Advanced Telemedicine Services through Context-aware Medical Networks

Transcription

1 Advanced Telemedicine Services through Context-aware Medical Networks Charalampos Doukas, Ilias Maglogiannis, and George Kormentzas Abstract The aim of this paper is to present a framework for advanced telemedicine services provision and personalization, through network and patient-state awareness. Under this scope a context-aware medical networking platform is described. The developed platform enables proper medical data coding and transmission according to both a) network availability and/or quality and b) patient status, optimizing thus network performance and telediagnosis. I. INTRODUCTION emote patient care and telemedicine platforms have R been proved during the last years significant tools for the optimization of patient treatment in isolated areas ([1]- [12]). Transport, accommodation and medical personnelrelated costs are reduced, and a full time 24 hours per day, 7 days per week patient status monitoring is provided ([7], [8]). Health monitoring may be delivered not only in a hospital environment but at home as well, through the establishment of modern patient telemonitoring systems. The reasons are better possibilities for managing chronic care, controlling health delivery costs, increasing quality of life and quality of health services and distinct possibility of predicting and thus avoiding serious complications. Methods of treating and monitoring patients remotely include the use of bio-sensors, additional monitoring devices (e.g., video cameras) and patient-physician interaction applications (e.g., video conference, prescription management, etc.). Due to the remote locations of the involved actuators, a network infrastructure (wired and/or wireless) is needed in order to enable the transmission of the medical information. Telemedicine systems cannot however always perform in a successful and efficient manner; Issues, like large data volumes (e.g., video sequences), unnecessary data transmission occurrence and limited network resources can cause inefficient usage of such systems ([13], [6]). In addition, wired and/or wireless network infrastructures often fail to deliver the required quality of service (e.g., bandwidth requirements, minimum delay and jitter requirements) due to network congestion and/or limited network resources. Manuscript received June 30, Charalampos Doukas is with the Dep. of Information and Communication Systems Engineering, University of the Aegean, Samos, Greece ( Doukas@ieee.org). Ilias Maglogiannis is with the Dep. of Information and Communication Systems Engineering, University of the Aegean, Samos, Greece ( imaglo@aegean.gr). George Kormentzas is with the Dep. of Information and Communication Systems Engineering, University of the Aegean, Samos, Greece ( gkorm@aegean.gr). Context-aware medical networks can overcome the aforementioned issues, through performing appropriate content adaptation. This paper presents an improved patient state and network aware telecare and telemonitoring framework. The scope of the framework is to allow medical data transmission only when determined necessary and encode the transmitted data properly according to the network availability and quality, and to the patient status. The framework s architecture is open and does not depend on the monitoring applications used, the underlying networks or any other issues regarding the telemedicine system used. A prototype evaluation platform has been developed in order to validate the efficiency and the performance of the proposed framework. The rest of the paper is organized as follows; Section 2 provides background information on telemedicine and telecare systems adopting the context aware and personalization concept, among with a brief literature overview of proposed schemes and implementations. Section 3 describes the architecture of the proposed framework, provides implementation details and presents its main functionality. Performance aspects in the context of data coding and transmission are presented in Section 4. Finally, Section 5 concludes the article and discusses future work. II. RELATED WORK AND BACKGROUND INFORMATION Despite the numerous implementations and proposals of telemedicine and e-health platforms found in the literature (an indicative reference collection can be found in [1]-[12]), only a few works include context awareness. The main goal of context aware computing is to acquire and utilize information about the context of a device to provide services that are appropriate to particular people, place, time, events, etc. ([19]). According to the latter, the work presented in [17] describes a context-aware mobile system for interhospital communication taking into account patient s and physician s physical location for instant and efficient messaging regarding medical events. J. Bardram presents in [18] additional use cases of context-awareness within treatment centers and provides design principles of such systems. The project AWARENESS (presented in [21]) provides a more general framework for enhanced telemedicine and telediagnosis services depending on the patient status and location. To our best of knowledge, there is no other work exploiting context awareness for optimizing network utilization and efficiency, within the context of medical

