基于3G的远程医疗系统

Abstract—Telemedicine systems have become an important supporting for the medical staffs. As the development of the mobile phones, it is possible to apply the mobile phones to be a part of telemedicine systems. We developed an innovative Remote Patient Monitoring System using a Java-enabled 3G mobile phone. By using this system, doctors can monitor the vital biosignals of patients in ICU/CCU, such as ECG, RESP, SpO2, EtCO2 and so on by using the real-time waveform and data monitoring and list trend data monitoring functions of installed Java jiglet application on the mobile phone. Futhermore, doctors can check the patients' information by using the patient information checking function. The 3G mobile phone used has the ability to implement the application as the same time as being used to mak a voice call. Therefore, the doctor can get more and more information both from the browsing the screen of the mobile phone and the communicating with the medical staffs who are beside the patients and the monitors. The system can be conducted to evaluate the diagnostic accuracy, efficiency, and safety of telediagnosis. I. INTRODUCTION HE rapid development of information and telecommunication technology has brought great revolutions in our everyday lives. The impact to medical field is being felt with many new applications using these technologies. Telemedicine is essentially the use of both information technology and telecommunications to provide health services or support health service provision over a distance. Telemedicine systems can be devided into two fields: one is that the mobile devices are used by patients to measure the medical data of themselves. For example, the European EPI- MEDICS project has developed a Personal ECG Monitor having the capabilities of recording ECG at anywhere and anytime [2]. A Peak flow monitoring system via a mobile phone for asthma is developed in UK[5]. The other field is that the mobile devices are used by medical staffs as an assistant for their daily work. A paramedical system has been developed using PDA to get efficient medical information in India[1]. T During the past decade, several new kinds of mobile devices, such as PDA, mobile phones and some intelligent Pu Zhang, Yuichi Kogure, Hiroki Matsuoka, Masatake Akutagawa, Yohsuke Kinouchi are with the Institute of Technology and Science, the University of Tokushima, Tokushima, 770-8506, JAPAN. (corresponding author to provide phone: +81-88-656-7576; e-mail: zhangpu@ee.tokushima-u.ac.jp) Qinyu Zhang is with ShenZhen Graduate School of Harbin Institute of Technology, ShenZhen, China and wearable medical devices have been designed. Combined to the recent proliferation of wireless communication solutions, this provides new development opportunities for telemedicine. However, there are some inherent limitations of mobile devices, such as memory capacity, CPU speed, small screen area, etc. These make internet access from mobile devices more difficult as compared to that from desktop computers. In Japan, mobile phone is recognized as “Ubiquitous service platform”. The total number of the subscribers of mobile telephone(including PHS) is about 95 million. The big market is shared by three carrier companies. More than 52 million users belong to NTT DoCoMo Inc., while over 27 million users belong to KDDI Corporation, and about 15 million users belong to SOFTBANK Corporation(as of January 31, 2007, according to Telecommunications Carriers Association, Japan). The statistic data mean that almost each person in Japan has a mobile phone in hand. In another aspect, the mobile phones have become more and more powerful. Although remain small screen size of the mobile terminals, the functions, such as memory capacity, CPU speed, Java-enabled applications and camera performance are continuously getting improved. Futhermore, third- generation(3G) mobile communication networks(WCDMA) combine standardized streaming with a range of unique services to provide high-quality Internet contents. The phone is possible to transfer large amount of data, such as photos, movies and application files etc, because of its wide transmission band. This paper presents a novel Remote Patient Monitoring System for doctor. By using a 3G mobile phone, the doctor can monitor the realtime biosignals of patients in ICU/CCU through the screen of the mobile phone while making a voice call. The software developed on the mobile phone can realize three main monitoring functions, including realtime waveform and data monitoring, list trend data monitoring and the patient information checking. The doctor can select which he wants to check using the keyboard of the mobile phone. The whole system has been implemented in the laboratory environment. The main function of the system, i.e., transferring the waveform data and text data of patients to the doctor's mobile phone, has been successfully accomplished as well as other auxiliary functions. The goal of this study is to construct a new bridge between the medical staffs in hospital and the doctor using the mobile phone to discuss and diagnose the patients' state of illness as quickly and accurately as possible to save A Remote Patient Monitoring System Using a Java-enabled 3G Mobile Phone Pu Zhang, Yuichi Kogure, Hiroki Matsuoka, Masatake Akutagawa, Yohsuke Kinouchi, Qinyu Zhang Proceedings of the 29th Annual International Conference of the IEEE EMBS Cité Internationale, Lyon, France August 