Proceedings of the 26th Annual International Conference of the IEEE EMBS San Francisco, CA, USA September 1-5,

, 2004

1 of 4

A Wireless ECG System for Continuous Event Recording and Communication to a Clinical Alarm Station
Agder University College, Faculty of Technology, Grimstad, Norway Hospital of Buskerud, Department of acute medicine, Drammen, Norway 3 Aalborg University, Department of Health Science and Technology, Aalborg, Denmark
2 1

R. Fensli1, E. Gunnarson2, O. Hejlesen3

Abstract Development of new wearable biomedical sensors within a wireless infrastructure opens up possibilities for new telemedical applications leading to significant improvements in continuous monitoring, and thereby to better quality of patient care. In this paper we describe a new concept for a wireless electrocardiogram (ECG) system intended for continuous monitoring of ECG activity especially designed for arrhythmia diagnostic purposes. The patient is wearing an ECG sensor, a smart electronic electrode, with wireless transmission of ECG signals to a dedicated Hand Held Device (HHD). This device is monitoring the continuously recorded ECG signal, and can detect abnormal ECG activity using an automatic arrhythmia detector. Based on this, the device will transmit alarm conditions to a remote Clinical Alarm Station (CAS). The system will act as a continuous event recorder, which can be used to follow up patients who have survived cardiac arrest, ventricular tachycardia or cardiac syncope but also for diagnostic purposes for patients with diffuse arrhythmia symptoms. This paper describes the principle design requirements for the new wireless ECG sensor and system design for the HHD in order to transfer detected alarms to the CAS. Keywords wireless ECG monitoring, tele-medicine, ambulatory ECG, remote consultation, safety alarm system.

I. INTRODUCTION Patients who have survived cardiac arrest, ventricular tachycardia or cardiac syncope, have an increased risk of sudden cardiac death. Many of those patients are normally living at home without any kind of arrhythmia monitoring system or cardiac alarm solutions. By using a wireless and wearable monitoring system for detection of arrhythmia situations, it is possible to provide quick alarms to a central safety alarm system and thereby take necessary action for an emergency rescue. Advanced monitoring solutions using telecommunicating technologies are used for remote ECG diagnosis, and The American College of Cardiology and The American Heart Association have published guidelines for ambulatory electrocardiography [1]. Such monitoring systems can be divided in two modes of operation: Real-time mode and store-and-forward-mode. Today there are several solutions available for portable Holter monitoring [2] and cardiac event recording. The systems are either based on use of standard ECG electrodes and a wired connection to a recording unit, or by pressing a recording device directly to the patients chest when a symptom arises [3]. The use of telecommunications for remote diagnosis is growing rapidly,

and there are several products and projects [4] within mobile ECG recording using Internet solutions [5], Bluetooth technology [6], cellular phones [7], WAP-based implementations [8] and wireless local area networks, WLAN [9]. The EPI-Medics project [10] will implement a solution based on the interpretation of ECG-derived cardiac syndromes and will develop a friendly and easy-to-use intelligent Personal ECG-monitor capable of synthesizing standard 12-lead ECG [11]. There is however, as yet no published information about the recording sensor and a standard 12-lead ECG recording system will normally require wired connection to a wearable device. Schulze and Ehrhardt [12] studied the use of an electronic emergency electrode in order to improve operation and safety in telemetric signal transmission of ECGrecordings by eliminating electric cables and their resulting interference factors. This device is useful in case of emergency for an ECG-recording under mechanical stress. They used a rather large array of electrodes, which can be quite inconvenient if the patient has to wear the system for several days of continuous recordings. A remote diagnostic system integrating digital telemetry has been developed, using a wireless patient module, a homecare station and a remote clinical station [13]. This system is based upon three standard ECG-electrodes connected with wires to a wearable module. The homecare station is based on ISDN transmission. Even though the patient has some degree of freedom to move around, he has to stay within a radius of 10-20 meters from the homecare station. A new ECG Holter recorder from Schiller [14] uses an automatic data transmission via integrated mobile GSM module, however also this system is based upon wired connection to the patient. Several ongoing international projects, where wireless sensors are used within the framework of a standardized Body Area Network, are focusing on improving the patients ability to freely move around in a daily situation while being monitored by a wearable system. Sachpazidis et al. [15] are trying to develop a robust platform for real-time monitoring of patients staying in their home transmitting data to doctors working at the hospital. This @HOME concept aims at measuring several vital parameters, but, so far, the applied sensor technologies have not been published. Several authors describe solutions based on sensors using a wireless Bluetooth communication protocol and a standard Personal Digital Assistant (PDA) [16], [17]. Jovanov et al. [18] propose the use of a Personal Area Network with wireless intel-