2 networks and telemedicine services. The presented telemedicine context aware framework is based on the active services concept presented in [20]. Active services can dynamically adapt themselves to various underlying networking environment conditions, according to the instruction of appropriate networking entities (i.e. a Network Broker) and the requirements posed by the end user. The Network Broker is a special software agent, which monitors network resources and activities. Its intelligence enables it to take decisions regarding services and network usage leading this way to optimum network utilization. Given a specific conventional service, the corresponding active service constitutes the outcome of the application to the conventional service. Taking into account that the particular format of the services functions depends on the underlying networking environment, an active service can have various instances according to the particularities of the defined functions. In our case, the active services consist of the patient monitoring tools that can dynamically adapt the coding of the generated data (in terms of rate, compression and/or encryption used) to both underlying network conditions and the patient status itself. A more detailed description of the discussed context-aware medical framework is provided in the following section. III. THE CONTEXT-AWARE MEDICAL NETWORKING FRAMEWORK The architecture of the proposed context-aware medical networking framework is illustrated in Figure 1. temperature sensors) and corresponding vital signals. Defined threshold values in the latter signals determine the case of an immediate data transmission (alarm event) to the monitoring unit. Depending on network availability and quality, periodical transmission of the patient data is performed for evaluation by physicians. According to the network interface used, appropriate coding (e.g. video compression, data encryption, etc.) is applied on the transmitted medical data, avoiding thus possible transmission delays and optimizing the whole telediagnosis procedure. The framework s architecture is open and does not depend on the monitoring applications used, the underlying networks or any other issues regarding the telemedicine system used. For this purpose, Web Services ([22], [23]) have been used as a communication mechanism between the major framework components and the external patient monitoring applications used. The message exchange has been implemented through SOAP [16], a simple yet very effective and flexible XML-based communication mechanism. The latter involves the session initialization (which more precisely includes user authentication and service discovery) and the exchange of status and control messages. The status messages include information regarding the patient data as generated from the monitoring sensors and the underlying network status and quality, whereas the control messages contain instructions regarding the proper coding of the transmitted data. It should be noted that the involved modules for the aforementioned communication (see Figure 1) can all reside at the patient s site, or alternatively the Medical Broker can reside at the remote treatment site for the direct collection of medical data and the reactive instruction s provision. Figure 1. The context-aware medical networking framework architecture The major modules consisting the proposed framework are; a) the network monitoring module that determines the current network interface used and the corresponding status, b) the patient status monitoring module that collects patient data and determines the patient status, c) the data coding module which is responsible for properly coding (compressing, encrypting, etc) the transmitted patient data, according to instructions given by d) the medical broker (i.e. usually a repository containing predefined or dynamically defined threshold values for determining patient and network status). The patient state can be determined through a number of health sensors (i.e. heart rate and body Figure 2. Message Exchange between the framework s modules Figure 2 illustrates the aforementioned message exchange between the framework s components. IV. A FRAMEWORK EVALUATION PLATFORM This section provides the description of a prototype telemedicine platform specially developed for the evaluation of the presented framework. The main components that the platform consists of are the attached biosensors on the patient, a PDA that collects corresponding signals, a

3 software module installed on a laptop or desktop computer that determines the data coding depending on the patient and network status, a network infrastructure for data transmission and the monitoring units (e.g., a treatment center, an ambulance or a physician at a remote site). TABLE I PATIENT DATA AND DATA LEVELS INDICATING AN URGENT STATUS ACQUIRED PATIENT DATA ECG (Electrocardiogram, 3 leads) DATA LEVELS INDICATING AN URGENT STATE ST Wave elevation & depression T-wave inversion BP (Non invasive Blood Pressure) PR (Pulse Rate) 90 mm Hg > systolic > 170 mm Hg 50/min > PR > 110/min HR (Heart Rate) 50/min > HR > 110/min SpO2 (Hemoglobin oxygen saturation) < 90 (%) Figure 3. The evaluation Platform Architecture Figure 3 illustrates the connection scheme of the aforementioned components. Biosensors that can monitor patient state through vital signs are attached on the patient and transmit data on a PDA device carried by the patient (A). The latter data are transmitted to a computer for evaluation through wireless access (e.g., Bluetooth). In addition, a camera monitors the patient site enhancing the overall patient status monitoring. The developed software monitors both patient and network status (B), determining the proper coding of the collected patient data. The data are transmitted through wireless or wired access provided by network operators (C) to the