Platform-Based Design of a Smart 12-Lead Electrocardiogram Device by Using Multiple Criteria Decision-Making Methods ^†

Abstract

Smart telemedicine represents an innovative application of information and communication technology within the healthcare sector, encompassing healthcare delivery, disease management, public health surveillance, education, and research. The commercialization of 5G and the extensive adoption of the Internet of Things (IoT) enable smart telemedicine devices to mitigate geographical and transmission delays, hence enhancing the quality of treatment provided to individuals. Although intelligent medicine is significant, previous studies emphasize the implementation and adoption of systems or technologies with few studies conducted on the platform of smart telemedicine equipment. This study aims to address the research gap by forecasting future developments and delineating smart telemedicine device designs utilizing platform-based design. We introduce a hybrid multi-criteria model that delineates the components of the intelligent medical platform. A portable 12-lead electrocardiogram (ECG) system is used by a global telemedicine technology company to assess the viability of the suggested framework. The portable 12-lead ECG device integrates artificial intelligence (AI), cloud computing, and 6G technology. The results of this study provide a basis for product creation by other smart telemedicine companies, while the platform-based analytical methodology can be employed for future product design.

1. Introduction

Smart telemedicine is an innovative application of information and communication technology across various dimensions of healthcare, including disease management, public health surveillance, education, and research. With the increasing availability of 5G and the growing prevalence of the Internet of Things (IoT), smart telemedicine devices efficiently mitigate geographical and transmission delays, hence enhancing healthcare for individuals. Platform-based design is a multifaceted methodology that meets diverse market demands across industries by emphasizing efficiency, modularity, and systematic approaches. A platform must have the ability to address diverse market segments and meet particular performance goals [1]. Platforms employ supplements including shared subsystems and components. This facilitates the development of platforms customized for a particular market sector and easily modified for other segments or elevated tiers within the same segment [1]. A platform-based design is utilized for the development of smart medicine and telemedicine products in various geographic regions or economic sectors. Although intelligent medicine is significant, the implementation and adoption of systems or technologies are emphasized, with limited research on the platform of smart telemedicine equipment.

Research and development costs can be reduced when the probability of failure is considered for business stakeholders. We utilized the knowledge and expertise of telemedicine specialists and adopted innovative methodologies such as decision-making trial and evaluation laboratory (DEMATEL) and DEMATEL-based analytic network process (DANP) to derive critical components. Furthermore, we employed the VlseKriterijumska Optimizacija I Kompromisno Resenje (VIKOR) method, utilizing an established platform for intelligent medicine and telemedicine equipment to ascertain the potential smart healthcare system.

In this study, an empirical investigation was employed to develop portable 12-lead electrocardiogram (ECG) equipment with improved practicality. The portable 12-lead ECG devices enhance telemedicine and smart medicine, providing essential advantages for the diagnosis and monitoring of cardiovascular disorders. These devices are economical, portable, and proficient in high-quality signal capture for remote patient care. A portable ECG device records 3-channel 12-lead signals and delivers findings over email, demonstrating accuracy and transmission speeds [2]. A finger-ring-shaped ECG sensor captures 12-lead ECGs by placing the ring on various body regions, facilitating wireless data transmission for timely interventions [3]. Moreover, a 12-lead ECG patch facilitates continuous monitoring through machine learning, improving power efficiency and data transfer [4]. These developments underscore the integration of telemedicine and smart medicine, enhancing cardiovascular treatment and patient outcomes in telemedicine. Smart medical equipment is appropriate for the empirical case study due to its inherent compatibility with a smart telemedicine apparatus.

In this study, we reviewed the literature on intelligent healthcare and platform architecture. Then, multiple criteria decision-making (MCDM) methods were introduced on platform-based design. The designed 12-lead ECG device on a platform can fulfill the market needs for telemedicine. Also, the proposed analytic framework was verified.

2. Smart Medical Devices

2.1. Smart Medical Devices

Smart medical devices represent a significant advancement in healthcare, incorporating cutting-edge technology to enhance the processes of diagnosis, treatment, and patient monitoring. These devices represent innovations in materials science, electronics, and sensor technology to deliver wearable, insertable, and implantable solutions that monitor health conditions and administer medicines with enhanced efficacy.

