2014 2nd International Conference on Information and Communication Technology (ICoICT)
Embedded Gateway Services for Internet of Things Applications in Ubiquitous Healthcare M F A Rasid1, W M W Musa1, N A A Kadir1, A M Noor1, F Touati2, W Mehmood2, 4 L Khriji3, A Al-Busaidi3, A Ben Mnaouer 1
Wireless and Photonics Network Research Centre, Universiti Putra Malaysia, Selangor, Malaysia {fadlee,muzaaliff,nurulazman,amirullah}@upm.edu.my 2 Department of Electrical Engineering, Qatar University, Doha, Qatar
[email protected] 3 Department of Electrical and Computer Engineering, Sultan Qaboos University, Muscat, Oman
[email protected] 4 School of Engineering, Applied Science and Technology, Canadian University of Dubai, UAE
[email protected]
needs for in-home medical monitoring as compared to health facility institutions for the elderly.
Abstract—The continuous advancement in computer and communication technologies has made personalized healthcare monitoring a rapidly growing area of interest. New features and services are envisaged, raising users’ expectations in healthcare services. The emergence of Internet of Things brings people closer to connect the physical world to the Internet. In this paper, we present embedded services that are part of a ubiquitous healthcare system that allows automated and intelligent monitoring. The system uses IP connectivity and the Internet for end-to-end communication, from each 6LoWPAN sensor nodes to the web user interface on the Internet. The proposed algorithm in the Gateway performs multithreaded processing on the gathered medical signals for conversion to real data, feature extraction and wireless display. The user interface at the server allows users to access and view the medical data from mobile and portable devices. The ubiquitous system is exploring possibilities in connecting Internet with things and people for health services.
In addition, the advancement of IPv6 technology coupled with the introduction of 6LoWPAN for short range communications opens up possibilities to introduce a more patient centric healthcare monitoring systems. With IPv6 addresses, each of the physiological sensors can be assigned with an IP address that allows more features to be introduced in real-time patient monitoring. Management of various sensors and later to the concept of multi sensors for multiple patients in a hospital or an elderly institution can be done in the similar way network devices are managed in an enterprise network. The emergence of Internet of Things (IoT) links the physical and virtual objects with the Internet. The 6LowPAN that is part of wireless sensor networks and IPv6 are seen as enablers for IoT. The IoT builds on well-known concepts such as ubiquitous computing, pervasive computing, cyber physical systems, ambient intelligence or technologies such as wireless sensor networks and RFID [6]. According to ITU, The Internet of Things is a technological revolution that represents the future of computing and communications, and its development depends on innovation in a number of important fields [7].
Keywords—ubiquitous health, Internet of Things, 6LoWPAN, medical sensor networks, embedded system.
I.
INTRODUCTION
The growing trends in healthcare applications from homebased care to sports health management drive the interest to incorporate pervasive technologies for embedded intelligent services in healthcare monitoring. The gradual shift of expectations from the traditional hospital-based to the patient centered personalized healthcare services create the demand and opportunities for more automated monitoring applications. Efforts can be seen from the emergence of web applications of e-health to the recent ubiquitous healthcare applications [1-4]. As the general growth in world population for people over the 65 years of age in the coming years, technology can play a significant role in improving the quality of aging life. The population projections from the US Bureau of Censors in 2005 indicated that the number of old people (65-84years old) is predicted to double from 35 million to 70 million by 2025 [5]. This scenario has prompted researchers to develop technologies and applications that can possibly address the
978-1-4799-3580-2/14/$31.00 ©2014 IEEE
In the global scenario, one of the main initiative on IoT is by the EU-funded Seventh Framework Programme (FP7) CASAGRAS project. CASAGRAS or the Coordination And Support Action for Global RFID-related Activities and Standardisation, aims at addressing key issues that are important in providing the foundation and cooperation necessary for realising the Internet of Things [8]. CASAGRAS has defines IoT as a global network infrastructure that links physical and virtual objects through the exploitation of data capture and communication [8]. There is a need to coordinate the convergence of ongoing IoT activities in order to address and facilitate the large potential for IoT-based capabilities. IoT may not need one single killer application, but a lot more applications that are inter-related to each other to make it more useful and solves everyday problems. Application needs to link the technologies and requirement from the society for IoT to be
145
2014 2nd International Conference on Information and Communication Technology (ICoICT)
from each sensor nodes to the web user interface on the Internet.
useful. With this view, an introduction of IoT in healthcare sector may perhaps spurs the expansion and adoption of IoT. This paper therefore presents embedded gateway services that has been developed to introduce IoT applications in healthcare. The work here is part of a 6LoWPAN-based ubiquitous healthcare system presented in [9]. The physiological medical sensors integrated with 6LoWPAN nodes and ubiquitous computing with broadband connectivity provides intelligent services for IoT healthcare applications. The rest of this paper is organized as follows. The types of medical sensors data and the connectivity in ubiquitous healthcare monitoring system are briefly discussed in Section II. The features and services for the embedded gateway in the system for IoT applications are introduced in Section III. Section IV propose and demonstrate the results and finally Section V concludes the paper. II.