corresponding monitoring units (D). Two patient states have been defined; normal and urgent. The patient data that are monitored through corresponding sensors are: ECG, BP (Non invasive Blood Pressure), PR (Pulse Rate), HR (Heart Rate) and SpO2 (Hemoglobin oxygen saturation). TABLE I presents the data levelsthresholds that indicate the patient status. Under normal patient status no real time data transmission is acquired, unless defined otherwise by the physicians. Daily status reports containing information regarding vital signs average values can be reported at predefined time periods. In case of an emergency (e.g., heart rate falling below a predefined threshold), the real time data transmission mode is, alerting the monitoring physicians regarding the patient status and allowing thus efficient remote care (e.g., medication advise, urgent patient transfer to hospital, etc.). In addition, according to three network condition levels, determined both by the serving network type and the available transmission bandwidth, proper data coding and encryption schemes are applied. In case of good network conditions (e.g., WLAN with good radio coverage), SSL [14] encryption is used, whereas video image is transmitted uncompressed, if the latter is required. When an average network connection is detected (e.g. multi-user WLAN environment), video images are compressed using the H.263 [15] standard. Compression can be scalable; low compression is applied on data in case of average network quality whereas high compression can optimize transmission in case of a slow network connection (e.g. GPRS). TABLE II PATIENT DATA CODING ACCORDING TO NETWORK QUALITY AND PATIENT STATUS Good Network Quality 112 kbps) Average Network Quality 50 kbps Low Network Quality 24 kbps) Normal Patient Status PATIENT STATUS SSL encryption used, Good video quality (no compression), SSL encryption used, (Low H.263 compression), Low video quality (High H.263 compression), Urgent Patient Status* Good video quality (no compression), (Low H.263 compression, (High H.263 compression, * Patient Urgent Status indicated by data levels-thresholds provided in TABLE I

4 TABLE II summarizes the data coding schemes used, according to patient and network status. Appropriate graphical user interface (see Figure 4) provides visualization of the collected sensor data among with additional functionality, like event logging, network setup (i.e. sensor data threshold definition, data coding according to network condition), and direct contact with the monitoring units. All software modules have been developed in Java, supporting usage under any operating system. The basic data exchange mechanism, like daily reports, communication with medical personnel, and real time data transmission activation is based on SOAP. During real time data transmission, both video data and bio-signals are transmitted through the RTSP protocol. V. PERFORMANCE ASPECTS The purpose of the presented platform is to optimize patient telecare and telediagnosis through proper data coding and transmission. On one hand, unnecessary high rate demanding data (i.e. video) is avoided; Camera transmission is not, unless requested or patient status indicates so. Additionally, scalable data compression achieves better network utilization in case of poor network conditions (i.e. low transmission rates, low connections speed, etc.). TABLE IIIpresents the transmission rate requirements for each data type (i.e., video sequences ) that is transmitted, according to the H.263 compression factor used. The decrement of the demanded bandwidth in the case of maximum compression applied (i.e. compression factor 0.2) indicates the video sequences can be transmitted even through slow and unreliable wireless network infrastructures (e.g., GPRS or UMTS). Thus, the deployment of the proposed patient and network-state aware telemedicine platform provides better utilization of network resources, especially in urgent events that require immediate patient status monitoring. TABLE III DATA TRANSMISSION RATE REQUIREMENTS ACCORDING TO COMPRESSIION FACTOR USED. TRANSMISSION RATE REQUIREMENTS Compression Factor Video Transmission Rate kbps kbps kbps Factor 1.0 applies to no compression, whereas factor 0.2 applies to maximum compression Figure 4. Application screenshots of the Patient Status Monitoring module; appropriate GUI provides the sensor data, the network status and additional functionality like network setup, event logging and medical personnel contact. VI. CONCLUSION Patient status monitoring is one of the most representative features of telemedicine and telediagnosis systems. The proper coding and transmission of the data can optimize the efficiency, effectiveness and utilization of such systems. An innovative telecare and telediagnosis platform has been developed, that considers both network and patient status for proper coding of patient data; unnecessary data transmission is thus avoided, unless an urgent case is detected or it is required by the medical personnel. Proper data encryption provides the essential security, and compression is applied optimizing transmission, when network conditions are considered inadequate. Future work could include the enhancement of the platform with advanced bio-signal and video compression techniques (e.g., H.264 or MPEG4) or with advanced video processing for automated patient status determination. In addition, the establishment of the innovative context-aware telecare and telediagnosis platform into a real patient-care environment would provide helpful information regarding the assessment of the platform in use.