23-26, 2007. FrP2D2.9 1-4244-0788-5/07/$20.00 ©2007 IEEE 3713 Authorized licensed use limited to: Shenzhen Institute of Advanced Technology. Downloaded on January 29, 2009 at 21:19 from IEEE Xplore. Restrictions apply. patients' lives. II. SYSTEM ARCHITECTURE A. System Overview The overview of the Remote Patient Monitoring System is shown in Fig.1. The system is mainly composed of three parts: patients monitoring system, Remote Information Server built in hospital and a Java-enabled 3G mobile phone. They are connected by the Internet and 3G mobile networks. The patients monitoring system is composed of Bedside Monitor, Central Station Monitor and PC used by Medical staffs in hospital. Conventionally, several patients' information can be collected by the Bedside Monitors and stored in the Central Station Monitor. In the proposed system, the patients' information in the Central Station Monitor, including numeric signal data, text data and realtime waveform data can be extracted by the Remote Information Server and changed into the recognizable pattern for Java jiglet application on the mobile phone, by using a PHP web application. The 3G mobile phone used should carry multi- tasking function, such as using a Java based application during a voice call. Via the Internet and 3G mobile networks, the Java jiglet application on the mobile phone receives the data and displays the visual information on the screen of the mobile phone to the doctor. Fig.1: the overview of the system B. Patients Monitoring System The Bedside Monitor is called DS-7100 system while the Central Station Monitor is called DS-5700 system. They are both produced by Fukuda Denshi Company and being widely used in Japan. The Bedside Monitor can monitor patient's ECG, RESP, SpO2, EtCO2, IBP, NIBP and TEMP. In the experimentation stage, we use the demonstration function of it to generate the biosignals we need for transmission. The Central Station Monitor stores data from Bedside Monitor. The upper limit of the capacity is up to storing 96 hours continuous waveform data. Medical staffs in hospital can also monitor the state of patients by using the PC which can access the Remote Information Server via the Local Area Networks in the hospital. C. Remote Information Server Remote Information Server is a web server, which is technically call an Apache server running on a SUSE Linux operating system. To create HTML pages dynamically, the PHP Hypertext Preprocessor(Apache Software Foundation, 2004) is used. Remote Information Server is the focus of the system's data transferring process. As is shown in Fig.2, the server can be accessed both from the 3G mobile phone and PC used by medical staffs in hospital. In the mobile phone's case, PHP application on the server get the requests from the mobile phone and access Central Station Monitor to extract the specific data, such as patients' data, event data and alarm data. These data are translated to the form that can be recognized by the Java jiglet application of the mobile phone. Then the application creates a unit of vital data and sends it back to the mobile phone. In PC's case, the medical staffs can access the server by using the web browser of the PC to get the data they need. Fig.2 the working process of the system D. The Java-enabled 3G mobile phone We use FOMA F901iC mobile phone released by NTT DoCoMo, Inc., which is Japan's biggest mobile communication carrier. The specification of it is shown in Table I. TABLE I THE SPECIFICATION OF FOMA F901IC MOBILE PHONE Application size 100KByte Scratchpad Size 400KByte Communication Protocol HTTP Type of Application Client/Server model Display Size QVGA TFT(240*320) Display output rate 63 points/s The networks between the Remote Information Server and the 3G mobile phone is shown in Fig.3. Patients Bedside Monitor Central Station Monitor Remote Information Server Medical Staffs Doctor 3G Mobile Phone 3G mobile phone Remote Information Server ICU/CCU Central Monitor PC Bedside Monitor PHP Application Patient's data Event data Alarm data JAVA jiglet application display 3714 Authorized licensed use limited to: Shenzhen Institute of Advanced Technology. Downloaded on January 29, 2009 at 21:19 from IEEE Xplore. Restrictions apply. Fig.3 the network structure between the remote information server and the 3G mobile phone The i-mode server is the connector of the Internet and the 3G mobile network. It is provided by NTT DoCoMo Inc.. It can support high-speed packet data communication via 3G mobile networks. The Java jiglet application which is based on the Java platform can be used not only on the NTT DoCoMo's mobile phones, but also can be used on other mobile phones released by the other two carriers. It exchanges information with the Remote Information Server through the i-mode server. Another function of the Java jiglet application is to display the data the mobile phone got and create a interface between the doctor and mobile phone. III. APPLICATION DEMONSTRATION The Java jiglet application on the mobile phone can implement three functions. They are realtime waveform and data monitoring, list trend data monitoring and the patient information checking. A. Realtime Waveform and Data Monitoring Realtime waveform and data monitoring can monitor the patients' biosignals simultaneously with about 20 seconds delay after the Bedside Monitor in hospital. The doctor can choose by pressing the mobile phone's keyboard which he would like to check. The