CC BY
22
НАУЧНОЕ ПЕРИОДИЧЕСКОЕ ИЗДАНИЕ «IN SITU» №5/2015 ISSN 2411-7161
МЕДИЦИНСКИЕ НАУКИ
УДК 004.272.44
Korsakov Igor
PhD in Mathematics, National Research University Higher School of Economics
Mikhaylova Anna
National Research University Higher School of Economics
Antonova Ksenia
National Research University Higher School of Economics
USING SINGLE-CHANNEL ECG RECORDER IN REMOTE HEALTHCARE MONITORING
Summary
This study demonstrates that the feasibility of creating a 12-lead ECG with one-lead ECG could be easy. The validity of the recordings would seem to suggest that this technology could become an important useful tool for home use. Distance patient monitoring during daily activity can provide medical information from different sensors, one of them is one-channel ECG, but physician needs 12-leads standard ECG. Producing 12 leads at home using one-channel ECG can be done using 5-leads EASI method and derived method to generate 12 leads ECG.
Keywords
ECG; electrocardiogram; cardiography; Holter monitor; EASI, information technologies in health care, remote
monitoring system.
Cardiograph is one of the most important medical devices for remote monitoring system [1, P. 273]. The medical facilities commonly use twelve-lead electrocardiographs that to the maximum extent allow diagnosing diseases of the cardiovascular system, but unfortunately such electrocardiographs are large in size and complexity and also cost much. Recently, inexpensive single-channel mobile cardiographs that are available for a large number of patients have appeared on market, but from a practical point of view, the data of the electrocardiogram collected from one lead is ineffective for physicians. Using a single-tap electrocardiogram in determining the diagnosis is practically pointless, because the doctor was trained by standard twelve ECG leads and has practical experience (thousands of decrypted electrocardiograms), and, most important, the correct diagnosis is possible only in conjunction with all leads, and sometimes resorting to the records from the Holter monitor. As a solution of the practical application of single-channel mobile ECG in remote health monitoring system is taking a customary ECG in 12-leads sequential via a single-channel device.
In this research as a single-lead cardiograph we use a device with AD620 compressor by Analog Device, 32 bit microcontroller with finalized software for easier sequential recording of ECG in 12 leads (FIG.1). Also we considered MAC 500 by GE Medical Systems as a standard of 12-leads cardiograph.
гххот -СЭ-
Figure 1 - the technical scheme of single-lead cardiograph
_НАУЧНОЕ ПЕРИОДИЧЕСКОЕ ИЗДАНИЕ «IN SITU» №5/2015 ISSN 2411-7161_
Our device has a LCD touchscreen display to monitor and manage ECG recordings [3, P.425]. An electrocardiograph has a differential input which allows connecting all leads one by one, including bipolar and unipolar leads. This involves a special command, which hints how and when you need to wire up an electrode. An
overall algorithm of ECG recording is illustrated on FIG.2.
Отведение + ЭКГ - ЭКГ Примечание
I LA RA
II LL RA
III RL LA
aVR RA LL,LA дополнительный шнур
aVL LA LL,RA дополнительный шнур
aVF LL LA,RA дополнительный шнур
VI VI LL
V2 V2 LL
V3 V3 LL
V4 V4 LL
V5 V5 LL
V6 V6 LL
Figure 2 - a sequence of ECG recordings
The data of measurements is written to cardiograph's internal memory (2GB flash drive) as 12 files: measurement 1.I, measurement 1.II and so forth to measurement 1.V6. Device has 12 bit RAM with sampling rate 400 SPS. The process of measurement lasts approximately 10-15 minutes, and then files are submitted to the central server and
converted into a format that is used to physicians (FIG.3).
Figure 3 - Standard 12-leads ECG in format 3x4+1
Despite the fact that patient doesn't need to do such a procedure often, once per day or in cases of decline, there is still a chance to mistake the electrodes or connect them in a wrong way. This procedure considers more attention from patient, which itself can contribute to stress.
_НАУЧНОЕ ПЕРИОДИЧЕСКОЕ ИЗДАНИЕ «IN SITU» №5/2015 ISSN 2411-7161_
This bring us to reduce an amount of leads for a patient, hence a number of measurements as well. It makes sense if the patient is followed-up with a certain type of cardiovascular disease and the physician needs several certain leads for monitoring purposes. In this case there is a risk of failing to detect a new cardiovascular disease.
Materials and Methodology There is a method of ECG recording - EASI, suggested by G.E. Downer in 1988 - to deal with this problem. It achieves standard 12-leads ECG while using only 5 electrodes [4, P.185]. Such an arrangement of EASI and standard electrodes is illustrated on FIG.4. In EASI electrodes are easily embedded not due to the reduction of general amount of electrodes, but also the location is non-critical. That's why this method is suitable for a patient.
Figure 4 - Location of EASI-electrodes (left) and its standard positioning (right)
G.E. Dower observed three modified vector signals, recorded by bipolar pair of electrodes:
1. A-I (horizontal axis X);
2. E-S (vertical axis Y);
3. A-S (including X, Y and Z).
Each of the 12 leads can be calculated from the formula:
D. = ai * (A-I) + bt *(E-S) + q*(A-S) (1)
Also Dower suggested empirical coefficients a, b and c, which allow to set a all 12 leads. Over the last few years, works in computation of coefficients have reached an advanced stage. Of particular interest in mathematical methods of computation is the paper [6, P.287], which considers these methods:
- Multilayed perceptron;
- Linear regression;
- Least-square method;
- Algorythm of a random tree.
