Sage Journals: Discover world-class research

Abstract

The implementation of the Internet of things in healthcare is a promising challenge to achieve coverage for a bigger number of users in different places at lower costs. Internet of things might mean better technology regarding response time and proper control of medical parameters. This study introduces an Internet-of-things system for healthcare with possibilities to control medical variables according to recent breakthroughs in sensors and data processing. The goal of the system is to optimize the development of applications to obtain variables in real time and with less energy consumption. The proposed model is validated on the measurement and monitoring of oxygen saturation, heart rate, and body temperature in patients with respiratory disorders. This was achieved by the optimization of data acquisition, integrated into a secure architecture using Message Queuing Telemetry Transport protocol. A cloud architecture with interconnection to low-cost and open-source devices was implemented, which interconnect to the sensors and actuators’ network. The experimental results were statistically treated against the device pattern data, through hypothesis tests for mean differences to probe the accuracy of the model. Finally, the proposed model demonstrates an efficient performance in several clinical parameters, such as oxygen saturation and heart rate per minute.

Keywords

Healthcare monitoring Internet of things respiratory disorder model

Introduction

Nowadays, the technology used in healthcare areas is exponentially growing; however, acquisition of optimal medical data and safety have become a challenge. The Internet of things (IoT) is the most recent and optimal technology in health areas for three reasons. First, IoT allows for the control and monitoring of previously diagnosed illnesses. Second, it helps to complement other technologies such as through sensors and high-response medical equipment in real time. Third, users suffer from health deterioration. This last advantage is presented in useful and practical applications to be available for various patients. It is a great approach to developing applications on self-care to prevent respiratory disorders due to highest global socioeconomic impact in home care.

Various related studies have been undertaken to control and mitigate respiratory disorders. Dymerski et al.¹ presented an electronic nose technique in the field of fast diagnostics of patients suspected of chronic obstructive pulmonary disease. The technique uses a simple electronic nose prototype equipped with a set of six semiconductor sensors to record chemical signals. These chemical signals are a mash of references of different volatile organic compounds. As a result, the relevance of using electronic devices in chronic obstructive pulmonary disease cases was obtained. Fernandez-Granero et al.,² subsequently, presented a pilot study of early detection of acute exacerbations. Through machine learning techniques and with an electronic sensor, a scenario was designed to monitor breathing in elderly people.³ This study achieved detection rates of 75.8%, with an average of 5 ± 1.9 days in advance at medical attention. This and other studies validate the use of emerging technologies to optimize respiratory disorder detection, control, and prevention.

Blood oxygen saturation (SpO₂) can be measured using the analysis of infrared light patterns in a finger, an earlobe, or a toe. This parameter allows to determine if body cells are receiving enough oxygen percentage.⁴ An oxygen level under 90% or 85% can result in respiratory disturbances, cognitive disturbances, and gastrointestinal problems.⁵ Improving the precision for SpO₂ measurement is one of the greatest challenges faced; different methods have been implemented to this purpose. Some are based on the sampling per division of frequency using two-level pulses (bi-level pulsed frequency-division multiplexing (BPFDM)), others are based on time-division multiplexing (TDM) and on simple frequency-division multiplexing (FDM).⁶ The importance of measuring this parameter has led to the development of different types of devices for its measurement and monitoring.⁷ The use of communication and computer devices to develop measurement and monitoring modules for SpO₂ levels, heart rate, temperature, and other vital signs has contributed to guaranteeing the safety of patients in different ways and places.⁸ These personal monitoring systems are not only used when doing physical activities, but they also operate in other states from the daily life of a patient, for example, their implementation in a portable and automated sleep apnea detector to measure SpO₂.⁹

