RU2489964C2 - Method of determining indices of variability of operator's heart rate in real-time mode and device for its realisation - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: invention relates to medicine. In method realisation electrocardiosignal is filtered, discretised by time. Support points are singled out in q first cardiocycles, durations of cardiocycles, average duration of cardiocycles and number of discrete counts N, corresponding to this duration, average power P₀ of counts are determined and two threshold levels of power Δ₁=1.5P₀ and Δ₂=0.5P₀ are formed. At each step of discretisation total power of following each other N counts is determined, obtained value is compared with threshold levels Δ₁ and Δ₂. On each cardio cycle quantity of counts, corresponding to average duration of all cardio cycles is determined, quantity of taken in turn counts of total power, which exceed Δ₁, and do not exceed Δ₂ is calculated. Device contains filter, unit of discretisation, generator of timing pulses, unit of formation of support points, unit of determining value of average duration of cardiocycles, counter of cardiocycles, comparators, unit of squaring, units of determining average and total power of counts, formers of threshold levels, pulse counters, unit of calculating values variability indices.

EFFECT: invention makes it possible to increase reliability and authenticity of determining index of operator's heart rate variability.

2 cl, 2 dwg

Description

The invention relates to the field of ergatic systems and can be used to determine indicators of heart rate variability of the operator of human-machine systems. The method implemented in the device improves the reliability of determining the functional state of the operator in real time.

The analysis of heart rate variability is a method for assessing the state of the mechanisms of regulation of physiological functions in the human operator’s body, in particular, the general activity of regulatory mechanisms, neurohumoral regulation of the heart, and the relationship between the sympathetic and parasympathetic parts of the autonomic nervous system.

The known method implemented in the device [1], which consists in the fact that allocate QRS-complexes of the electrocardiogram and determine the duration of each RR interval. Substituting the obtained duration values into the formulas, the values of the heart rate variability indicators are calculated. In particular, the value of the average duration of the cardiocycle is determined by the formula:

$$\overline{T}=\frac{\mathrm{one}}{Q}{\displaystyle \sum _{j=\mathrm{one}}^Q{T}_j},\left(\mathrm{one}\right)$$

standard deviation of the full array of normal cardio intervals -

$$SDNN=\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^Q(T_j}-{\overline{T})}^2,\left(2\right)$$

RMS difference characteristic -

$$RMSSD=\sqrt{\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^{Q-\mathrm{one}}(T_j}-{T_{j+\mathrm{one}})}^2},\left(3\right)$$

where T _j is the duration of the j-th RR-interval, Q is the number of RR-intervals.

The disadvantages of this method are:

1. Dependence on the shape of the elements of the EX. Some types of forms of ECS elements can lead to the false isolation of QRS complexes or to the omission of such.

2. Impulse noise can lead to false allocation of QRS-complexes or to the omission of such.

3. The reliability of the allocation of QRS complexes decreases with increasing heart rate.

A known method for detecting arrhythmias, implemented in the device [2], which is the closest to the proposed method (prototype). This method, which consists in determining the duration q of the first cardiocycles, the average duration of these cardiocycles, the number of discrete samples N corresponding to this duration, the average power P ₀ samples, form two threshold power levels Δ ₁ = 1.5P ₀ and Δ ₂ = 0.5 P ₀ and the threshold value of the change in the duration of the cardiocycle N _pore = kN, where 0 <k <1. Then, at each subsequent sampling step, the total power of N samples following one another is determined, the obtained value of the total power is compared with threshold levels Δ ₁ and Δ ₂ , the number of successive samples of total power exceeding Δ ₁ or not exceeding Δ _{2 is} calculated, and in the case the achievement of the calculation of the threshold value of N _then changes in the duration of the cardiocycle form a signal of the presence of arrhythmia.

The disadvantage of this method is that the output signal indicates only the presence of arrhythmia, and does not carry information on indicators of heart rate variability, which can be used to determine the functional state of a human operator.

