RU2546103C9 - Method of determining parameters of heart rhythm variability - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention relates to medical equipment. Measurement of successively adjacent to each other R-R intervals is performed. Data array is formed. Average value of measured R-R intervals and frequency of the pulse are calculated as value, reverse to average value of measured intervals, as well as value of standard deviation of said R-R intervals from the average value. Maximal and minimal values of measured R-R intervals and difference between them are calculated. Said difference is divided by number of channels Nc, with determination in this way of width of time "window" of each channel of formed in this way Nc-channelled time discriminator. Number of elements of obtained array of measurements, present in each channel of discriminator, is counted. Function of distribution of number of array elements, present in each channel is formed depending on channel number and on respective value of R-R-interval. Diagram of obtained function of distribution of measured R-R intervals is presented on display-screen together with other parameters of heart rhythm and its variability. "Effective width" of distribution function, value of which is also presented on display, is found by division of number of array elements by maximal ordinate of distribution function.

EFFECT: method makes it possible to increase reliability of determining parameters of heart rhythm variability, which is reached due to obtaining function of distribution of measured R-R intervals, envelope of which contains data about distribution of frequency components in electrocardiosignal spectrum.

Description

Technical field

The invention relates to medical equipment and is intended to determine the parameters of human heart rate variability during measurements of heart rate (HR) in various conditions and for pulse diagnosis.

State of the art

The heart rate was the most accessible physiological parameter for registration, which reflects the processes in the cardiovascular system (CVS) of a person. The method of analyzing heart rate variability (variational pulsometry), which began to be developed in the 60s of the previous century in connection with the needs of aviation and astronautics, showed very significant potential opportunities for recognizing pre-painful conditions of the human body, determining the effectiveness of medical procedures, etc. It found present widespread in cardiology medicine [1, 2, 3 ...].

To measure heart rate in pulsometry, it is first necessary to obtain electrocardiosignals (EX) either using metal plate electrodes [1, 2, 3, 9, 10], as in classical electrocardiography (ECG), which is not very convenient for the subjects, or using any or other sensors (more compact and convenient, without electrical contacts) [6, etc.]. So, optoelectronic sensors are widely used, in which the LED transmits the finger (hands), and the photodiode (or phototransistor) receives the optical signal transmitted through the finger (go earlobe), modulated by the blood flow, which has a pulsed nature during heart work [7 et al. .]. Other non-contact sensors are known: compact fiber optic [4], piezoelectric [5], etc.

The electrocardiosignal (ECS) received with the help of sensors has the character of voltage pulses, only slightly exceeding the level of noise and interference, therefore the ECS needs preliminary amplification, filtering, isolation and formation of clear pulses corresponding to pointed R-teeth on the ECG. Using pulses obtained from the ECS, heart rate is then measured in various devices (mainly digital) heart rate monitors, a large number of which are presented not only in the specialized literature, but also in the popular technical periodicals.

The current state of measuring and computing technology allows you to create multifunctional measuring systems that can operate in real time [2, 4, 6 ...]. A variety of heart rate monitors are currently on the market, built into electronic watches of various companies (American POLAR, Finnish SUUNTO, German BEURER, etc. including various blood pressure monitors with a heart rate measurement function) with an indication on a matrix liquid crystal display (LCD). There is also already information about using i-Phone and i-Pad mobile devices for receiving EX. However, these simple heart rate monitors measure, as a rule, only heart rate and do not allow determining the parameters of heart rate variability.

Since the pacemaker is a low-frequency quasiperiodic signal with a normal pulse of 60 beats / min, the heart rate is 1 Hz, heart rate is measured in almost all of the above devices according to the known rules for measuring low frequencies: time intervals between adjacent pulses of the pacemaker (so-called RR intervals) are counted-pulse method [8] (filled the measured intervals with counting pulses from a highly stable reference frequency generator, usually quartz, followed by counting the number of these counting pulses filling the measured interval pulse counter). Heart rate is found as the reciprocal of the measured R-R interval (its average value).

The measurement error of a pulse-counter meter includes the error and frequency instability of the counter pulse generator (usually negligible if the oscillator is quartz) and additionally the specific error of such digital meters equal to ± the period of this counter pulse generator (it corresponds to ± 1 count).

Of the specific developments, one can note an inexpensive microcontroller monitor [7] with an optoelectronic sensor (with LED and photo diode), worn on the finger of the hand, and with LCD. The microcontroller with the Fourier transform allows the LCD to obtain, in addition to the heart rate, the spectral diagram of the EX, which shows the 1st (72 bps) and 2nd (144 bps) harmonics of the spectrum. The measurement error of heart rate was units%.