0-7803-8439-3/04/$20.002004 IEEE

2208

2 of 4

ligent sensors to perform data acquisition. The use of patient operated ambulatory ECG recording equipment will normally be dependent on the patients condition, as it requires a conscious patient to operate the equipment for making the ECG recordings. This method is also time consuming as the recording time and time for transmitting the recorded signal takes several minutes. Patients undergoing acute myocardial infarction and subsequent arrhythmia leading to ventricular fibrillation and sudden cardiac arrest will not be able to operate such equipments. For a reliable monitoring system it is necessary to develop a completely automatic recording and analyzing unit, which can detect cardiac abnormalities and automatically send alarm conditions to a central safety alarm system. It has been clearly documented that time is a critical factor in order to perform early cardiopulmonary resuscitation (CPR) and early defibrillation, and if the time to start CPR after a cardiac arrest increases, the chance of survival is reduced [19]. If it is possible to develop a simple but wireless ECG sensor which is easy to use, and let the patient wear a device containing computer intelligence, it is assumed that this system can detect critical cardiac conditions and give early alarm signals even if the patient is unconscious or unaware of cardiac arrhythmias. One important aspect will be to eliminate a wired connection between the sensor electrodes and this device. With the use of such a system, it is possible to make easier and more cost efficient ambulatory ECG recordings compared to existing solutions on the market, and the patient can be continuously monitored 24 hours a day. This solution can be useful as an advanced safety alarm system for patients suffering from severe cardiac diseases (primary prophylaxis), and can probably also be used as a secondary prophylaxis for patients who have survived cardiac arrest, ventricular tachycardia or cardiac syncope. It is also intended for diagnostic purposes for patients with diffuse arrhythmia symptoms as an alternative to Holter-monitoring. This paper describes the system design to a new solution for continuous event recording and wireless monitoring of ECG. II. METHODOLOGY In order to perform continuous ambulatory ECG recordings, a new wireless ECG electrode sensor has been developed. This sensor measures the ECG-signal and will continuously transmit the signal wireless to a developed receiver integrated in the HHD. The HHD will receive and analyse the ECG-signal, and if an abnormal cardiac signal is encountered, the device will automatically start transmitting the recorded ECG-signal to the CAS using a standard telecommunication facility. With this solution, only one lead is required for recording of the ECG-signal. This is accomplished by using a

compact double-electrode with no wires connected, as this electrode is equipped with a wireless transmitter and battery supply for several days of continuous usage. This new concept of a wearable cardiac alarm system has recently been developed in our laboratory. It was developed as a surveillance system for long-time ECG-monitoring, and ambulatory ECG, designed as an alternative to conventional Holter monitoring and cardiac event recorders. The system uses a combined real-time mode and store-andforward-mode, and can be described as a continuous event recorder. Fig.1 shows the principal connection between the ECG sensor, the HHD that consist of an RF-receiver connected to a standard PDA and using General Packet Radio Services (GPRS) to the remote CAS. The ECG-signal is measured through a sensor placed on the patients chest which continuously transmits the signal using an incorporated RFradio transmitter. The HHD receives and analyzes the ECG-signal and also stores a predefined length of the recordings. If an abnormal cardiac signal is encountered, the HHD unit will start transmitting the recorded ECG-signal to CAS. The CAS in a hospital or a private alarm service centre will give an alarm to the operating personnel and will display the actual recorded ECG-signals from the patient. In order to reduce problems of real-time ECG signal transmission, we use the data transmission and error correction protocol based on the TCP/IP protocol.

ECG Sensor RF-radio PDA GPRS RF-Radio Receiver Clinical Alarm Station (CAS)

Fig. 1 shows the principal components of the continuous event recording system, where the ECG sensor is wirelessly transmitting the recordings to a handheld device which detect cardiac arrhythmias and transmit the recordings to a remote alarm station, CAS.

Fig. 2 A picture of the ECG sensor prototype (to the left) and the receiver connected to the PDA (to the right) using RS232 cable interface.