The proposed method for determining indicators of heart rate variability of the operator in real time eliminates these disadvantages of the prototype.

The essence of the proposed method is as follows. The electrocardiogram is filtered, sampled by time. In q first cardiocycles, reference points are distinguished, the durations of these cardiocycles are determined, the average duration of these cardiocycles and the number of discrete samples N corresponding to this duration are determined, and a time window equal to this duration is formed. During the q first cardiocycles, the average power P _{0 of the} samples is determined, two threshold power levels Δ ₁ = 1.5P ₀ and Δ ₂ = 0.5P _{0 are formed} . The first threshold level is higher than the total power of one cardiocycle P ₀ and its excess indicates that the duration of the cardiocycle has become less than the duration of the time window. The second threshold level is below the total cardiocycle power P ₀ and its non-exceeding indicates that the cardiocycle duration became longer than the time window duration. The movement of the time window is set, and at each step, the total power of the electrocardiogram samples that fall into the time window is determined, and the obtained value is compared with two threshold levels Δ ₁ and Δ ₂ . The value of the total power greater than the threshold level Δ ₁ indicates a decrease in the duration of the cardiocycle relative to the average value of the duration (increase in heart rate), and the value of the total power less than Δ ₂ indicates an increase in the duration of the cardiocycle relative to the average value of the duration (decrease in heart rate). Count the number of alternately taken power readings exceeding Δ ₁ and not exceeding Δ ₂ . On each cardiocycle, the number of samples corresponding to the average duration of all cardiocycles is determined. The obtained values of the number of samples going beyond the levels Δ ₁ and Δ ₂ , the number of samples corresponding to the average duration of all cardiocycles, the number of samples corresponding to the average duration q of the first cardiocycles are used to calculate the values of heart rate variability parameters.

When working with discrete samples, the values of T _j can be replaced by the number of discrete samples of the j-th cardiocycle N _j multiplied by the sampling period Δt:

$$T_j=N_j\cdot \Delta t.\left(\mathrm{four}\right)$$

Formula (1) taking into account (4) can be rewritten in the form

$$\overline{T}=\frac{\mathrm{one}}{Q}{\displaystyle \sum _{j=\mathrm{one}}^Q{T}_j}=\frac{\mathrm{one}}{Q}{\displaystyle \sum _{j=\mathrm{one}}^Q(N_j}\cdot \Delta t)=\Delta t\frac{\mathrm{one}}{Q}{\displaystyle \sum _{j=\mathrm{one}}^Q{N}_j}=\overline{N}\cdot \Delta t,$$

- число дискретных отсчетов, приходящихся на один кардиоцикл средней длительности,Where $$\overline{N}$$

- the number of discrete samples per cardiocycle of medium duration,

$$$$

$$$$

$$\overline{N}=\frac{\mathrm{one}}{Q}{\displaystyle \sum _{j=\mathrm{one}}^Q{N}_j}.\left(5\right)$$

In this way

$$\overline{T}=\overline{N}\cdot \Delta t.\left(6\right)$$

Formula (2) taking into account (4) and (6) can be rewritten in the form

$$SDNN=\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^Q(N_j}\Delta t-{\overline{N}\Delta t)}^2=\Delta t\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^Q(N_j}-{\overline{N})}^2.\left(7\right)$$

Formula (3) taking into account (4) can be rewritten in the form

$$RMSSD=\sqrt{\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^{Q-\mathrm{one}}(T_j}-{T_{j+\mathrm{one}})}^2}=\sqrt{\Delta t\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^{Q-\mathrm{one}}(N_j}-{N_{j+\mathrm{one}})}^2}.\left(8\right)$$

Consider the Q neighboring cardiocycles of one electrocardiogram. These cardiocycles, depending on their number of discrete samples with respect to the value of N, can be divided into three groups:

a) m cardiocycles with the number of samples N _d , N _d > N, where 0≤d≤m, a 0≤m≤q;

b) l cardiocycles with the number of samples N _b , N _b <N, where 0≤b≤l, and 0≤l≤q;

c) a cardiocycles with the number of samples N _z , N _z = N, where 0≤z≤a, a 0≤a≤q.

