Desktop version

Home arrow Computer Science

  • Increase font
  • Decrease font


<<   CONTENTS   >>

II: Designing IoT-Based Smart Solutions for Monitoring and Surveillance

Smart Interfaces for Development of Comprehensive Health Monitoring Systems

DINESH BHATIA

Department of Biomedical Engineering, North Eastern Hill University, Shillong, Meghalaya 793022, India, E-mail: This email address is being protected from spam bots, you need Javascript enabled to view it

ABSTRACT

With the rapid increase in the older population coupled with enhancement in their life span, the number of patients who require continuous monitoring rises tremendously. It would lead to higher costs of hospitalization and patient care globally. Hence, the requirement of smart interfaces and systems for observing health may be employed in lessening the hospitalization stay, weight on clinical staff, counseling time, holding up records, and, in general, social insurance costs. Different smart interfaces frameworks are characterized into three subcategories: remote health monitoring system (RHMS), mobile health monitoring system (MHMS), and wearable health monitoring system (WHMS). The RHMS alludes those with remote access or frameworks to communicate back and forth information or multiple patient parameters from a remote location or region. MHMSs refer to mobile phones, personal digital assistants, and pocket-personal-computer-based monitoring systems, which are utilized as the principle preparing station or at times as the primary working modules. Tire RHMS and MHMS are considered to be more advantageous and practical than the traditional institutional care mechanism. They empower patients to stay at then respective locations while getting access to proficient healthcare. WHMSs refer to wearable gadgets or biosensors that can be worn by patients comprising of WHMS, RHMS, and MHMS. Shrewd health monitoring systems are referred to as trendsetting innovations with regard to patient’s continuous health monitoring (R. Roine, A. Ohinmaa, and D. Hailey, “Assessing telemedicine: A systematic review of the literature,”

Can. Med. Assoc. J., vol. 165, no. 6, pp. 765-771, 2001). They comprise smart gadgets that could be employed to address several health-related issues. The devices measure heart rate, blood pressure, electrocardiogram, oxygen saturation levels, body temperature, and respiratory rate. This chapter would explore the development of smart interfaces for available comprehensive health monitoring systems built on artificial intelligence tools, cognitive computing systems, and machine learning algorithms.

INTRODUCTION

In today’s advanced technological era, continuous health monitoring plays a vital role due to the rapid rise in the elderly population that enjoys a long life span. It is more prominent due to the increase in a large number of nuclear families, wherein the elderly adults are living separately from their children. Continuous health monitoring of such a population would lead to higher costs of hospitalization and patient care globally. In developed countries such as the USA, the death rate for the older population is more than 770,000 persons eveiy year. Mistaken diagnosis, measurement errors, and delay in intercessions lead to increased hospitalization. The treatments costs are ranging between $1.5 billion and $5 billion every year [2]. Hence, development of smart healthcare interfaces and systems for continuously observing the health of such population can play an essential part in reducing the duration of hospitalization, number of caretakers, counseling tune, and maintaining records with reduced social insurance costs. Smart healthcare frameworks could be classified as remote health monitoring system (RHMS), mobile health monitoring system (MHMS), and wearable health monitoring system (WHMS). The RHMS alludes those with remote access or frameworks to communicate to and fro information from a remote location or region. This framework can help in monitoring single or multiple parameters covering different patient symptoms employed at the user homes and a doctor’s clinical facilities. MHMSs could be referred to mobile phones, personal digital assistants, and pocket-personal-computer-based monitoring systems, which are utilized as the principle preparing station or at times as the primary working modules [17]. RHMSs and MHMSs are thought to be more advantageous and practical than traditional institutional care systems. They empower patients to stay at their respective locations while availing better healthcare facilities [17,22]. WHMSs may refer to as wearable gadgets or biosensors that can be worn by patients comprising of WHMS, RHMS, and MHMS. Shrewd health monitoring systems (HMSs) consist of smart gadgets to address multiple health-related issues and referred to as trendsetting innovations in continuous health monitoring [34,35]. General HMSs refer to systems that monitor different parameters and general symptoms. The devices measure heart rate (HR), blood pressure (BP), electrocardiogram (EGG), oxygen saturation (SpO,) levels, body temperature, and respiratory rate (RR) [39]. This chapter discusses the development of smart interfaces for comprehensive HMSs built on artificial intelligence tools, cognitive computing systems, and machine learning algorithms.