2209

3 of 4

Fig. 2 shows a photograph of the ECG sensor prototype, and the handheld device with the two parts, the RF-receiver and the PDA, connected by an RS232 cable. The ECG sensor includes two electrical contact points with conducting gel applied to the patients skin, for obtaining ECG-signals from the patient. These contact points are electrically connected to electronic circuits for amplifying a differential signal by a gain of 250 with an input range of +5 mV, and with a 0,5 Hz High Pass Filter and a 150 Hz Low Pass Filter. The sensor contains a battery supply of 2x3 V Li/MnO2 and a modulated RF-transmitter, Nordic VLSI nRF401, with a carrier frequency of 434.33 MHz. A corresponding receiver using Nordic VLSI nRf401, is continuously receiving the ECG signal and demodulate the signal to an analog signal. This signal is connected to a microprocessor, ATmega8L-8AI, containing an analog to digital converter with 8 bit resolution and a sampling rate of 500 Hz. This receiver uses an interface with RS232 for connection between the receiver and an iPAQ h5550 PDA, where an application is programmed in Labview 7 for Pocket PC 2002. The PDA can be supplied with a GPRS interface in order to transmit the recorded ECG signal to the CAS. The ECG sensor prototype is a compact unit with a size of 70 x 35 x 18 mm, and a weight of 22 g which at a later stage, may be reduced even more. It is supplied from two small batteries of 230 mAh, which currently can perform ECG-recording and transmission for more than five hours of continuous operation. The signal is stored at the PDA in a binary byte stream format together with a timestamp and information identifying the actual recordings. The ECG data is processed in realtime using the R-peak detection algorithm, which is one of the most fundamental ECG analysis methods. This is based on the conclusions of Friesen et Al [20] who made a comparison of the noise sensitivity of nine QRS detection algorithms and found that the use of digital filters would be the obvious choice for many applications since this can be incorporated into a power line notch filter. This is integrated in the Labview application for detection of heart rate, and individual lower and upper limits for detection of bradycardia and tachycardia respectively, can be adjusted. III. RESULTS Fig. 3 shows a picture of the displayed ECG-signal recorded using a test subject. It can be seen that even though it is recorded in a very different manner, it has striking resemblance to a standard Lead I ECG (Eindhoven Lead I), which should make it easier to incorporate into most clinical practices. Preliminary informal assessments indicate that the functionality from the cardiologists point of view is very useful. A study of the functionality and benefits from the patients point of view is on the way.

mV 2,0 1,5 1,0 0,5 0,0 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 sec

Fig.3 A recorded ECG-signal at the HHD using a test subject.

IV. DISCUSSION Normally ECG recordings for monitoring purposes will use three electrodes, where two electrodes separated by approximately 0.3 meter (1 foot) are used for the actual differential recording, Lead I (RA to LA), and the third electrode is used only for noise suppression (ground lead). Thakor and Webster [21] found that a two-electrode design could be useful for biotelemetry, portable Holter monitors, and portable arrhythmia monitors. Our solution is in some ways similar to this ground-free model, as the ECG sensor is in no way connected to an electrical ground. Our concept is, however, based on two electrodes or contact points located close to each other (at a distance of approximately 3 cm) and built into a common unit with electronic circuits and battery supply, and surrounded by adhesive tape for applying this double electrode directly to the patients chest. The optimal position of the sensor is in the 4th or 5th intercostals space at the apex of the heart, but it is possible to obtain adequate ECG-signals with the electrode placed anywhere on the patients chest. Because of the short distance both between the electrodes and the direct input to the amplifier (without any wired connection), only limited 50Hz noise is registered. The ECG-signal obtained differs in some ways from a standard Lead I recording as it only uses two electrodes that are placed close to each other. It is therefore, necessary to further investigate the use of this recording principle for diagnostic purposes for example, with regard to diagnosis of cardiac infarction. On the other hand, the recordings in our system are supposed to be comparable to the recordings from an implantable loop recorder used by Krahn et al [22] who used the Medtronic ILR, and found that this technology is a powerful tool in arrhythmia diagnosis. Even though proper clinical trials are clearly needed to verify our hypothesis, it therefore seems reasonable to assume that our new ECG-monitoring system will be able to, reliably, detect critical cardiac arrhythmias. The prototype sensor unit seems too large for routine use by outpatients, and therefore, miniaturization of the unit is an important future task. Otherwise we have so far, found no critical problems using this technique.

2210

4 of 4

V. CONCLUSION Our system acts as a continuous event recorder, which can be used to follow up patients who have survived cardiac arrest, ventricular tachycardia or cardiac syncope in both ambulatory settings and in hospitals. In conclusion, our system is designed for ease of use and it is designed so that the patient will be able to wear the smart electronic electrode for several days before it is necessary to replace it with a new one. This will be a big advantage compared to the commercial products, which are available today. Hopefully, our system will, after further refinement, be able to contribute to a better quality of life for patients by offering easy-touse continuous ECG-monitoring outside the hospital.

[13]

[14] [15]

[16] [17]

[18]

ACKNOWLEDGMENT
[19]

The study is supported from Srlandets Teknologienter, Grimstad, Norway as a FORNY project.
[20]