Respectively

$$Q=m+l+a.\left(9\right)$$

с учетом (5) можно рассчитать следующим образом:Then $$\overline{N}$$

taking into account (5), one can calculate as follows:

$$\overline{N}=\frac{{\displaystyle \sum _{d=0}^m{N}_d+{\displaystyle \sum _{b=0}^l{N}_b+{\displaystyle \sum _{z=0}^a{N}_z}}}}{Q}.\left(\mathrm{one}0\right)$$

The total power of successive N samples, resulting from the conversion of the electrocardiogram, represents the power level of one cardiocycle, relative to which peaks are formed. Moreover, the width of the peaks is determined by the difference between the value of N and the number of samples corresponding to the duration of the current cardiocycle (figure 1). Therefore, the values of the number of discrete samples N _d and N _b can be expressed in terms of the number of samples of sections of the conversion result beyond the threshold levels Δ ₁ and Δ ₂ : ΔN1 _b is the number of discrete samples of the bth section of the conversion result that goes beyond the upper threshold level Δ ₁ , ΔN2 _d is the number of discrete samples of the d-th section of the conversion result that goes beyond the lower threshold level Δ ₂ . Then

N _b = N-ΔN1 _b , N _d = N + ΔN2 _d , N _z = N.

In this case, expression (10) can be written as

, $$\overline{N}=\frac{({\displaystyle \sum _{d=0}^{m}N+{\displaystyle \sum _{d=0}^m\Delta N{2}_d})+({\displaystyle \sum _{b=0}^{l}N-{\displaystyle \sum _{b=0}^l\Delta N{\mathrm{one}}_b})+{\displaystyle \sum _{z=0}^{a}N}}}}{Q}$$

,

after simplification of which, taking into account (9), we obtain

$$\overline{N}=\frac{({\displaystyle \sum _{d=0}^{m}N+{\displaystyle \sum _{b=0}^{l}N+{\displaystyle \sum _{z=0}^{a}N})+({\displaystyle \sum _{d=0}^m\Delta N{2}_d}-{\displaystyle \sum _{b=0}^l\Delta N{\mathrm{one}}_b})}}}{Q}=N+\frac{{\displaystyle \sum _{d=0}^m\Delta N{2}_d-{\displaystyle \sum _{b=0}^l\Delta N{\mathrm{one}}_b}}}{Q}.\left(\mathrm{one}\mathrm{one}\right)$$

может быть определено как сумма значения N определения суммарной мощности отсчетов и отношения разности суммарных длительностей участков ΔN2_d результата преобразования, не превышающих нижний пороговый уровень Δ₂, и суммарных длительностей участков ΔN1_b результата преобразования, превышающих верхний пороговый уровень Δ₁, к числу кардиоциклов Q.Thus, the number of discrete samples corresponding to the average duration of cardiocycles $$\overline{N}$$

can be defined as the sum of the values N for determining the total power of the samples and the ratio of the difference in the total durations of the sections ΔN2 _{d of} the conversion result not exceeding the lower threshold level Δ ₂ and the total durations of the sections ΔN1 _{b of} the conversion result exceeding the upper threshold level Δ ₁ to the number of cardiocycles Q .

можно записать какTo determine the standard deviation of the full array of SDNN cardio intervals, the difference $$N_j-\overline{N}$$

can be written as

, $$N_j-\overline{N}=N\pm \Delta N_j-\overline{N}$$

,

where + ΔN _j = + ΔN1 _j or -ΔN _j = -ΔN2 _j . Here, a negative sign in front of ΔN _j is placed if the result of the conversion exceeds the threshold level Δ ₁ , and a positive sign does not exceed Δ ₂ .