LITERATURE SURVEY ON THE CURRENT STATE-OF-THE-ART MONITORING SYSTEMS

Rapid development in the cutting-edge health monitoring procedures and techniques in the past decade has enabled healthcare experts to precisely monitor grown-ups or adults in connection to age-related diseases such as dementia, Alzheimer's, and Parkinson's [7,25,30]. Since there are no confinements to HMS applications, they can be utilized in the clinic [10], home [14], and outdoor settings using either global positioning system [39] or radio frequency identification technology [9]. In spite of slow advancement of innovation, there are worries with regard to the nature of medical information, the security of patient data, reliability of sophisticated monitoring systems, ease of usage, adequacy by the therapeutic staff and patients, and the recurrence of false alerts [10]. However, various studies and research investigations over the past two decades have primarily diminished such worries. For example, Imhoff and Kuhls [16] have figured out that up to 90% of all alarms in critical care monitoring are false positives. Different scientists proposed measures to lessen these false positives by incorporating adjusting the scope of parameters, decreasing limit esteems, or joining a period delay in producing such alerts [40].

7.2.1 WEARABLE HEALTH MONITORING SYSTEMS

A smart vest [6] is an example of a WHMS, which is a wearable physiological monitoring framework, fused in a jacket. The device is an assortment of different biological sensors coordinated into the piece of clothing’s texture that gathers biosignals in a noninvasive and simple way without causing subject discomfort or pain. The parameters estimated by the vest comprise of ECG, photoplethysmography (PPG), HR, BP, body temperature, and galvanic skin reaction. The patient’s ECG can be recorded without using electrode jelly and is free from benchmark confusion and development collectibles due to execution of high pass, low pass, and notch filters in the device. BP is measured from acquired ECG noninvasively by employing either the auscultatory or oscillometric methods. Results from approval preliminaries affirm the precision of estimated physiological parameters. LOBIN [26] presented an e-textile wearable wireless healthcare monitoring system comprising of sensors to record ECG, HR, and body tempera tine. Similarly, Blue Box [4] developed a novel hand-held device capable of collecting and wirelessly transmitting vital cardiac parameters such as ECG, PPG, and bioimpedance. It could assist in measuring patient’s RR intervals and QRS duration, HR. and systolic time intervals, as well as assessing their values in correlation with cardiac output measured by an echo-Doppler. An in-shoe device was developed by Saito et al. [20] to monitor plantar pressures under real-life conditions. A pressure-sensitive conductive elastic sensor measures plantar weight and approval performed by the F-scan framework. SMARTDIAB [29] device intends to help the monitoring, administration, and treatment of patients with type 1 diabetes mellitus by combining with a patient unit and patient administration unit. The pilot form of the SMARTDIAB was actualized and assessed in a clinical setting. TELEMON [11] is an electronic informatics-telecom and versatile framework permitted automatic and ongoing telemonitoring by mobile correspondences for monitoring the essential indications of constantly sick elderly patients.

DESIGN METRICS AND DESIGN FLOW OF HEALTH MONITORING SYSTEM DESIGN

The design of the above-discussed systems is governed by a specific set of rules called design metrics [37] that were defined for the ease of design engineers and helped in improving the overall design of an HMS. The rules are applicable for any healthcare device as well as for any other embedded system. However, some additional constraints that are specific to the HMS design are discussed with design goals. The design metrics are explained in detail in the following sections.

7.3.1 DESIGN METRICS

The design of a system is influenced by parameters [ 19], for example, consider the design of house floor if we want to improve the quality of the flooring, by improving the quality of the tiles. It will increase the cost of the building. Additionally, consider the computational system in which decreasing the price of a processor may lower the processing capability. Similarly, for designing health or any other embedded systems, parameters are interdependent and affect the performance of the design. Increasing one parameter may decrease or compromise on the other aspect [36]. These parameters are referred to as design metrics and have shown conflicting requirements that need to be addressed before any system design. During the designing stage, a system design engineer tries to find out an optimal set of solution for these metrics. Earlier, the design metrics solution was evaluated after the prototype development of the device, which was a time-consuming, stressful, costly, and tedious process. At that stage, if constraints were not as per our expectations and required modification, the design engineer v'as bound to change the developed prototype or model. However, nowadays, availability of advanced system simulation software allows us to evaluate the design metrics of the concerned system by creating a virtual model [24]. The following points and Figure 7.1 show's design metrics for a modem HMS, winch needs to be care- fiilly optimized during any design development process.