A device for measuring heart rate (according to the patent of the Russian Federation of 1995 [9]) contains a two-electrode EX sensor, a differential amplifier, a filter, a heart rate detection unit, an executive unit and an additional power source, a resistor and a key. Heart rate is determined by the time between two adjacent beats (heart rate). In the event of a contact failure (in the electrodes), a signal is generated in the executive unit that blocks the analysis of time intervals, and the previous heart rate information remains on the indicating device. Thus, the reliability of the device. Its disadvantage is the impossibility of determining the parameters of heart rate variability.

In [10], a wearable monitor was patented for long-term ECG monitoring. It contains an ECG removal unit with two matrixes of electrodes, an ADC, a removable memory card associated with a microcontroller (MK), means for generating SMS messages via a GSM telephone modem, a three-axis accelerometer, a real-time clock, a USB port, an in-phase noise suppression unit, associated with MK and random access memory (RAM). MK is made with the possibility of entering in the RAM the individual ECG parameters of the patient and comparing with them the ECG parameters of the current monitoring. If the parameters differ more than critical, an SMS message is generated with a diagnosis through the GSM modem of the phone. The disadvantage of this device is its hefty complexity, as well as inability to determine the parameters of heart rate variability.

A device for controlling arrhythmia (RF patent No. 2077863 [11]) allows you to automatically register the moment of increase in the arrhythmia of the patient’s cardiac activity. It contains a clock generator, a unit for measuring the difference between adjacent R-R intervals, a comparison unit, a control unit, and a control unit. With an increase in arrhythmia in the patient's EX, the corresponding signal appears at the device output. The disadvantage of this device is its limited functionality - it performs only the functions of a signaling device. It does not provide opportunities for determining heart rate variability parameters.

The method for determining the functional state of the cardiovascular system (CVS), patented in [12], consists in registering ECS and determining pulse parameters and their coordinated changes at various loads. In this case, 10 RR intervals are recorded, their duration is measured and the amount of K _{RRs of the} same duration is determined. Do this before the load and after the load. When the durations after the load change by more than 30%, the overstrain of the CCC adaptation reactions is diagnosed. At the same time, the duration of the intervals is reduced to 300 ms, fibrillation readiness of the heart is determined and timely preventive measures are taken to prevent sudden death of the patient. This method is not focused on determining the parameters of heart rate variability.

Closest to the proposed purpose is the "Method for the study of heart rate variability" (RF patent No. 2326587 [13] - prototype). This method includes registering an R-R intervalogram and performing its analysis in the frequency domain by the continuous wavelet transform method, in which the cardiointervalogram frequency power is determined by the formula given in the claims. Based on wavelet coefficients, scaleograms are constructed on which physiologically significant frequency ranges are distinguished and then changes in the wavelet power density are determined. The disadvantage of this method is the lack of a visible connection between the obtained parameters and generally accepted parameters of variability in the frequency analysis of ECS, as well as in excessive implementation complexity.

SUMMARY OF THE INVENTION

The proposed method for determining heart rate variability parameters is based on the analogy of the variability and instability of the frequency of quasiperiodic oscillations in radio engineering and on the application of low-frequency EX methods of analysis in the time domain [14, 15]. The method includes converting a heart pulse into an ECM using known sensors, its preliminary amplification and frequency filtering (if necessary), the selection of amplified and filtered ECM pulses corresponding to R-pulses of the ECG.

Next, a series of measurements is made of successive (consecutive) time intervals RR with a digital meter of a pulse-counting type, which includes a software-computing unit with a memory device and a liquid crystal matrix indicator (LCD) -display, the results of these measurements are stored in Memory and form an array of the results of many measurements of a certain volume (the number of elements of this array is predefined). From this array, the average value of the measured intervals and heart rate is calculated in a known manner, as the reciprocal of this average value of the measured R-R intervals, as well as the mean square deviation (standard deviation [3]) of these R-R and heart rate intervals from the average value.

Next, the maximum and minimum values of the measured RR intervals and the difference between them are determined from the resulting array, divide this difference by a certain number of identical (in duration) channels (this number of channels Nк can be determined (set) in advance), thus determining the width of the temporary “window »Of each channel of the resulting multichannel temporary discriminator, count the number of elements in the array of measurement results that fall into each channel of the obtained Nk - channel temporary discriminator, and form the distribution density function of the number of array elements falling into each channel, depending on the channel number (and on the corresponding R-R interval value). A diagram of the obtained distribution function of the measured R-R intervals is presented on the display along with other heart rate parameters and its variability. Finally, dividing the number of array elements by the maximum ordinate of the distribution function, they find the “effective width” of the distribution function (in [14] it is called the effective width of the “spectral line” of quasiharmonic oscillations). This parameter is also displayed.

The shape of the envelope of the distribution function of the measured RR intervals (values of the periods of the heart rhythm) should be for a healthy person under normal conditions, unchanged during the measurement of a set of measurements (then the EX can be considered stationary), a one-humped curve with one maximum (similar to the normal (Gaussian) distribution). The presence of a side maximum in this distribution will indicate the presence of cardiac arrhythmias in such a patient.