After squaring

$$(N\pm \Delta N_j-{\overline{N})}^2={(N-\overline{N})}^2\pm 2\Delta N_j(N-\overline{N})+{\left(\Delta N_j\right)}^2.$$

Then expression (7) can be written as

$$\begin{array}{c}SDNN=\Delta t\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^Q\left({\left(N-\overline{N}\right)}^2+2\Delta N_j\left(N-\overline{N}\right)+{\left(\Delta N_j\right)}^2\right)=}\\ =\Delta t\frac{\mathrm{one}}{Q-\mathrm{one}}\left(Q{\left(N-\overline{N}\right)}^2+2\left(N-\overline{N}\right){\displaystyle \sum _{j=\mathrm{one}}^Q\pm \Delta N_j}+{{\displaystyle \sum _{j=\mathrm{one}}^Q\left(\Delta N_j\right)}}^2\right)\end{array}.\left(\mathrm{one}2\right)$$

To determine the mean square difference characteristic RMSSD, the difference N _j -N _{j + 1} can be written as

N _j -N _{j + 1} = N ± ΔN _j - (N ± ΔN _{j + 1} ) = ± ΔN _j - (± ΔN _{j + 1} ).

Then expression (8) can be written as

$$RMSSD=\sqrt{\Delta t\frac{\mathrm{one}}{Q-\mathrm{one}}{\displaystyle \sum _{j=\mathrm{one}}^{Q-\mathrm{one}}{\left(\pm \Delta N_j-\left(\pm \Delta N_{j+\mathrm{one}}\right)\right)}^2}}.\left(\mathrm{one}3\right)$$

Expressions similar to (11) - (13) for other indicators of heart rate variability can be obtained in the same way. Thus, the proposed method of analysis of the electrocardiogram allows you to get the values of the heart rate of the operator in real time.

The proposed method allows, in comparison with the known method (prototype), to determine the heart rate variability of the operator for a wide class of electrocardiograms with various modifications of the shape of the elements under the influence of noise.

The invention and a possible implementation of the proposed method is illustrated by the following graphic material:

figure 1 - analysis of the electrocardiogram using a time window;

figure 2 - functional diagram of the device.

To achieve a technical result, which consists in increasing the reliability of determining the heart rate variability of the operator in real time under the action of noise and variations in the forms of cardiac cycle elements, and implementing the proposed method in a device containing a filter, the input of which is the input of the device, and the output is connected to the first input of the sampling unit, the output of the sampling unit is connected to the input of the squaring unit and the information input of the reference point formation unit, the output of the gene the clock pulse generator is connected to the second input of the sampling block, the control input of the block for generating reference points, the first information input of the block for determining the average duration of cardiocycles, the first control input of the block for determining the average cardiac cycle power and the control input of the block for determining the total power of samples, the output of the block for generating the reference points is connected to the input of the cardiocycle counter and to the second information input of the unit for determining the average duration of cardiocycles, the output of the cardiocycle counter is connected to the control input of the unit for determining the average duration of cardiocycles, the output of the squaring unit is connected to the information input of the unit for determining the average power of samples and the first information input of the unit for determining the total power of the samples, the output of the cardiocycle counter is connected to the second control input of the unit for determining average power counts, the output of the block determining the value of the average duration of the cardiocycle is connected to the second information input of the block distribution of the total power of the samples, the output of the unit for determining the average power of the samples is connected to the input of the first driver of the threshold level and the input of the second driver of the threshold level, the output of the unit for determining the total power of the samples is connected to the first input of the first comparator and the second input of the second comparator, the output of the first driver of the threshold level is connected to the second input of the first comparator, and the output of the second threshold level driver is connected to the first input of the second comparator, additionally the first and second pulse counter, a unit for calculating the values of heart rate variability indicators are provided, and the counting inputs of the first and second counters are connected to the output of the clock generator, the output of the first comparator is connected to the zero setting input of the first pulse counter, the output of the second comparator is connected to the second zero input pulse counter, the first, second, third, fourth inputs of the unit for calculating the values of the variability parameters are connected respectively to the outputs of the first and second count Ikov pulse counter cardiocycles, the average value determining block length cardiocycles, calculation unit outputs variability values are the outputs of the device.