The breakthroughs in the devices connected to the network have managed to increase IoT applications that are currently limited by the structure inflexibility from the existing network. Creating new services and more efficient applications to cover the real requirements is a challenge for this type of systems.¹⁰ IoT systems and their applications in the monitoring of vital signs can be divided into three large groups: monitoring, control, and data analysis, all these services are subject to possible security vulnerabilities.¹¹ To provide privacy and security to these new developments, light protocol designs have been developed to control IoT devices, external devices, and the effective central server against conventional attacks without the need of a highly computational capacity to implement measurements and controls in the operation.¹²

The growing number of applications that use virtual and physical devices, which, in turn, constitute monitoring, homecare and self-care systems, may condition or interfere with the functioning of other elements or systems in their operating and surrounding areas.¹³ This is a consequence of the need to use different frequencies for different wireless devices in a limited range of frequencies without causing interferences, restriction or limitation.¹⁴ However, any device can generate electrical noise, which can be transmitted through physical means, which, in turn, can interfere with other devices that are in the same means, system or surroundings, through transients and waves that are over the limits of the proper functioning of a device.¹⁵ Due to the previous premise, electromagnetic compatibility (EMC) is an essential requirement applied to electronic devices to guarantee non-interference toward the environment.^16,17

In Colombia, healthcare attention to patients with respiratory disorder is high. In fact, it is the seventh highest cause of mortality.¹⁸ Consequently, for countries with large geographical distances to health centers and a wide digital gaps, IoT technology convergence is a promising solution to detect and prevent respiratory disorders by means to pulmonary function tests (PFTs) using specialized spirometry equipment. In 2014, a medical cloud-based platform for respiration rate measurement and hierarchical classification of breath disorders was introduced. This study shows the development of Hub-IoT to record and process respiratory disorder signals, but not intended to replace PFTs but to improve patient monitoring to prevent respiratory disorders.¹⁹ Real-time monitoring of oxygen saturation patterns is shown through body sensor networks.²⁰ Signals are recorded through oxygen saturation patterns and body temperature. Records are available in real time to access health data via cloud computing. Data remote accessibility allows the specialist to have complete tracking at time intervals, averages, critical and historical values, among others.

Materials and methods

IoT architectural design is based on signal acquisition through sensors and, then, processed using a programmable card. Database (DB) signals are then recorded in a Hub-IoT cloud server. Hub-IoT is a technology service for multiple programming languages DB access used in this study to acquire, save, process, and analyze by means of an application programming interface (API). Optimization capabilities for application development rely on communication channels, protocols (IoT, transport, and priorities), low resource-consumption technology, among others. A performance effectiveness test application is then developed (see Figure 1). The main acquisition module includes a high-efficiency wireless personal area network (WPAN) which is composed of low energy consumption sensors, and it also has a discontinuous control system to clinical parameters.

Figure 1.

IoT general architecture to data acquisition.

WPAN module is composed of two sensors interconnected to an e-health card developed. The first is a digital Expense body temperature sensor. The second sensor is a CMS50DL Pulse Oximeter. For this device, a synchronization protocol for e-health communication is established prioritizing SpO₂. E-health cards allow users to access biometric and medical applications where body monitoring takes place. E-health cards support the use of nine multiple sensors: heart beat, blood oxygen SpO₂, air flow (breathing), body temperature, electrocardiogram (ECG), glucometer, galvanic skin response (GSR)—sweating, blood pressure (sphygmomanometer), and patient position sensor (accelerometer). This allows for an efficient use of sensors, thus generating low energy consumption during data transmission. This work uses both body temperature thermometers and pulse oximeters to measure heart rate and oxygen saturation.

Wireless sensor networks nowadays provide considerable benefits over traditional approaches for multiple applications, including intelligent homes, health attention, environmental monitoring, and safety.^21,22 Sensors are connected to everyday objects (doors, appliances and, even, the human body). Wireless body area network (WBAN) allows for medical attention applications that are in an early development stage, but offers valuable contributions at monitoring, diagnostic and therapeutic levels. However, the comparison of technologies and protocols on WBANs allows to analyze the development of new monitoring architectures for the collection of medical information in real time with secure data communication and low energy consumption.²³

To search, low-error range and comfortable-for-the-patient sensors that range human body temperature, oxygen saturation, and heart rate are examined. Tables 1 and 2 show multiple temperature, and heart rate and oxygen saturation sensors analyzed under different specifications, respectively.