The shape of the envelope of the distribution function of the period of quasiharmonic (quasiperiodic) oscillation with chaotic (noise) modulation of the frequency (period), as shown in the authoritative sources mentioned in [14], repeats in a linearly changed scale the shape of the envelope of the power spectrum of this quasiharmonic oscillation. Given the inverse proportionality of the period and frequency for such oscillations will be:

$$T(t)=To(\mathrm{one}+\delta T)=\mathrm{one}/f(t)=\mathrm{one}/fo(\mathrm{one}+\delta t),\left(\text{one}\right)$$

where δT = ΔT / To and δf = Δf / fo are small relative fluctuations of the period and oscillation frequency, respectively. Expanding the right-hand side of (1) in a binomial series and restricting ourselves to the first approximation, we obtain:

$$\delta T=-\delta f,\left(\text{2}\right)$$

respectively. It follows from this that the shape of the envelope of the distribution function of the periods of quasiperiodic oscillations can also be considered as the shape of the envelope of the power spectrum of this oscillation, in which the frequency axis (abscissa) runs parallel to the time axis (period axis) only from right to left (in the direction opposite to the time axis).

In order to minimize the total measurement time and set the aforementioned array of R-R interval measurement results, it is very important that each of the consecutive intervals is measured (without exception). Then, during a measurement time of 5 minutes (accepted in clinical practice [1, 2]), an array of 300 measured values can be obtained for a normal pulse (69 beats per minute) (this is not so much for the formation of a distribution function). For a 20-minute measurement time (this is also practiced in some cases [1, 2, 6]), an array of 1200 measurement results can be obtained (this is already enough to form a distribution function) with the number of channels of the temporary discriminator of about 20-30.

In the device for measuring intervals R-R EX, can be used, in principle, and industrial digital frequency counter-periodometers. Already there are models of such devices in which mathematical processing of measurement results is provided and some dependencies are built on the LCD (for example, PENDULUM frequency meters, CNT-90XL and CNT-91 models costing from 150 to 650 thousand rubles or AKTACOM frequency meters, models ASN-8322, 8324, 8326 worth from 20 to 40 thousand rubles). However, industrial frequency meters measure only every second period of the input signal, since the single-channel pulse counter used in them with an input selector, having measured one period, is restored to the initial state during the next period and is ready for the next measurement one more time after the measured one. To ensure the possibility of measuring each input signal period in a digital RR interval meter according to the proposed method, one should use, as proposed, for example, in [15], a two-channel measurement circuit with two pulse counters (one in each channel), one of which considers “odd” ”, And the other -“ even ”RR intervals, and the results of their measurements are stored in the memory so that they form a single array of measurement results.

The proposed method can be slightly modified to facilitate the procedure for finding the "width" of the learned distribution. To do this, the process of counting the number of hits of the elements of the array of measurement results in each channel (i.e., the set of ordinates of the distribution function) needs to be combined with the process of typing the array itself and end these processes simultaneously when the number of elements in a channel reaches 10 or better than 100 . Then, to determine the "width" of the distribution, the resulting number of array elements (it must be calculated by a special counter) will need to be divided by 10 or 100, for which it is enough to move the comma in the number of elements in the array and one or two decimal places to the left. In this case, the difference between the maximum and minimum values of the measured R-R intervals and the width of the time “window” of each channel must be determined in advance (for example, as a result of preliminary measurements in a shorter time).

To implement the proposed method, it is not necessary to digitize the ECS in the ADC (as in [10] and others) and perform complex computational procedures (digital multiplication, etc.), requiring complex computing devices with a large amount of memory. Therefore, it seems possible to implement this method in fairly simple and compact devices such as electronic watches with LCDs, mobile phones, i-Phone, etc.

The technical result consists in obtaining a distribution function of the measured R-R intervals (and heart rate), the envelope of which contains information about the distribution of frequency components in the spectrum of the electrocardiogram (EKS). To implement the proposed method, it is not necessary to digitize the ECS in the ADC (as in [10] and others) and perform complex computational procedures (digital multiplication, etc.), requiring complex computing devices with a large amount of memory. Therefore, it seems possible to implement this method in fairly simple and compact devices such as electronic watches with LCDs, mobile phones, i-Phone, etc.

Information sources

1. New methods of electrocardiography / Edited by G.G. Ivanova, S.V. Gracheva, A.P. Syrkina. - M .: Technosphere, 2007 .-- 552 p.

2. Kuznetsov A.A. Methods of analysis and processing of electrocardiographic signals: New approaches to the selection of information: monograph. - Vladimir: Vladimir State Publishing House. University, 2008 .-- 140 s.