original information can been seen in Fig. 4(a). And the menu that the doctor can select from the mobile phone is shown in Fig. 4(b). Take ECG waveform for instance. By selecting the HR item in ECG1's submenu(the right column in the screen), approximately 6 seconds waveform can be seen on the screen of the mobile phone, as is shown in Fig. 4(c). The bottom of the screen also shows the basic information of the waveform, including the Patient ID, the name of the waveform, the time on the Bedside Monitor and the heart rate of the patient is 60, which is shown on the screen. Fig.4(a) the screen of Bedside Monitor Fig. 4(b) The select submenu in realtime waveform and data monitoring function Fig.4(c) the screen of realtime waveform and data monitoring B. List Trend Data Monitoring List trend data monitoring allows the doctor checking the patients' data of the last 4 days, as is shown in Fig.5. The data shown involve parameters of ECG, BP, SpO2 and so on. The data are taken on a trend of the changing of the parameters every 10 minutes. The doctor can select the time point in which he is interested. This function makes development of the patient's state of illness clear to the doctor and help the doctor to analyze and make the right decision. Fig.5 The screen of the list trend data monitoring 3715 Authorized licensed use limited to: Shenzhen Institute of Advanced Technology. Downloaded on January 29, 2009 at 21:19 from IEEE Xplore. Restrictions apply. C. Patient Information Checking The patient information checking function displays the basic information of the patient, as is shown in Fig. 6. Because it is a demonstration, some parameters are the initial value of the Monitor. Fig.6 the screen of the patient information checking IV. PERFORMANCE ANALYSIS AND DISCUSSION After the system was successfully implemented, some test has been done. One of the most important results is about the time used when the application is brought into practise. As mentioned in the last chapter, the delay for realtime waveform monitoring is about 20 seconds. The time used varied with different mobile phones. Take the NTT DoCoMo's FOMA F901iC for instance: 1) As mentioned in the last chapter, the delay for realtime waveform monitoring is about 20 seconds. 2) The total time used for starting the application from the beginning on average is 17.3 seconds. 3) Time for receiving realtime waveform on average is 4.7 seconds. 4) Time for receiving list trend data on average is 8.1 seconds. 5) Time for receiving patient's information on average is 4.6 seconds. The transferring data rate is about 120kbps on average. This means it is fast enough to afford the realtime monitoring. On the other hand, some problems need to be solved. When the mobile phone cannot be connected to the base stations because of being out of service area or other unknown reasons for even a few seconds, the realtime monitoring waveform data become too confusing to be distinguished. Another problem is that the time used for starting the application is long. The third one is that i-mode service cannot be used all over the world. V. CONCLUSIONS AND FUTURE WORKS In this paper, we developed a novel Remote Patient Monitoring System for doctors to monitor the patients' biosignals in ICU/CCU on a Java-enabled 3G mobile phone. By using this system, doctor can monitor patients' realtime data, list trend data and check the patients' information during a voice call from other medical staffs in hospital. It can help doctor with diagnosing the patients' situation more quickly and accurately. In the future, a Remote Patient Monitoring System will be developed by adopting the smartphones, which based on the Symbian OS or Windows Mobile OS. The new system can be used all over the world and save the time for starting the jiglet platform. Additional functions will be added to avoid the data confusion. REFERENCES [1]Suman Kundu, J. Mukherjee, A.K. Majumdar, B. Majumdar, Sirsendu Sekhar Ray, “Algorithms and Heuristics for Efficient Medical Information Display,” Computers in Biology and Medicine(2006), www.intl.elsevierhealth.com /journals/cobm [2]Paul Rubel, Jocelyne Fayn, Giandomenico Nollo, Deodato Assanelli, etc, “Toward personal eHealth in cardiology. Results from the EPI- MEDICS,” Journal of Electrocardiology 38(2005) 100-106, www.elsevier.com/locate/jelectrocard [3]Mehmet Engin, Yalcin Yamaner, Erkan Zeki Engin, “A biotelemetric system for human ECG measurements,” Measurement 38(2005) 148-153, www.elsevier.com/locate/measurement [4]N. Maglaveras, V. Koutkias, I. Chouvarda, D.G. Goulis, A. Avramides, D. Adamidis, G. Louridas, E.A.Balas, “Home care delivery through the mobile telecommunications platform: the Citizen Health System(CHS) perspective,” International Journal of Medical Informatics 68 (2002) 99-111 [5]Yuichi Kogure, Hiroki Matsuoka, Masatake Akutagawa, Yoshihiro Shimada and Yohsuke Kinouchi, “The Applications of Remote Patient Monitoring System using a Java-enabled 3G Mobile Phone,” IFMBE Proceedings WC 2006 “World Congress on Medical Physics and Biomedical Engineering” , Vol. 14, pp.3522-3525, (2006) [6]M. F. A. Rasid, B. Woodward, “Bluetooth Telemedicine Processor for Multichannel Biomedical Signal Transmission via Mobile Cellular Networks,” IEEE Trans Inf Technol Biomed, vol.9, pp.35-45, (2005) 3716 Authorized licensed use limited to: Shenzhen Institute of Advanced Technology. Downloaded on January 29, 2009 at 21:19 from IEEE Xplore. Restrictions apply.