Table 1.

Body temperature sensor specification.

Sensor	Temperature range	Voltage	Accuracy	Type
LM53	−55 to 150°C	4–30 V	±0.5°C	No linear
Testo 830-T4	−30 to 400°C	9 V	±1.5°C	Linear
Exsense	−40 to 85°C	9 V	±0.1°C	Negative coefficient

Table 2.

Heart rate and oxygen saturation sensor specification.

Sensor	Pulse range	Voltage	Accuracy	Type
Datex ohmeda	0–100	2.6 to 3.6 V	±1%	Oximeter
Vitacare	0–100	2.6 to 3.6 V	±1%	Oximeter
JZX-301	0–100	2.6 to 3.6 V	±2%	Pulseoximeter

Then, the environmental conditions during testing stage: room temperature of 21°C and a relativity humidity of 19%. The environmental conditions are not considered to change the patient diagnostic because the sensors are connected directly to the patient. E-health platform carries out an analogue-to-digital converter (ADC) process and, then, data may be used digitally. However, sensors acquire data analogously and include an ADC module.

Transmission module is composed of Arduino Uno (1) and Raspberry Pi Model 3B (Rpi3B) programmable cards. Arduino is an open-source electronics prototyping platform based on hardware and software.²⁴ This device conducts serial communication with the Raspberry over baud rate of 9600. Rpi3B programmable card is a small host connecting multiple sensor interfaces to Internet through multiple processing languages. Its main function is to be subscribed to the cloud broker and send data received from the Arduino. All data are then recorded in an SQL language DB, based on Not Only Structured Query Language (NoSQL) DB has better performance than the model on SQL DB.²⁵ To analyze the data collected, an end-user application is developed in Java language. A graphic interface shows the sensor values on considerable times of acquisition.

Message Queuing Telemetry Transport (MQTT) was the communication and control protocol used over IoT. This protocol was selected based on the analysis conducted on a decision model to work with IoT protocols based on cloud computing platforms.^26,27 MQTT capabilities, management, and safety were decisive to carry out communication between broker, sensors, and applications. However, variables control was implemented using an Android application connected to the broker, thus allowing to add buttons, modify topics, publications, and subscriptions as well as to create a personal configuration per person, handling MQTT protocol, users, and passwords. Figure 2 shows the final architecture used in the analysis scenario (MQTT protocol) with temperature, oxygen saturation, and heart rate sensors. For the evaluation of magnetic compatibility, reference was made to the tests performed with the devices used in.²⁸ The above results are within the ranges and standards of compliance established in.¹⁴ For the above reason, we consider that the results obtained and disclosed by are a benchmark and apply to the edge devices used in the implementation of the IoT system.^28,29

Figure 2.

IoT architecture with tele-health of Group Telemedicine Research, TIGUM center on real scenario.

Performance analysis

It is important to facilitate aggregate sensor traffic with variable report times and to control data traffic with respect to channel capacity based on frame size during transmission.¹⁵ Direct real-time optimization is carried out during sensor processing with definite periods in algorithms according to traffic and use of small frames toward Hub-IoT, DB, and final user. Thus, Hub-IoT is similar to Hub-Sensor because both have an environment to support IoT-related application and service development.²⁸ For algorithm optimization over bandwidth and efficiency over data transmission, see application flow diagrams. Figure 3 and 4 shows algorithms implemented.

Figure 3.

Algorithm deployment on programmable Arduino Uno (1).

Figure 4.

Algorithm deployment on programmable Rpi3B.

When serial communication receives values containing letters O, T, and P, the program decides to send data from oxygen saturation, temperature, and heart rate sensors recording the current data of the sensor connected to the Arduino.