Design metrics constraints involved in the design of an HMS

FIGURE 7.1 Design metrics constraints involved in the design of an HMS.

7.3.2 POWER CONSUMPTION

Nowadays, in the era of wearable technologies, the amount of battery and pow'er embedded within the device is limited [31]. It necessitates engineers to develop devices that are capable of operating with ultralow-power consumption. The amount of power dissipated by the device requires careful evaluation before marketing. Consider a commonly employed method in cardiac cases, namely, a pacemaker. If the pacemaker under development has better accuracy, although it requires frequent charging, the demand for it may be quite low. Therefore, design engineers need to incorporate a longer power life in their design to allow the recommending doctors to suggest the device to the patients. As per the design metrics depicted in Figure 7.1, a better model of the invention is expected to have a longer life of the embedded power source.

Evaluation of the power requirements of the healthcare devices (illustrated using pacemaker battery life)

FIGURE 7.2 Evaluation of the power requirements of the healthcare devices (illustrated using pacemaker battery life).

To improve the power requirement of HMSs, before implementing them on hardware level, a design engineer typically assesses the power consumed by different modules, components, connecting wires, energy consumed in wireless transmission and reception of the data, etc., via simulating the design on simulation software [38]. This simulation software has the real-time behavior of components, modules encoded into functions, and hence, via accessing them, the power of the system can be estimated [34]. By looking at the requirement, specific to the application, we can adjust the value by replacing the components through suitable alternatives. Figure 7.2 shows how the life of the embedded power sources in implantable pacemakers has increased over the years by using different design methods and would further improve by the year 2020. Several other ways to increase the batteiy life of the healthcare devices and monitoring systems are to reduce the number of sensors, to use appropriate fabrication technology, to decrease the size of transmission and reception data bits, etc. [3].

7.3.3 SIZE

Another constraint that affects the suitability and popularity of an optimized HMS is the size of the system [19]. For a design engineer, the size of the system does not mean physical dimensions but measured in bits for any software design and the number of gates for any hardware design. An engineer always tries to reduce the size concerning these two parameters [18]. For this purpose, several optimizing techniques and software have been developed by the researchers, which evaluate their design before actual implementation. The functionality coding of the models in hardware descriptive languages such as Verilog, VHDL, etc., allows a designer to evaluate and optimize the requirement of the gate without implementing it on hardware. Other techniques employed for minimizing the size are logic optimization, technology scaling, device modeling, etc. [3]. Figure 7.3 illustrates the development using the dimensions of different electrocardiogr aph machines, from the first Einthoven’s ECG machine to modem EGG machine available today.

7.3.4 NONRECURRING COST

Besides, the power requirement and size of the nonrecurring cost (NRC), the cost involved in the designing of the system is considered to be an essential parameter. Again, this is a conflicting cost that depends upon the number of units to be manufactured. If the number of units to build is more, then we can compromise with higher NRC if it can reduce die manufacturing cost in a considerable amount [19]. Hence, the design engineer needs to put their efforts in reducing the NRC without degrading the other constraints. The alternatives to lessen the NRC are design reuse, shared design programs, design modification, etc. [3].

7.3.5 PERFORMANCE

Performance is a broader categoiy of constraints for HMSs. Performance is the ability of the device to correctly present the parameters of interest with minimum delay [19]. To get the idea of the device performance, engineers employ several advanced design software tools such as MATLAB, Lab View, etc., to evaluate the performance of the design before use [36]. It helps the designer to improve the performance of their device and other design metric constraints. However, the larger goal for the engineer is to find out an optimized metrics for their application.

 
<<   CONTENTS   >>

Related topics