Figure 2. Data flow of the system.
III.
UBIQUITOUS HEALTHCARE SYSTEM
Generally a pervasive computing power is envisaged to have the processing capability in medical applications to provide some intelligent, automated point of care and emergency management services. Our Gateway is implemented to collect medical data from medical sensor networks running on 6LoWPAN. Since scalability is always one of the main concerns for any medical monitoring systems, our system is designed to handle multiple nodes with multi sensors that represent multiple patients operation for the monitoring. Since the continuous sensing and processing takes place from low-power, computationally constrained nodes, thus the power consumption and complexity of the processing algorithms can remain at a minimum level. The large amount of collected data from medical sensors continue to increase as new measurements are taken over the monitoring duration. This situation prompts for a reliable data measurement and data management at the Gateway in order to produce high confidence data for further medical diagnosis and treatment. Naturally, the system capacity will work at minimal if it is used for personal monitoring at a home environment.
Various efforts were made since the last decade in wireless biomedical systems and applications [9]. Some works were focusing on wearable medical sensors that looks into hardware efficiency for such healthcare applications. The proposed healthcare system here is set to be used in either at home or at hospital environment. The physiological medical sensors and the connectivity between sensor nodes as well as wide area connectivity is depicted in the overall diagram of the system in Fig. 1. The details of ECG sensor, temperature sensor, accelerometer sensor, 6LoWPAN nodes and edge router are described in length in [10]. Each of the nodes has their own IP address. The communication between nodes and edge router are utilizing 6LoWPAN packets, which later converted to IPv6 packets at the edge router. The role of PC in [10] is enhanced with the introduction of embedded gateway with intelligent features, which will be further described in Section III.
A first lab prototype of the Gateway that provides a wide area mobile communication between sensors and healthcare providers is produced. The current Gateway is running on INTEL processor. The Gateway will be ready to communicate with an edge router that features 6LoWPAN for short range communication with sensor nodes via serial interface. A 3G/4G function is featured as the Gateway’s mobile capability for seamless connection of services for the patient. An open source platform is adopted for the Gateway operation. Other main hardware functions available in the Gateway is depicted in Fig. 3. The Gateway has substantial processing power and memory to handle multiple sensors from many patients. Some of the Gateway services available for the ubiquitous system includes automated connection for reliable communication, auto update and remote reboot for ease of operation and maintenance, on board database for temporary storage of data, wireless display module, listener and sender algorithm that will manage identification and interpretation of patient data as well as QRS detection algorithm for ECG feature extraction. The number of QRS peaks detected and the calculated heart rate in beat per minute are then displayed. Each of the patient will have their own unique patient ID and identifier for each of the sensors connected.
Figure 1. Overall diagram of ubiquitous healthcare system.
The data flow of the system starting from the received medical signals at the edge router to the web application for the users at the server are depicted in Fig. 2. A wireless display module will fetch the data from the Gateway for local display. The proposed ubiquitous healthcare system uses IP connectivity and the Internet for end-to-end communication,
978-1-4799-3580-2/14/$31.00 ©2014 IEEE
GATEWAY FUNCTIONALITY
146
2014 2nd International Conference on Information and Communication Technology (ICoICT)
The data processing will then take place to read and convert raw medical data into real data. In the case of ECG signals, the program will invoke QRS detection algorithm. Since the system is running an embedded application, the choice of the QRS detection algorithm is set for a reasonably less complexity algorithm with high accuracy. A real-time QRS detection algorithm that is based on analysis of slope, amplitude and width of QRS complexes algorithm is adopted [11]. This PanTompkins algorithm has been found to have a higher accuracy for various beat morphologies than other traditional real-time methods developed before 1990 [12]. The algorithm includes a series of filters and methods that perform lowpass, highpass, derivative, squaring, integration, adaptive thresholding, and search procedures for QRS peaks detection. At the same time, medical data will be processed at the Gateway to allow the wireless display to read and display the current data of the patients individually. The wireless display can be viewed from patient’s own smart devices. If the system is handling multiple patients, then each one of them is able to view their own set of medical data at the wireless display simultaneously. All these functionalities and services are multithreaded automated programs pervasively computed in the Gateway that does not require user interactions.
Figure 3. Functional block diagram of the Gateway.
A flowchart of the listener, data processing, wireless display and data uploading program to the server via http connection is shown in Fig. 4. The program is to handle multiple users with multi sensors data and therefore are designed with multithreaded process approach. The listener functions will read raw medical data received at the edge router and parsed the gathered data following a predetermined packet format structure before storing the data in a temporary database in the Gateway.