Rpi3B card is programmed to satisfy user needs regarding particular data or a specific sensor; all this in accordance with e-health parameters. Depending on the sensor and data, it transmits a letter: “T” is transmitted to identify temperature, “O” to identify oxygen saturation, and “P” to identify heart rate data the sensor is reading at that moment.

After entering the topic message, the program decides which sensor to use and, then, sends such data to the Arduino, which, in turn, returns sensor value through serial communication. Thanks to recent technology advances, it is now easier to own low-cost computers, for example, Rpi cards, Intel Galileo, and small credit-card like devices, thus allowing for low-cost IoT with higher resources.

New studies about expediting Internet connection via MQTT communication protocol are based on low bandwidth use, low data weight, and information safety configuration, thus permitting speed communication between sensors and Arduino and Rpi3B e-cards. In this study, MQTT user device can publish date to the broker-configured subscribed cloud with the purpose of monitoring data or send such data to the other subscribed broker-configured subscribed cloud. This study was based on protocol and architecture to bring IoT proposed.¹⁴ Figure 5 shows sensor connectivity to client process.

Figure 5.

Sensors connectivity to client process.

Data sent by sensors can be seen in applications, PCs, or mobile applications, where ongoing applications can be seen while recording data from subscribed topics under a TCP/IP protocol.

Results

Technical system validation

Initial tests based on data recording and statistical analysis are conducted in the following order: heart rate, oxygen saturation, and finally, temperature (see Figures 6 –8).

Figure 6.

Heart rate: (a) IoT system and (b) Agilent M3046Am device.

Figure 7.

Oxygen saturation: (a) IoT system and (b) Agilent M3046Am device.

Figure 8.

Body temperature: (a) IoT system and (b) Begut MT 402 device.

The statistical treatment applied to the experimental samples obtained from the measurements was a hypothesis test for the mean differences, where a null hypothesis with the sample mean ${\bar{x}}_{1} = {\bar{x}}_{2}$ and an alternative hypothesis with sample mean ${\bar{x}}_{1} {\bar{x}}_{2}$ were considered. To carry out the procedure, $n_{1} = n_{2} = 838$ was considered in the 17 sample intervals for each volunteer

σ^{2} = \frac{(n_{1} - 1) s_{1}^{2} + (n_{2} - 1) s_{2}^{2}}{n_{1} + n_{2} - 2}

$s_{1}^{2} and s_{2}^{2}$ are the sample standard deviations, $σ^{2}$ is the demographic variance, and the bias for this case is 0. A distribution t with a significance level $α = 0.05$ was used, through which the statistical critical value of $t_{α} = 1.960$ was obtained

t = \frac{({\bar{x}}_{1} - {\bar{x}}_{2}) - δ}{\sqrt{\frac{σ^{2}}{n_{1}}} + \frac{σ^{2}}{n_{2}}}

Table 3 shows the measurements obtained in the 838 heart rate samples made in 17 measurement intervals for each of the volunteers, and it also shows the standard deviation and the statistical operator: in the case of volunteer 1, t = 0.5234, where $t < t_{α}$ ; in the case of volunteer 2, t = 1.623.

Table 3.

Heart rate outcome on IoT system versus Agilent M3046A.

Heart rate	$\bar{x}$	s	$σ^{2}$	t
Volunteer 1
Agilent M3046A	92.9439	4.3665	18.8369	0.5234
IoT System	92.8329	4.3137	18.8369	0.5234
Volunteer 2
Agilent M3046A	82.0656	3.7072	13.8168	1.6231
IoT System	81.7709	3.7269	13.8168	1.6231

IoT: Internet of things.

Tables 4 and 5 show that t values obtained for SpO₂ (0.6964, –0.3248) and for body temperature (1.0436, 0.6964), are less than $t_{α}$ , indicating that, for these cases, they are in the acceptance zone of the null hypothesis. Therefore, it can be stated that the mean difference is statistically significant.