Testing of these models revealed that the best result was gained by linear regression [6, P.288]. Using it, we calculate new empirical coefficients:
aVF = 0,2143 X E + 0,1146 X A - 1,0935 X S + 0,7287 X I - 3,0685
aVL = -0,1298 X E + 0,5988 X S- 1,6804 X I + 2,3043
aVR = -0,0845 X E - 0,1195 X A + 0,4929 X S + 0,9408 X I + 0,7811
I = -0,0302 X E + 0,083 X A + 0,0718 X S- 1,7402 X I + 1,0043
II = 0,1992 X E + 0,1561 X A - 1,0576 X S- 0,1414 X I - 2,5664
III = 0,2295 X E + 0,0731 X A - 1,1295 X S+ 1,5988 X I - 3,5707 V1 = 0,6344 X E + 0,0799 X A + 0,501 X S + 0,4933 X I + 4,0389 V2 = 1,0836 X E- 0,095 X A + 0,5252 X S- 1,249 X I + 13,6635 y3 = 0,7993 X E + 0,2801 X A + 0,0881 X S- 2,3115 X I + 5,0573
НАУЧНОЕ ПЕРИОДИЧЕСКОЕ ИЗДАНИЕ «IN SITU» №5/2015 ISSN 2411-7161
V4 = 0,368 х£ + 1,2349 X А + 0,0869 X S - 1,1872 X I - 2,2414 V5 = 0,1384 X E + 1,5578 X A + 0,0865 X S + 0,3616 X I + 0,024 V6 = 0,0362 X E + 1,2552 X A - 0,1469 X S + 0,706 X I - 1,2352
Figure 5 - A comparison of method 1 [4, P.185], method 2 [8, P.26] and method 3 [6, P.288]
Now we can use a formula and generate a 12-leads-cardiogram by EASY-method and single-lead cardiograph. The results are presented in FIG.6, 12-lead ECG on above and calculated results are below.
Figure 6 - A comparison of ECGs (measured vs calculated)
Conclusion
Remote monitoring system is getting on a daily basis as more new services from different providers appear nowadays. We get new medical devices which are simple to use at home conditions. In order to encourage physicians to trust such devices more researches and guidelines are in need. This paper shows that it's possible to reduce an amount of electrodes and make ECG-recording process at home available to patients themselves. To increase the accuracy and reliability of computation of leads there is a need to produce a study which includes individual features into computed coefficients.
The developers of «Health monitor» system thank the CJSC «AhTh» and National Research University Higher School of Economics for a tight and productive cooperation and the support of the Ministry of Education and Science
_НАУЧНОЕ ПЕРИОДИЧЕСКОЕ ИЗДАНИЕ «IN SITU» №5/2015 ISSN 2411-7161_
of Russian Federation (the contract № 02. G 25.31.0033) without which it would be impossible to complete the research.
References:
1. Kupcov S.M., Korsakov I.N., Atabaeva V.D. Avtomatizirovannaya sistema distancionnogo monitoringa zdorovya cheloveka // Innovacii i progress v kardiologii. Materialy Rossijskogo nacionalnogo kongressa kardiologov 24-26 sentyabrya 2014. P.273-274.
2. AD620 Data Sheet revision F. Analog Devices, 2003. URL: http://www.analog.com/media/en/technical-documentation/data-sheets/AD620.pdf
3. Korsakov I. N., Kuptsov S. M., Raznometov D. A. Personal Medical Wearable Device for Distance Healthcare Monitoring X73-PHD //International Journal of Scientific & Engineering Research. 2014. Vol. 5. No. 1. PP. 422429.
4. G. E. Dower, A. Yakush, S. B. Nazzal, R. E. Jutzy, and C. E. Ruiz Deriving the 12-lead electrocardiogram from four (EASI) electrodes. Journal of Electrocardiology. 1988. Vol. 21. PP. S182-S187.
5. G. E. Dower EASI 12-lead electrocardiography. Totemite Publishers. Vol. 2. No. 4. PP. 303-314.
6. Wojciech Oleksy, Ewaryst Tkacz, and Zbigniew Budzianowski Improving EASI ECG Method Using Various Machine Learning and Regression Techniques to Obtain New EASI ECG Model. International Journal of Computer and Communication Engineering. 2012. Vol. 1. No. 3. PP. 287-289.
7. W. Oleksy and E. Tkacz Investigation of a Transfer Function between Standard 12-Lead ECG and EASI ECG. Analysis of Biomedical Signals and Images. 2010. PP. 322-327
8. D. Q. Feild, C. L. Feldman, and B. M. Horacek Improved EASI Coefficients: Their Derivation, Values, and Performance. Journal of Electrocardiology. 2002. Vol. 35, PP. 23-33.
© Korsakov I., Mikhaylova A., Antonova K.,2015
