoidal numbers C~=(a,b,c,d) by the Eqs. (2)–(5) after Eq. (1).

(2) a=min{aLi,aMi,aMi+1,……aMj,aRj}=aLi

(3) d=max{aLi,aMi,aMi+1,……aMj,aRj}=aRj

(4) b={aMi,ifi+1=jOWAw2(amj,…..ami+j2),ifi+jisevenOWAw2(amj,…..ami+j+12),ifi+jisodd}

(5) c={aMi+1,if i+1=jOWAw2(amjamj−1,…..am(i+j)2),ifi+jisevenOWAw2(amj,amj−1…..am(i+j+1)2),ifi+jisodd}

After imposing the Eqs. (3)–(5), the experts decide the first and second form of weights η, i.e., the number between [0, 1] and Eqs. (6)–(7) applied by the experts to obtain these numbers.

1st type weights (W1= ( w11,w21,……..wn1 )):

(6) w11=η2,w21=η2(1−η2),…….wn1η2(1−η2)n−2

2nd type weights (W2=(w12,w22,……..wn2)):

(7) w12=η1n−1,w22=(1−η1)η1n−1

The numerical form for the highest rank in the formula η1=g−(j−1)g−1 s, and η2=g−(j−1)g−1 is g and lowest, highest rank factors are shown by i and j, respectively.

Phase 4: Eqs. (8)–(9) are used by the experts after evaluating the entire previous approach to satisfy the remaining comparison matrix attributes. Thereafter, experts use Eq. (10) to defuzzify the matrix to determine the comparison matrix.

(8) A~=[1⋯c~1n⋮⋱⋮c~n1…1]

(9) c~ji=(1ciju,1cijm2,1cijm1,1cij1)

(10) μx=l+2m1+2m2+h6

Phase 5: The phase of defuzzification provides correct values. The experts examine the Consistency Ratio (CR) by applying the Eqs. (11)–(12) to analyse the CR of these values.

(11) CI=γmax−nn−1

(12) CR=CIRI

Phase 6: In this step, by Eq. (13), the experts assess the geometrical mean of the values.

(13) r~i=(c~i1⨂c~i2……⨂c~in)1/1nn

Phase 7: The most significant criterion in the entire set is evaluated by experts by applying the Eq. (14).

(14) w~i=r~1⨂(r~1⨁r~2…….r~n)−1

Phase 8: Examiners analyze the defuzzified values by Eq. (15).

(15) μx=l+2m1+2m2+h6

Phase 9: By applying the Eq. (16), experts transform the defuzzified values into normalized values or weights.

(16) w~i∑i⁡∑j⁡w~j

Now after identifying priority list for selected attributes the second adopted methodology of TOPSIS is used for testing the effectiveness of obtained results. TOPSIS is effective as a MADM technique in recommending the most preferred option for use. The definition of the TOPSIS approach was introduced by Lai et al. The synthesis of positive and negative ideas is the TOPSIS methodology; the most accurate and effective option is the most precise and reliable factor. The worst option, on the other hand, is an irrelevant factor. The authors utilized the hesitant fuzzy AHP TOPSIS approach to test and assess the security of healthcare devices [14]. The TOPSIS method associates the distance between two linguistic values such as H1s and H2s and performs its computations. Below, the procedure has been clarified (Eq. (17)):

(17) d(H1s,H2s)=|q∗−q|+|p∗−p|

Phase 10: The following terms are described as the starting process:

The following written formulas are applied as (C={C1,C2,…..CE}) and n criteria (C={C1,C2,…..Cn}) to define alternatives and criteria in TOPSIS.

Similarly, k is used to show the numeric count of experts in TOPSIS ex Denotes the experts.

The equation X~l=[HSijl]E×n is used in TOPSIS technique to represent HF matrix.