Table 4.

SpO₂ outcome on IoT system versus Agilent M3046A.

SpO₂	$\bar{x}$	s	$σ^{2}$	t
Volunteer 1
Agilent M3046A	94.5131	2.8823	8.2729	0.6964
IoT System	94.4153	2.8702	8.2729	0.6964
Volunteer 2
Agilent M3046A	94.3437	2.8718	8.1674	−0.3248
IoT System	94.3890	2.8438	8.1674	−0.3248

IoT: Internet of things.

Table 5.

Body temperature outcome on IoT system versus Begut MT 402.

Body temperature	$\bar{x}$	s	$σ^{2}$	t
Volunteer 1
Agilent M3046A	35.0647	0.6221	8.2729	1.0436
IoT System	35.0328	0.6278	8.2729	1.0436
Volunteer 2
Agilent M3046A	94.5131	2.8823	8.2729	0.6964
IoT System	94.4153	2.8702	8.2729	0.6964

IoT: Internet of things.

Discussion

The system proposed in Chan et al.,⁸ evaluated the ease of use, reliability, and validity of measurement from a pulse oximeter, through the design of customized applications in smartphones, during the sessions of exercise and rest from the users with and without lung disease. The IoT system proposed collects the measurements from the same parameters in any type of session, not only in exercise and rest. In addition, the proposed IoT system has a strong computing architecture in the cloud that allows for better monitoring and data analysis.

The design shown in Son et al.⁴ has a similar design to the one proposed regarding the functioning, and this design proposes a mobile device for the monitoring in real time of vital parameters, such as SpO₂ and heart rate, but does not include body temperature, nor actuators for remote control. It also considers the open-source and open-hardware principles, but does not include low energy consumption devices. The standard deviations shown for contrasted heart rate (6.7, 5.9, 4, 4.5, 5.1), shown in Table 3 (4.3665, 4.3137, 3.7072, 3.7269), indicate that the IoT system proposed accomplishes more significance in heart rate measurement. For the case of SpO₂,⁴ it shows the following deviations (1.1, 4.4, 2, 1.2, 4), compared to the ones obtained by our IoT system (2.8823, 2.8702, 2.8718, 2.8438), and this comparison shows a similar average value for the deviations.

The Web-based system for the measurement and monitoring of heart rate and temperature shown by Raharja and Wijaya⁷ contains the Web visualization of parameters, just like the IoT system proposed in this article (http://tigum.umng.edu.co/tigum/?page_id=593). However, the visualization only shows data in real time, but does not show historic data for monitoring or alerts. However, the system proposed shows historical data. The system shown by Raharja and Wijaya,⁷ just like the proposed IoT system, considers the use of open hardware, but only the proposed IoT system considers the use of low energy consumption devices. The average standard deviations from Raharja and Wijaya⁷ are heart rate = 2.48 and temperature = 0.63, and for the proposed IoT system, the average standard deviations are heart rate = 4.028 and temperature = 0.6245. Finally, the system⁷ does not consider an actuators’ network for remote control.

Conclusion

The proposed IoT system used the measurement of oxygen saturation in the blood (SpO₂) as the main medical parameter, which has been playing an increasingly important role in medical monitoring and health detection. This measurement can become a control tool for patients with respiratory disorders, even when the patient is in an unconscious state.

The use of authentication, authorization, and encrypted communication based on MQTT protocol and TLS Handshake Protocol caused that the connection was the operation that takes longer.

When the client (into IoT) is already connected, the resources it consumes and the time it takes to conduct in both cases are practically the same and do not represent a significant impact in the performance of the MQTT broker.

Even if the inflexibility from the traditional network architectures is inefficient, new ways of using infrastructure and technological communications in health are required. Open, low-cost, and low energy consumption devices can lead to more extensive monitoring periods for greater control using IoT platforms.