The development of flexible electronic sensors, compatible with biological systems and capable of detecting physiological signals from the skin, exemplifies the potential of intelligent healthcare devices. These sensors are integrated into wearable devices that exhibit properties of elasticity, self-power generation, and multifunctionality. This integration improves the practicality of the devices in personal healthcare [5]. The integration of smart devices on eHealth platforms enables preventative healthcare by allowing users to monitor their activities, engage with communities, and communicate with healthcare practitioners. The incorporation revolutionizes the relationship between patients and healthcare providers and benefits health management [6]. In biomedical sensors, the implementation of transducer electronic data sheets (TEDS) enables the storage of patient-specific information. This also allows devices to adjust their diagnostic algorithms according to individual requirements, resulting in enhanced accuracy and reliability in diagnoses [7].

The collection of health-related data by smartphones and smartwatches presents the potential of these devices in clinical practice. These devices provide information on activity levels and heart rate measures, which can be used to enhance patient care [8]. Such advances emphasize the revolutionary influence of smart medical devices on healthcare, providing tailored, accurate, and proactive medical solutions that improve patient care and results.

2.2. 12-Lead ECG Device

A 12-lead ECG equipment consists of multiple components to accurately gather and process cardiac signals. The components consist of the electrodes and leads, which are responsible for sensing the cardiac electrical activity. The ECG system requires a belt and limb straps equipped with wet Ag/AgCl electrodes for the best contact between the skin and electrodes which is needed to obtain reliable signals [9].

The signal acquisition subsystem utilizes instrumentation amplifiers and analog-to-digital converters (ADCs). The signal processing unit incorporates digital signal processing (DSP) to filter and improve the ECG signals. The ECG recording system employs frequency division multiplexing (FDM) and twin DC servo loops to effectively handle and analyze the signals, resulting in excellent resolution and minimal noise [10].

In addition, the system monitors heart rate variability (HRV) and other physiological signals. The enhanced heterogeneous oscillator model employs improved van der Pol and FitzHugh–Nagumo equations to accurately mimic genuine ECG waveforms, including pathological states [1]. The user interface and data transmission subsystems are crucial components for ensuring usability and enabling remote monitoring. Mobile or web-based applications are frequently employed to manage the device, exhibit the ECG information, and transmit it to healthcare practitioners for examination, as observed in the ECGraph and STM32F-microcontroller-based systems [10,11]. The combination of these subsystems guarantees that a 12-lead ECG equipment accurately gathers, processes, and sends cardiac data for efficient diagnosis and monitoring.

The advent of portable 12-lead ECG equipment has greatly enhanced telemedicine and smart medicine, offering crucial advantages in the diagnosis and monitoring of cardiovascular diseases. These devices are specifically engineered to have a low price, be easily carried, and have the ability to acquire signals of excellent quality. They must be accessible and efficient for providing medical treatment to patients who are not physically present. The portable ECG equipment possesses such telemedicine integration. This device captures three-channel 12-lead signals and transmits the results through email. This technology exhibits exceptional precision and rapid transmission speeds, hence improving diagnostic capabilities and increasing the availability of medical services [2].

2.3. Platform-Based Design

Platform-based design is a cutting-edge method that utilizes a shared infrastructure to improve efficiency, flexibility, and scalability in house building, digital architecture, and clinical trials. Platform-based design simplifies system design by organizing it into numerous layers, each representing distinct levels of abstraction and refinement. This method enables designers to concentrate on high-level functionalities without being overwhelmed by low-level implementation. In this context, a platform collects components and connections, which are described by models that define their capabilities and performance indicators. This allows for a structured design process [12]. In electronics, platform-based design improves the functioning and adaptability of electronic systems. It also promotes sustainable practices and fosters the creation of novel technologies in several domains.