REFERENCES
[1] [2] [3] [4] [5] M. H. Crawford, "ACC/AHA Guidelines for ambulatory electrocardiography.," Journal of the American College of Cardiology, vol. 34, pp. 912-48, 1999. K. Y. Kong, C. Y. Ng, and K. Ong, "Web-Based Monitoring of Real-Time ECG Data," Computers in Cardiology, vol. 27, pp. 189-192, 2000. Vitaphone, "Hertz Handy." http://www.vitaphone.de, 2000. C. Finch, "Mobile Computing in Healthcare," Health Management Technology, vol. 20, pp. 63-4, 1999. A. I. Hernndez, F. Mora, G. Villegas, P. G., and G. Carrault, "Real-Time ECG Transmission Via Internet for Nonclinical Applications.," IEEE Trans. Inf. Technol. Biomed, vol. 5, pp. 2537, 2001. J. Andreasson, M. Ekstrm, A. Fard, J. G. Castano, and T. Johnson, "Remote System for Patient Monitoring Using Bluetooth," presented at Sensors, 2002, 2002. R. H. Istepanian, B. Woodward, E. Gorilas, and P. A. Balos, "Desing of mobile telemedicine systems using GSM and IS-54 cellular telephone standards.," J Telemed Telecare, vol. 4, pp. 80-2, 1998. K. H. Hung and Y.-T. Zhang, "Implementation of a WAP-Based Telemedicine System for Patient Monitoring," IEEE Trans. Inf. Technol. Biomed., vol. 7, pp. 101-107, 2003. D. L. Rollins, C. R. Killingsworth, G. P. Walcott, R. K. Justice, and R. E. Ideker, "A Telemetry System for the Study of Spontaneous Cardiac Arrhytmias," IEEE Trans on Biomed Eng, vol. 47, pp. 887-92, 2000. EPI-MEDICS, "IST-2000-26164," 2000. P. Rubel, F. Gouaux, J. Fayn, D. Assanelli, A. Cuce, L. Edenbrandt, and C. Malossi, "Towards Intelligent and Mobile Systems for Early Detection and Interpretation of Cardiological Syndroms.," Computers in Cardiology, vol. 28, pp. 193-196, 2001. H.-J. Schulze and B. Ehrhardt, "Die telemetrische Notfallselekreokardiographie - praktische Anwendung und methodische Effektivitt.," Z Gesamte Inn Med, vol. 41, pp. 685-9, 1986. [21] [22]

Y. H. Nam, Z. Halm, Y. J. Chee, and K. S. Park, "Development of remote diagnosis system integrating digital telemetry for medicine," presented at Proceedings of the 20th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1998. SCHILLER, "ECG Holter Recorder MT-120." http://www.schiller.ch/products.html, 2003. I. Sachpazidis, A. Stassinakis, D. Memos, S. Fragou, S. Nachamoulis, A. Vamvatsikos, A. Stavropoulou, M. Fonseca, R. Magalhes, B. Valente, A. D'Aquila, M. Fruscione, J. Ferreira, and C. Aguiar, "@HOME ein neues Eu-projekt zum Tele Home Care," Biomed Tech (Berl), vol. 47, pp. 970-2, 2002. C. Moor, M. Schwaibold, H. Roth, J. Schchlin, and A. Bolz, "Entwicklung Drahtloser Sensoren auf basis von Bluetooth.," Biomed Tech (Berl), vol. 47, pp. 325-7, 2002. J. Becker, D. Gebauer, L. Maier-Hein, M. Schwaibold, J. Schchlin, and A. Bolz, "The Wirelessmonitoring of Vital Parameters: A design study.," Biomed Tech (Berl), vol. 47, pp. 851-3, 2002. E. Jovanov, D. Raskovic, J. Price, J. Chapman, A. Moore, and A. Krishnamurthy, "Patient Monitoring Using Personal Area Networks of Wireless Intelligent Sensors," Biomed Sci Instrum, vol. 37, pp. 373-8, 2001. T. J. Bunch, R. D. White, B. J. Gersh, R. A. Meverden, D. O. Hodge, K. V. Ballman, S. C. Hammill, W. K. Shen, and D. L. Packer, "Long-term outcomes of out-of-hospital cardiac arrest after successful early defibrillation," New England journal of medicine, vol. 348, pp. 2626-33, 2003. G. M. Friesen, T. C. Jannett, M. A. Jadallah, S. L. Yates, S. R. Quint, and H. T. Nagle, "A Comparison of the Noise Sensitivity of Nine QRS Detection Algorithms," IEEE Trans on Biomed Eng, vol. 37, pp. 85-98, 1990. N. V. Thakor and J. G. Webster, "Ground-Free ECG Recording with Two Electrodes," IEEE Trans on Biomed Eng, vol. BME27, No 12, pp. 699-704, 1980. A. D. Krahn, G. J. Klein, R. Yee, and C. Norris, "Final Results From a Pilot Study With an Implantable Loop Recorder to Determine the Etiology of Syncope in Patients With Negative Noninvasive and Invasive Testing," The American Journal of Cardiology, vol. 82, pp. 117-119, 1998.

[6] [7]

Addresses for correspondence Ass.prof. Rune Fensli, Agder University College, Grimstad, Norway. http://ikt.hia.no/runef/ e-mail: rune.fensli@hia.no

[8] [9]

[10] [11]

[12]

2211