The implementation of this type of monitoring systems allows for the control of parameters in the environment of a patient due to their size, low energy consumption characteristics, and their compatibility with other systems. In addition, it allows for the analysis of historic data and behaviors, since it has a computing and storage capacity in the cloud systems that allows for the analysis and processing in short response times. These systems can generate alerts, control signals, or new diagnoses for patients in different environments and places because of the analysis and monitoring.

IoT technology has provided efficient solutions to health management, thus reducing care attention costs and increasing accessibility to medical data, patient safety, and time optimization in care services. All this has been achieved through multiple connections between institutions, patients, and technologies. Research outcomes enhance the benefits of IoT when monitoring and controlling vital signs.

The proposed IoT system can operate via Bluetooth in very low consumption mode, activated with ultra-low-power (ULP) protocols, with consumption under 10 mA, thus extending battery usage time. Besides, this also allows the IoT system to continue with its monitoring and control functions for the user without having an alternative power source for its energy supply during sleep states. This results in a low-cost tool that can benefit the daily life of users with respiratory disorders, thanks to the historic data and analysis available for possible diagnosis and subsequent treatments.

Footnotes

Handling Editor: Francesc Pozo

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the Universidad Militar Nueva Granada (Project code: IMP-ING-2660) and by the special cooperation agreement No. 002-IMP-ING-2660 of 2018 between USACH (Chile) and UMNG (Colombia).

ORCID iD

Leonardo Juan Ramírez López

References

Dymerski

Gębicki

Wiśniewska

et al . Application of the electronic nose technique to differentiation between model mixtures with COPD markers. Sensors 2013; 13: 5008–5027.

Fernandez-Granero

Sanchez-Morillo

Leon-Jimenez

Computerised analysis of telemonitored respiratory sounds for predicting acute exacerbations of COPD. Sensors 2015; 15(10): 26978–26996.

Sun

Tian

Qiu

et al . The influence of pre-existing health conditions on short-term mortality risks of temperature: evidence from a prospective Chinese elderly cohort in Hong Kong. Environ Res 2016; 148: 7–14.

Son

Thu

Kien

. Design an IoT wrist-device for SpO2 measurement. In: 2017 international conference on advanced technologies for communications (ATC), Quy Nhon, Vietnam, 18–20 October 2017, pp.144–149. New York: IEEE.

Kocher

Rohling

Tschupp

Performance of a digital PCO 2/SPO 2 ear sensor. J Clin Monit Comput 2004; 18(2): 75–79.

Song

Jiang

Hao

et al . Use of bi-level pulsed frequency-division excitation for improving blood oxygen saturation precision. Measurement 2018; 129: 5239.

Raharja

Wijaya

. Design of monitoring tool heartbeat rate and human body temperature based on WEB. In: MATEC web of conferences 2018, vol. 164, article no. 01017. EDP Sciences, https://doi.org/10.1051/matecconf/201816401017

Chan

Inskip

Kirkham

et al . A smartphone oximeter with a fingertip probe for use during exercise training: usability, validity and reliability in individuals with chronic lung disease and healthy controls. Physiotherapy. Epub ahead of print 26 October 2018. DOI: 10.1016/j.physio.2018.07.015.

Mendonça

Mostafa

Morgado-Dias

et al . A portable wireless device based on oximetry for sleep apnea detection. Computing 2018; 100(11): 1203–1219.

10.

Valdivieso Caraguay

Benito Peral

Barona Lopez

et al . SDN: evolution and opportunities in the development IoT applications. Int J Distrib Sens N 2014; 10(5): 735142.

11.

Lee

The Internet of things (IoT): applications, investments, and challenges for enterprises. Bus Horizons 2015; 58(4): 431–440.

12.

Wang

Ding

Han

et al . Secure and efficient control transfer for IoT devices. Int J Distrib Sens N 2013; 9(11): 503404.

13.

Michalak

Szafrańska

ME.