5 ACKNOWLEDGMENT This Research work is funded by the Community Initiative INTERREG III/A Greece Cyprus under the Grant "Telemedicine Network for Remote Locations in the North Aegean Region". REFERENCES [1] Lin J. C., Applying telecommunication technology to health care delivery, IEEE Engineering in Medicine and Biology Magazine, vol. 4, pp , 1999 [2] Pavlopoulos, S., Kyriacou, E., Berler, A., Dembeyiotis, S., Koutsouris, D, A novel emergency telemedicine system based on wireless communication technology-ambulance, IEEE Transactions on Information Technology in Biomedicine, vol. 4, pp , 1998 [3] Deb, S., Ghoshal, S., Malepati, V.N., Kleinman, D.L., Telediagnosis: remote monitoring of large-scale systems, in Proc. of IEEE Aerospace Conference, 2000, pp: [4] Choi, Y.B.; Krause, J.S.,H. Seo, C., K.E., K. Chung,, Telemedicine in the USA: standardization through information management and technical applications, IEEE Communications Magazine, vol. 44, pp , April 2006 [5] Anagnostaki, A.P.;, Pavlopoulos, S., Kyriakou, E., Koutsouris, D., A novel codification scheme based on the "VITAL" and "DICOM" standards for telemedicine applications, IEEE Transactions on Biomedical Engineering, vol. 49, pp , Dec [6] LeRouge, C., Garfield, M.J., Hevner, A.R., Quality attributes in telemedicine video conferencing, In Proc. of 35 th Annual Hawaii International Conference on System Sciences, 2002, pp [7] Hailey D.,Roine R., Ohinmaa A., Systematic Review of evidence for the benefits of telemedicine, Journal of Telemedicine and Telecare, vol. 8, 2002, pp [8] Traver, V., Monton, E., Bayo, J.L., Garcia, J.M., Hernandez, J., Guillen, S, Multiagent home telecare platform for patients with cardiac diseases, in Proc. of Computers in Cardiology, 2003, pp [9] Kollmann, A., Hayn, D., Garcia, J., Rotman, B., Kastner, P., Schreier, G., Telemedicine framework for manufacturer independent remote pacemaker follow-up, In Proc. of Computers in Cardiology, 2005, pp [10] Pattichis, C.S., Kyriacou, E., Voskarides, S., Pattichis, M.S., Istepanian, R., Schizas, C.N., Wireless telemedicine systems: an overview, IEEE Antennas and Propagation Magazine, vol. 44, pp , April 2002 [11] I. Maglogiannis : Design and Implementation of a Calibrated Store and Forward Imaging System for Teledermatology Journal of Medical Systems, Vol. 28, (5), pp Springer Science Business Publishers, 2004 [12] C. Doukas, V. Moulos, I. Maglogiannis, G. Kormentzas, Performance of a Java-based Mobile 3-tier PACS Application for DICOM compliant Medical Images over Heterogeneous Radio Networks, In Proc. of INC2005 Fifth International Network Conference, pp Samos Island, Greece July 2005 [13] Lage, A.-L., Martins, J., Oliveira, J., Cunha, W., A quality of service approach for managing tele-medicine multimedia applications requirements, in Proc. of IEEE Workshop on IP Operations and Management, 2004, pp [14] SSL 3.0 Specification, [15] The ITU H.263 compression standard, available in [16] SOAP specifications, [17] Miguel A. Muoz, Marcela Rodrguez, Jesus Favela, Ana I. Martinez- Garcia, Victor M. Gonzlez, Context aware mobile communication in hospitals IEEE Computer Magazine, vol. 36, pp , Sept [18] J. Bardram, Applications of context-aware computing in hospital work: examples and design principles, In Proc. of the ACM symposium on Applied Computing, 2004, pp [19] T. Moran, P. Dourish, Introduction on Context-Aware Computing, Special Issue on Human-Computer Interaction, vol. 16, 200. [20] C. Doukas, G. Kormentzas, Active Services Substantiation through SOAP, to be presented at the 6 th Networking Conference (INC2006), July 2006, UK. [21] Broens, T., van Halteren, A., van Sinderen, M. & Wac, K., Towards an application framework for context-aware m-health applications, In Proc. of the 11th Open European Summer School (EUNICE 2005), July 6-8, 2005, Madrid, Spain [22] Hori, M., Ohashi, M., Applying XML Web Services into Health Care Management, In Proc. of the 38th Annual IEEE Hawaii International Conference on System Sciences (HICSS 05), pp.155, Jan [23] Yugyung Lee, Patel, C., Soon Ae Chun, Geller, J., Towards intelligent Web services for automating medical service composition, In Proc. of the IEEE International Conference on Web Services, pp , July 2004