3. Research Methods

In selecting elements for the 12-lead ECG equipment, two methods were reviewed: DANP and modified VIKOR. At first, the direct influence relationship matrix $\mathit{A}$ is established. To measure the influence relationships between criteria, a direct influence relationship matrix A is established [13].

$$\mathit{A}=\left[\begin{array}{ccccc}{a}_{11}& \dots & a_{1j}& \dots & a_{1n}\\ ⋮& \ddots & ⋮& \ddots & ⋮\\ a_{i1}& \dots & a_{ij}& \dots & a_{in}\\ ⋮& \ddots & ⋮& \ddots & ⋮\\ a_{n1}& \dots & a_{nj}& \dots & a_{nn}\end{array}\right]$$

(1)

$$i,j=1, 2,\dots ,n$$

Based on the matrix A, Equation (2) yields the normalized matrix N.

$$\begin{array}{c}\mathit{N}=\phi \mathit{A}\\ \phi =\mathrm{m}\mathrm{i}\mathrm{n}\{1/\underset{i}{\max }{\displaystyle \sum _{i=1}^n}{a}_{ij},1/\underset{j}{\max }{\displaystyle \sum _{j=1}^n}{a}_{ij}\}\end{array}$$

(2)

After calculating the N matrix, the total influence relationship matrix T is computed using (3).

$$\begin{array}{ll}\mathit{T}& =\mathit{N}+{\mathit{N}}^2+{\mathit{N}}^3+\cdots +{\mathit{N}}^{\mathit{\theta }}\\ & =\mathit{N}{\left(\mathit{I}-\mathit{N}\right)}^{-1}\\ \theta & \to \mathrm{\infty }, \mathit{I} \mathrm{is} \mathrm{the} \mathrm{Identify} \mathrm{Matrix}\end{array}$$

(3)

where Tc represents the transpose of the total influence matrix T, denoted as T^t, the matrix Tc is expressed by (4). The process of normalizing matrix Tc involves assuming a normalization baseline q_j, represented by (5). By dividing the values of matrix Tc by q_j individually, the normalized matrix ${{\mathit{T}}_c}^N$ is obtained, as expressed in (6).

$${\mathit{T}}_{\mathit{c}}=\left[\begin{array}{ccccc}{t}_{11}& \dots & t_{1j}& \dots & t_{1n}\\ ⋮& \ddots & ⋮& \ddots & ⋮\\ t_{i1}& \dots & t_{ij}& \dots & t_{in}\\ ⋮& \ddots & ⋮& \ddots & ⋮\\ t_{n1}& \dots & t_{nj}& \dots & t_{nn}\end{array}\right]$$

(4)

$$q_{\mathit{j}}=\sum _{\mathit{j}=1}^{\mathit{n}}{\mathit{t}}_{\mathit{i}\mathit{j}}$$

(5)

$${{\mathit{T}}_c}^N=\left[\begin{array}{ccccc}{\displaystyle \frac{t_{11}}{q_1}}& \dots & {\displaystyle \frac{t_{1j}}{q_j}}& \dots & {\displaystyle \frac{t_{1n}}{q_n}}\\ ⋮& \ddots & ⋮& \ddots & ⋮\\ {\displaystyle \frac{t_{i1}}{q_1}}& \dots & {\displaystyle \frac{t_{ij}}{q_j}}& \dots & {\displaystyle \frac{t_{in}}{q_n}}\\ ⋮& \ddots & ⋮& \ddots & ⋮\\ {\displaystyle \frac{t_{n1}}{q_1}}& \dots & {\displaystyle \frac{t_{nj}}{q_j}}& \dots & {\displaystyle \frac{t_{nn}}{q_n}}\end{array}\right]$$

(6)

The weighted supermatrix W is obtained by multiplying the matrix ${{\mathit{T}}_c}^N$ by dimensions’ weights. The limiting supermatrix can be computed using (7).

$$\underset{p\to \infty }{\mathit{lim}}{\mathit{W}}^p$$

(7)

Criteria are not independent from each other. If simply rely on the additive property of simple weighting, we cannot discern the importance of each criterion. Therefore, the VIKOR is adopted to evaluate and select the most suitable elements. Then, the L_p-metric for modified VIKOR is calculated as follows:

$$L_l^p=\left\{\sum _{j=1}^n[w_j\left(\left|f_j^{*}-f_{lj}\right|\right)/(\left|f_j^{*}-f_j^{-}\right|\left)\right]\right\}$$