EMC in IoT world. Przegląd Elektrotechniczny 2018; 94: 50–53.

14.

Fukumoto

Kajimoto

Key challenges for Internet of things in consumer electronics industry. In: H

(eds) Artificial intelligence and robotics. Cham: Springer, 2018, pp.229–237, https://doi.org/10.1007/978-3-319-69877-9_25

15.

Aswin

Sundar

Electromagnetic interference (EMI) and electromagnetic compatibility (EMC) in real time applications—survey. Int J Mech Eng Technol 2018; 9(6): 474–482, https://www.iaeme.com/MasterAdmin/uploadfolder/IJMET_09_06_054/IJMET_09_06_054.pdf

16.

Veraguas

JP.

Compatibilidad electromagnética y seguridad funcional en sistemas electrónicos. Barcelona: Marcombo, 2010.

17.

Baligar

Lawrence

. The future of immunity testing. In: 2018 IEEE international symposium on electromagnetic compatibility and 2018 IEEE Asia-Pacific symposium on electromagnetic compatibility (EMC/APEMC), Singapore, 14–18 May 2018, pp.135–138. New York: IEEE.

18.

Ministerio de Salud y Protección Social. Enfermedad Pulmonar Obstructiva Crónica—EPOC, 2013, Colombia, https://www.minsalud.gov.co/Documentos%20y%20Publicaciones/epoc.pdf

19.

Fekr

Janidarmian

Radecka

et al . A medical cloud-based platform for respiration rate measurement and hierarchical classification of breath disorders. Sensors 2014; 14(6): 11204–11224.

20.

Enright

PL.

Oxygen desaturation during a 6-min walk identifies a COPD phenotype with an increased risk of morbidity and mortality. Eur Respir J 2016; 48: 1–2.

21.

Rüdiger

Yuce

Real-time performance of a self-powered environmental IoT sensor network system. Sensors 2017; 17(2): 282.

22.

Lee

Ouyang

Intelligent healthcare service by using collaborations between IoT personal health devices. Int J Bio-Science Bio-Technol 2014; 6(1): 155–164.

23.

Filipe

Fdez-Riverola

Costa

et al . Wireless body area networks for healthcare applications: protocol stack review. Int J Distrib Sens N 2015; 11(10): 213705.

24.

Miranda

Cabral

Wagner

et al . An open platform for seamless sensor support in healthcare for the Internet of things. Sensors 2016; 16(12): 2089.

25.

Liu

Cai

A dynamic processing system for sensor data in IoT. Int J Distrib Sens N 2015; 11(8): 750452.

26.

Guillén

Ruiz

Ubaque

. Decision model to deploy IoT solutions on cloud computing based platforms. In: Proceedings of the international multiconference of engineers and computer scientists 2017, vol. 2, Hong Kong, 15–17 March 2017, http://www.iaeng.org/publication/IMECS2017/IMECS2017_pp556-561.pdf

27.

Guillén

Sánchez

López

. IoT protocol model on healthcare monitoring. In: VII Latin American Congress on Biomedical Engineering CLAIB 2016, Bucaramanga, Santander, Colombia, 26–28 October 2016, pp.193–196. Singapore: Springer.

28.

De Miguel-Bilbao

Aguirre

Lopez Iturri

et al . Evaluation of electromagnetic interference and exposure assessment from s-health solutions based on Wi-Fi devices. Biomed Res Int 2015; 2015: 1–9, http://dx.doi.org/10.1155/2015/784362

29.

Jang

A scheme to smooth aggregated traffic from sensors with periodic reports. Sensors 2017; 17(3): 503.

Internet of things in healthcare monitoring to enhance acquisition performance of respiratory disorder sensors

Abstract

Keywords

Introduction

Materials and methods

Performance analysis

Results

Technical system validation

Discussion

Conclusion

Footnotes

Declaration of conflicting interests

Funding

ORCID iD

References