The Server and web application design diagram for the system is presented in Fig. 5. This is for healthcare management application for the healthcare provider to monitor and provide services to patients. The Gateway will send the data to the server using http connection and later will be stored in the main database. The transmission of the gathered data to the server are encapsulated in a proprietary packet data format. Users of the system (either patients, caretakers, doctors or family members) need to log in into the web user interface with secured password to access relevant information based on access control list (ACL). Since the application is a web based application, the user interface can be accessed through web browsers from various devices, allowing flexibility and mobility to the users.
Figure 4. Flowchart of codes and services in Gateway. Figure 5. Server and web application design diagram
978-1-4799-3580-2/14/$31.00 ©2014 IEEE
147
2014 2nd International Conference on Information and Communication Technology (ICoICT)
IV.
V.
RESULTS AND IMPLEMENTATION
CONCLUSION
In this work, we focused on design and development of an autonomous multithreaded algorithm for an embedded Gateway as part of a ubiquitous healthcare system for IoT applications. Medical signals are automatically gathered using IP-based sensor nodes and later processed by the algorithm in the Gateway for conversion to real data, feature extraction and wireless display. The multi sensors data from multiple patients are then made available to relevant users through anytime anywhere access to the web user interface. The exploitation of IP address and the Internet for communication from patients to medical personnel made it an application that link technologies with society in their everyday life. The ubiquitous healthcare application can be seen as a launching pad towards realizing the future potential of IoT, connecting Internet with things and people.
The embedded services for ubiquitous system was implemented using an open source platform. The Gateway is running an Ubuntu server 13.10 in performing the multithreaded processes described in Section III. The development of the ECG feature extraction algorithm was done in C++ before the final code is embedded in our Gateway application. The detected peaks in the sampled ECG signals are shown in Fig. 6. The flow on how the algorithm performs the peak detection and heart rate calculation is demonstrated in an input-process-output flow in Fig. 7. This algorithm is then compiled and the algorithm function is called as part of multithreaded process in the Gateway. The snapshot of the wireless display of medical signals are presented in Fig. 8.
ACKNOWLEDGMENT This publication was made possible by NPRP grant No [41207-2-474/1/2012] from the Qatar National Research Fund (a member of Qatar Foundation). The statements made herein are solely the responsibility of the authors.
Figure 6. Detected peaks in ECG signals.
REFERENCES [1]
R. Ranjan and S. Varma, “Object-oriented Design for Wireless Sensor Network assisted Global Patient Care Monitoring System,” Int. J. of Comput. Applicat. vol. 45, no. 3, pp. 8–15, May 2012. [2] F. Delmastro, “Pervasive communications in healthcare,” Comput. Commun., vol. 35, no. 11, pp. 1284–1295, Jun. 2012. [3] Touati, Farid, and Rohan Tabish. ”U-Healthcare System: State-of-theArt Review and Challenges,” J. of Medical Syst. 37.3, pp. 1-20, 2013. [4] Muzaaliff W. Musa, Mounir Mokhtari, Borhanuddin M. Ali, M. F. A. Rasid, and Mahmoud Ghorbel, "Seamless semantic service provisioning mechanism for ambient assisted living," in 8th International Conference on Advances in Mobile Computing and Multimedia, Paris, 2010, pp. 119-125. [5] Campbell, P., Current population reports (population projections: States, 1995-2025), pp. 25-1131. Census Bureau, 2005. [6] F. Elgar, “What is the Internet of Things – An Economic Perspective”, Auto-ID Labs, Eth, Zurich, WP-BIZAPP-053, Jan. 2010. [7] ITU, “ITU Internet Reports 2005: The Internet of Things”, Int. Telecommun. Union, Geneva, Nov. 2005. [8] Casagras, Internet of Things [Online]. http://www.iot-casagras.org. [9] Al-Busaidi, A.M. and Khriji, L. ‘Wearable wireless medical sensors toward standards, safety and intelligence: a review’, Int. J. Biomedical Engineering and Technology, Vol. 14, No. 2, pp.119–147, 2014. [10] Touati, Farid, Rohan Tabish, and A. Ben Mnaouer. ”Towards uhealth: An indoor 6LoWPAN based platform for real-time healthcare monitoring.” 6th Joint IFIP IEEE Wireless and Mobile Networking Conf. (WMNC), 2013. [11] J. Pan and W.J. Tompkins, “A real-time QRS detection algorithm”, Biomedical Eng. (BME), IEEE Trans., vol. 32, no. 3, pp. 230-236, 1985. [12] F. Portet, AI. Hernandez and G. Carrault, “Evaluation of real-time QRS detection algorithms in variable contexts”, Med. Biol. Eng. Comput., vol. 43, no. 3, pp. 379-85, 2005.
Figure 7. Input-Process-Output flow of QRS peak detection.
Figure 8. Snapshot of wireless display for medical signals from Gateway.
978-1-4799-3580-2/14/$31.00 ©2014 IEEE
148