3. Kuznetsov A.A. Application of the method for assessing heart rate variability in prenosological diagnosis of the functional state of the body. - Measuring equipment. - 2010. - No. 6. - S. 50-55.

4. Yavelov S.S., Kolpakov E.V. Computer analyzer of the pulse wave and electrical activity of the heart "PULSE". - Medical equipment. - 2003 .-- S. 11-16.

5. Ramazanov M.A., Panakhova Z.G. Piezoceramic sensor for registering arterial pulse waves. - Instruments and experimental technique. - 1997. - No. 5. - S. 132-133.

6. System for collecting and processing data on the functions and condition of a person / E.I. Shkelev, V.G. Kuzmin, I.Ya. Orlov et al. - Bulletin of Nizhny Novgorod State University. Series Radiophysics. - 2004. - Vol. 2. - S. 47-54.

7. Sharief F. Babiker, Liena Elrayah Abdel - Khair, Samah M. Elbasheer. Microcontroller Based Heart Rate Monitor using Fingertip Sensor. - University of Khartum Endgineering Journal (UofKEJ) Vol. 1 Issue 2 pp. 47-51 (October 2011).

8. Dalechin A.Yu., Sock C.A. Modern pulsodiagnosis - new possibilities of application in the clinic. - Automation. Automation. Electrotechnical complexes and systems (AAEKS), No. 2 (12), 2003, Information-measuring systems.

9. Device for measuring heart rate. RF patent RU 2042333 C1, IPC A61B 5/024, claimed 06/36/1988, publ. 08.27.1995 / Institute of Cybernetics named after V.M. Glushkova, Academy of Sciences of Ukraine, authors: Mamaev V.M., Onishchenko P.M.

10. Wearable monitor with automatic transmission of the diagnosis through the communication channel in the event of a critical situation. RF patent RU 2444986 C1, IPC A61B 5/0402, A61B 5/0205, A61B 5/024, claimed 07/27/2010 / LLC PO Neurokomelectrontrans, authors: Bonch-Bruevich VV, Kadin I.L., Filatov A.L., Shershukov A.S.

11. Device for controlling arrhythmia. RF patent RU 2077863 C1, IPC A61B 5/0245, A61B 5/0402, claimed 11/23/1993, publ. 04/27/1997 / Kazan State Technical University named after A.N. Tupolev, authors: Mikhailov A.L., Sedov S.S., Scherbakova T.F.

12. A method for determining the functional state of the cardiovascular system (CVS). RF patent RU 2313279 C1, IPC A61B 5/0425, claimed 07/05/2006 / Institute of Physiology. I.P. Pavlova RAS, authors: Kuznetsov D.V., Kuznetsova T.G., Kornyushina N.M.

13. A method for studying human heart rate variability. RF patent RU 233657 C1, IPC A61B 5/0452 / Voronezh State University, author Turovsky Y.A.

14. Ermolenko I.A. Measurement of short-term instability, real width and shape of the spectral line of oscillation of semiconductor generators. - Sat scientific., proceedings "Semiconductor self-oscillating systems and amplifying devices." - L .: LPI, 1984. - S.55-62.

15. The method of receiving signals with angular modulation. RF patent RU 2040860 C1, IPC H04L 27/22, claimed 06/09/1990, publ. 07.27.1995 / Baltic State Academy of the Fishing Fleet, authors: Ermolenko I.A., Pavlov K.P.

Claims (1)

A method for determining heart rate variability parameters, including converting a person’s heart rate into an electrocardiogram using known sensors, pre-amplifying and necessary frequency filtering this signal, extracting from the amplified and filtered cardiosignal pulses corresponding to R-pulses of the electrocardiogram, then making a plurality of measurements consecutive, RR intervals adjacent to each other by a digital pulse meter, incorporating software o-computing unit with a storage device and a liquid crystal matrix display, the results of these measurements are stored in the memory of the meter and form an array of the results of many measurements, the number of elements in this array is predefined, then the average value of the measured RR intervals and pulse rate are calculated from this array as the reciprocal of the obtained average value of the measured RR intervals, as well as the value of the mean square deviation of these RR intervals and frequencies of the average value, characterized in that it further determines the maximum and minimum values of the measured RR intervals and the difference between them, divide this difference by the selected number of channels Nk, which is selected in advance, and thus determine the width of the temporary “window” of each channel thus formed Nk - channel temporary discriminator and then calculate the number of elements of the resulting array of measurement results that fall into each channel of this discriminator and form the distribution function of the number of elements of the mass willows that fall into each channel, depending on the channel number and the corresponding RR interval value, a diagram of the obtained distribution function of the measured RR intervals is presented on the display screen along with other parameters of the heart rate and its variability, finally dividing the number of array elements by the maximum ordinate distribution functions, find the "effective width" of the distribution function, the value of which is also displayed.