The standards are written for TOPSIS to determine the criteria and effect of outcomes:

The standard for TOPSIS evaluation lies in between very poor and very good scale,

r11 = between medium and good (bt M&G)

r21 = at most medium (am M)

r12 = at least good (al G)

r22 = between very bad and medium (bt VB&M)

For HF matrix, the following formulas are used [9]:

envF (EGH (btM&G)) = T (0.3300, 0.5000, 0.6700, 0.8300)

envF (EGH (amM)) = T (0.0000, 0.0000, 0.3500, 0.6700)

envF (EGH (alG)) = T (0.5000, 0.8500, 1.0000, 1.0000)

envF (EGH (btVB&M)) = T (0.0000, 0.3000, 0.3700, 0.6700)

Phase 11: By applying the Eq. (18) formula, the associated combined matrix is created:

Tpij=min{mini=1K(maxHtijx),maxi=1K(minHtijx)}

(18) Tqij=max{mini=1K(maxHtijx),maxi=1K(minHtijx)}

Phase 12: The effective factor where most effective factor is indicated by Aj, is shown by alpha in the TOPSIS evaluation, and alpha shows the cost-related preferences. In addition, the latest efficient alternatives need high precision for cost related preferences. The following Eqs. (19)–(22) are used to define and compare cost as well as effective factors:

(19) V~pj+=maxi=1K(maxi(minHSijx)) j∈αb and mini=1K(mini(minHSijx)) j∈αc)

(20) V~qj+=maxi=1K(maxi(minHSijx)) j∈αb and mini=1K(mini(minHSijx)) j∈αc)

(21) V~pj−=maxi=1K(maxi(minHSijx)) j∈αc and mini=1K(mini(minHSijx)) j∈αb)

(22) V~qj−=maxi=1K(maxi(minHSijx)) j∈αc and mini=1K(mini(minHSijx)) j∈αb)

Phase 13: Experts evaluate TOPISIS +ve and −ve concepts components by applying following Eqs. (23)–(24).

(23) D+=[d(x11,V~1+)+d(x12,V~2+)+…d(x21,V~1+)+d(x22,V~2+)+…d(xm1,V~1+)+d(xm2,V~1+)+…+d(x1n,V~n+)+d(x21,V~n+)+d(xmn,V~n+)]

(24) D−=[d(x11,V~1−)+d(x12,V~2−)+…d(x21,V~1−)+d(x22,V~2−)+…d(xm1,V~1−)+d(xm2,V~1−)+…+d(x1n,V~n−)+d(x21,V~n−)+d(xmn,V~n−)]

Phase 14: Experts build and assess the closeness of positive and negative factors evaluated by Eqs. (25)–(26).

(25) CS(Ai)=Di+Di++Di−,i=1,2,….m

where

(26) Di+=∑j=1n⁡d(xij,Vj+) and Di−=∑j=1n⁡d(xij,Vj−)

Phase 15: The ranks are allocated to conclude the process, and the tabular forms of options are focused on their assessment of effectiveness.

In further parts of this study, a highly detailed and evaluated numerical assessment of healthcare device security has been conducted.

In this section, authors discussed the analysis of results of the proposed method, and compare the proposed method with existing methods to verdict the proposed method advantages.

The security of healthcare devices is compromised at the execution time by the method of data transfer, data storage, and migration process. By updating the software patch, using the hardware security guards and network encryption techniques all these problems can be addressed. Healthcare devices hold confidential information relating to the health and personal data of patients. We have established an approach for quantitative assessment of the security of healthcare devices from the proposed article through HF-AHP.TOPSIS approach in our framework, which is the best decision making and ranting approach. The decision-makers allocated the rating of the healthcare devices based on their security using this approach. The study has enlisted the help of 45 experts in different fields of security. Based on their experiences, they ranked the healthcare devices accordingly. Finally, on the provided data for performance assessment of the healthcare devices, HF.AHP.TOPSIS was applied. The findings of this research work as shows:

Most researchers work on the security of healthcare devices, but do not have sufficient guidelines for the development and design of the software and security of the device.

In the current situation, dependability on healthcare devices has improved enormously, more so in the aftermath of a health emergency such as the COVID-19 pandemic when home quarantine instead of attending hospitals was recommended to patients. For health care monitoring and treatment, doctors and patients alike depend on medical devices. Healthcare devices submit data from patients to physicians who prescribe the course of care after the data has been checked. However, the confidentiality of the data and system is under consideration. Even a small difference in the data of the patient can lead to an incorrect diagnosis, thus endangering the health and well-being of the patient. Quantitative and automated evaluation of the security of medical devices is an efficient solution for ensuring the security of the healthcare device. In the above observation, the A2 alternative obtains the highest ranks among the best alternatives. In the current analysis, this was done with the help of the HF-AHP.TOPSIS approach. Among the different alternatives, this method is best for decision making and provides corroborative results. This framework is well validated and tested; manufactures may use a tested approach to security checking to protect the healthcare devices.