The device consists (Fig. 2) of a filter 1, a sampling unit 2, a clock pulse generator 3, a reference point generation unit 4, a cardiocycle counter 5, a unit 6 for determining the average duration of cardiocycles, a unit 7 for calculating the values of the variability indicators, pulse counters 8, 16 , block 9 squaring, block 10 determining the average power of the samples, block 11 determining the total power of the samples, formers of threshold levels 12, 13, comparators 14, 15.

The input of the filter 1, which is the input of the device, receives an electrocardiogram. The output of the filter 1 is connected to the first input of the sampling unit 2, the output of the clock generator 3 is connected to the second input of the sampling unit 2, which controls the input of the reference point generating unit 4, the first information input of the unit 6 for determining the average duration of cardiocycles, the first control input of the unit for determining the average the power of the samples, the control input of the unit 11 for determining the total power of the samples and the counting inputs of the pulse counters 8, 16. The output of the sampling unit 2 is connected to the information input 4 in block form of control points and to the input of block 9 squaring. The output of block 4 of the formation of reference points is connected to the input of the counter of cardiocycles 5 and the second information input of unit 6 of determining the value of the average duration of cardiocycles. The output of the cardiocycle counter 5 is connected to the control input of the unit 6 for determining the average duration of cardiocycles, to the second control input of the unit 10 for determining the average power of the samples and to the third input of the unit 7 for calculating the values of the variability indices. The output of the unit 6 for determining the value of the average duration of the cardiocycles is connected to the fourth input of the unit 7 for calculating the values of the variability indicators and to the second information input of the unit 11 for determining the total power of the samples. The output of the squaring unit 9 is connected to the information input of the average sample power determination unit 10 and to the first information input of the total sample power determination unit 11. The output of the unit 11 for determining the total power of the samples is connected to the first input of the first comparator 14 and the second input of the second comparator 15. The output of the unit 10 for determining the average power of samples is connected to the input of the first driver of the threshold level 12 and to the input of the second driver of the threshold level 13, the output of the first driver of the threshold level 12 is connected to the second input of the first comparator 14, and the output of the second threshold level driver 13 is connected to the first input of the second comparator 15. The output of the first comparator 14 is connected to at the zero setting of the first pulse counter 16, the output of the second comparator 15 is connected to the zero setting input of the second pulse counter 8, the first and second inputs of the variability parameter value calculation unit 7 are connected respectively to the outputs of the first and second pulse counters 16, 8, the outputs of the value calculation unit 7 Variability indicators are the outputs of the device.

This device can be implemented both in analog and digital form. The implementation and operation of blocks 1-6 and 9-15 are given in [2].

The comparator 14 compares the output signal of block 11 with the output signal of block 12, and the comparator 15 compares the output signal of block 11 with the output signal of block 13. When the output signal of block 11 exceeds the first threshold level or does not exceed the second threshold level, then the output of the first or second comparators accordingly, a high level logic signal appears. Under the influence of this signal, the counters are respectively allowed to count 16 or 8 clock pulses received at the counting input from the output of the generator 3 clock pulses. The counting results, along with the output signals of blocks 5 and 6, go to block 7, where the direct calculation of heart rate variability indicators takes place.

Block 1 can be implemented on the K174UN19 chip, block 2 on the K1107PV1B chip, block 3 on the KR531GG1 chip, block 9 on the K561IP5 chip, block 14, 15 on the K561IP2 chip, block 8, 16 on the K561IE chip. Block 7 can be implemented on the basis of microprocessor technology (microcontrollers and microprocessors), the incorporated programs in which (and implementing, in particular, formulas (11) - (13)) will calculate heart rate variability indices from the values of four input signals [3] .