(8)

$$f_j^{*}=max{f}_{lj}; f_j^{-}=\min f_{lj}.$$

After that, S_j and Q_j is calculated as follows:

$$S_j=\sum _{j=1}^n\left[{\displaystyle \frac{w_j\left(\left|f_j^{*}-f_{lj}\right|\right)}{\left|f_j^{*}-f_j^{-}\right|}}\right]$$

(9)

$$Q_j=\mathrm{m}\mathrm{a}\mathrm{x}\left[{\displaystyle \frac{w_j\left(\left|f_j^{*}-f_{lj}\right|\right)}{\left|f_j^{*}-f_j^{-}\right|}}\right]$$

(10)

Finally, the index value R_l is derived accordingly, as in (11).

$$\begin{array}{l}{R}_l=\frac{v\left(S_l-S^{*}\right)}{S^{-}-S^{*}}+\frac{\left(1-v\right)\left(Q_l-Q^{*}\right)}{Q^{-}-Q^{*}}\\ S^{*}=\min S_l, S^{-}=max{S}_l\\ Q^{*}=\min Q_l, Q^{-}=\max Q_l\end{array}$$

(11)

4. Empirical Results

The empirical study was conducted on a 12-lead EKG/ECG platform provided by a Taiwan-based biotech medical company. The analytic framework was verified by using the platform. The DANP-based VIKOR framework was used to evaluate and select the most appropriate configuration.

4.1. Empirical Case Introduction

The multinational biotech medical company cooperated in this study. The company was established in August 2013 with an initial capital of USD 2.01 million. The company obtained a marketing license in Taiwan and 510(k) approval from the US Food and Drug Administration (FDA). A 510(k) is a submission of the FDA before marketing a product to prove that the device is safe and effective as a legally marketed device. The company’s major products are portable medical grade 12-lead ECG and services for cardiovascular disease. The company is the only one approved by the FDA for medical-grade 12-lead ECG devices for home use. The device enables cardiovascular disease screening and diagnosis.

4.2. Cloud-Based Platform

The cloud-based platform for the 12-lead ECG device consists of the 6G module, Bluetooth, Ethernet, Wi-Fi, high-resolution displays, microSD, cloud storage and computation, AI module, eLearning module, and edge computing module (

Table 1. Subsystems of 12-lead ECG device.

Component	Symbol
6G Module	SS1
Cloud-Based Computation and Storage	SS2
AI Module	SS3
Digital Learning Module	SS4
Edge Computation Module	SS5

) [14].

Based on the VIKOR results (

Table 2. Selected subsystems for the 12-lead ECG machine.

Subsystem	Score	Rank
6G Module	0.740	3
Cloud-Based Computation and Storage	0.466	2
AI Module	0.000	1
Digital Learning Module	0.854	4
Edge Computation Module	1.000	5

), the most appropriate modules for the 12-lead ECG devices were selected in terms of AI, cloud-based computation, and storage, as well as 6G.

5. Conclusions

Smart telemedicine is an advanced information and communication technology in healthcare. It enables disease management, public health monitoring, education, and research. Smart telemedicine devices are used to overcome geographical limitations and transmission delays, therefore improving care to patients, thanks to the emergence of 5G and the IoT. Most previous research has been conducted on the implementation and deployment of individual systems or technologies without a thorough investigation of the design of smart telemedicine devices. Therefore, we reviewed smart telemedicine devices and their structures by employing a platform-based design methodology. The viability of the suggested framework was assessed using the portable 12-lead ECG platform. Based on empirical evidence, scenarios for the advancement of smart telemedicine platforms were created. The result showed the importance of the integration of robust technology, a supportive environment, and legal backing, and that of robust technology and a supportive environment, which does not need legal support. The integration of poor technology and environment is also necessary with legal support. The portable 12-lead ECG device needs to incorporate AI, cloud computing, and 6G modules in the first and third scenarios. In the second scenario, the system needs to integrate AI, 6G technology, and digital learning modules. The findings provide useful information for smart telemedicine companies to develop future products. The platform-based method serves as a basis for designing next-generation smart telemedicine devices.