The technical and economic effect of the proposed method and device for its implementation is to increase the reliability and reliability of determining the heart rate variability of the operator in real time under the influence of noise on the cardiac signal and regardless of possible deviations from the norm of the parameters of the QRS complex (shape, amplitude, duration ) Reliable and reliable determination of the values of the heart rate variability of the operator provides a better definition of the functional state of the operator of the human-machine system.

Literature

1. Cardiomonitors. Equipment for continuous monitoring of the ECG / A.L. Baranovsky, A.N. Kalinichenko, L.A. Manilo et al .: Ed. A.L. Baranovsky and A.P. Nemirko. M .: Radio and communication. 1993. S.194-204.

2. RF patent No. 2323339, A61B 5/0402. A method for detecting cardiac arrhythmias in real time and a device for its implementation / A.N. Varnavsky, O.V. Melnik, A.A. Mikheev // Discoveries. Inventions 2008. No. 10.

3. Perelman B.L., Shevelev V.I. Domestic microcircuits and foreign analogues. Reference book, STC Mikrotekh, 1998 - 376 p.

Claims (2)

1. A method for determining the operator’s heart rate variability in real time, namely that the electrocardiogram is filtered, sampled by time, reference points are selected in the q first cardiocycles, the duration of these cardiocycles, the average duration of these cardiocycles and the number of discrete samples N, corresponding to this duration, the average power P ₀ samples, form two threshold power levels Δ ₁ = 1.5P ₀ and Δ ₂ = 0.5P ₀ , then at each next sampling step determine the total power the number of successive N samples, compare the obtained value of the total power with threshold levels Δ ₁ and Δ ₂ , calculate the number of alternately taken samples of the total power as a result of comparison with threshold levels Δ ₁ and Δ ₂ , characterized in that each cardiocycle is determined number of samples corresponding to the average duration of cardiocycles, the number of counts of total power exceeding Δ ₁ and Δ ₂ not exceeding received values along with the value of the number of discrete samples corresponding cf. days of the duration of the first q cardiocycles are used to calculate values of parameters of heart rate variability. 2. The device for determining the heart rate variability of the operator in real time, containing a filter, the input of which is the input of the device, and the output is connected to the first input of the sampling unit, the output of the sampling unit is connected to the input of the squaring unit and the information input of the reference point generation unit, the output of the clock generator is connected to the second input of the discretization block, the control input of the block of formation of reference points, the first information input of the block determining of the average duration of cardiocycles, the first control input of the unit for determining the average cardiocycle power and the control input of the unit for determining the total power of samples, the output of the unit for generating reference points is connected to the input of the cardiocycle counter and to the second information input of the unit for determining the average cardiocycle duration, the output of the cardiocycle counter is connected to the control the input of the unit for determining the value of the average duration of cardiocycles, the output of the squaring unit is connected to the input of the unit for determining the average power of the samples and the first information input of the unit for determining the total power of the samples, the output of the cardiocycle counter is connected to the second control input of the unit for determining the average power of the samples, the output of the unit for determining the average duration of the cardiocycle is connected to the second information input of the unit for determining the total power of samples, output the unit for determining the average power of the samples is connected to the input of the first driver of the threshold level and the input of the second form dividing the threshold level, the output of the total sample power determination unit is connected to the first input of the first comparator and the second input of the second comparator, the output of the first threshold level former is connected to the second input of the first comparator, and the output of the second threshold level former is connected to the first input of the second comparator, characterized in that the first and second pulse counters, a unit for calculating the values of heart rate variability indicators, the counting inputs of the first the second counters are connected to the output of the clock generator, the output of the first comparator is connected to the zero setting input of the first pulse counter, the output of the second comparator is connected to the zero setting input of the second pulse counter, the first, second, third, fourth inputs of the unit for calculating the variability parameter values are connected respectively to the outputs first and second pulse counters, cardiocycle counter, unit for determining the average duration of cardiocycles, outputs of the display value calculation unit Variability factors are the outputs of the device.

Images