ES2384834B2 - METHOD AND APPARATUS FOR CONTINUOUS CONTINUOUS DETECTION OF HEART FREQUENCY FOR IMPEDANCE PLETISMOGRAPHY WITH HID ELECTRODES. - Google Patents

Abstract

Method and apparatus for continuous non-conscious detection of heart rate by impedance plethysmography with hidden electrodes. # A method of measuring heart rate by electrical impedance plethysmography is proposed, the subject being seated and not aware of the measurement. Four electrodes of parallel conductive strips are hidden under the seat upholstery. An alternating current between 10 kHz and 250 kHz is injected with the two external electrodes and the voltage drop between the two internal electrodes is measured. The recorded signal has periodic peaks whose frequency coincides with the heart rate. The waveform and its sensitivity to impedance changes depend on the position and relative dimensions of the detection electrodes relative to those of injection. The heart rate is obtained by an algorithm based on the detection of the slope of the signal.

Description

METHOD AND APPARATUS FOR CONTINUOUS CONTINUOUS DETECTION OF THE HEART FREQUENCY FOR IMPEDANCE PLETISMOGRAPHY WITH HIDDEN ELECTRODES

The present invention relates to a method of detecting heart rate, more particularly to a method of detecting heart rate by impedance plethysmography.

OBJECT

The object of the present invention is to develop a method of continuous detection of heart rate by impedance plethysmography, based on hidden electrodes and without the subject being aware of the measurement. A second object is to develop an apparatus for recording the impedance plethysmogram (IPG) based on hidden electrodes, continuously and without the subject being aware of the measurement.

BACKGROUND OF THE INVENTION

Daily monitoring of health status outside clinical settings or medical centers contributes decisively to the discovery of possible health disorders before they become important [see, M. Ishijima, "Unobtrusive approaches to monitoring vital signs at home", Medical and Biological Engineering and Computing, vol. 45.11, pp. 1137-1141,2007] And the improvement of lifestyle, which can help prevent diseases. Daily control can also help the monitoring of chronic patients and the care of the elderly, thus contributing to their health and well-being.

Home health control offers greater comfort for patients, which is very important for the elderly, patients with chronic diseases and people with reduced mobility, for whom it can be difficult and expensive to visit a hospital or medical center to Your routine controls. For a greater advantage, health control during daily life should be carried out with a minimum participation of the subject and without interrupting their daily activities [see, M. Ogawa and T. Togawa, "The concept of the home health monitoring," in 5th International Workshop on Enterprise]. Measures that go unnoticed meet these requirements, and even more so if the measurement does not require any direct contact with the subject's skin.

Various methods of using physical methods for measuring parameters related to people's health have long been known in the art. Among them, impedance plethysmography has been continually refined and is finding new applications.

W02009072023 (A 1), by Herleikson Earl; Crone William describes a system and method for determining changes in blood volume / flow in a patient. The electrodes are placed on the patient's forehead and an alternating current is injected through a first group of a plurality of electrodes (16a, 16d). The voltage is measured at each of a second plurality of electrodes (16c, 16b) and from these voltages an impedance is calculated whose value is used to determine changes in blood volume / flow. It is appreciated that this method is based on measurement only at a certain time, on conscious disposition and making contact with the patient's skin.

US20080249433 (A 1) by Stahmann Jeffrey E; Hatlestad John; Moon Boyce discusses, among others, a device for managing the heart rhythm or other implantable medical device, which uses thoracic impedance to determine how much fluid is in the chest, so that congestive heart failure, pulmonary edema can be detected or predicted , pleural effusion, hypotension and the like. Changes in blood resistivity, which may be due to changes in hematocrit or other factors, are compensated so that they do not affect the determination of the amount of thoracic fluid. As we see, this method consists of an implant that the patient carries in his body.

Document ES2296474 (A1) by Pallás Areny Ramón; Piedrafita Jaime Oscar Houses; Gonzalez Landaeta Rafael, is based on the measurement of the electrical impedance between the feet due to the ejection of blood at each heartbeat. One way to implement this method is to simply stand on a surface so that each foot comes into contact with one or two conductive areas through which an alternating electric current is injected and the voltage drop is measured, whose amplitude reflects the change of the internal electrical impedance of the body produced by each beat. The document includes the proposal of a device to obtain a heart rate measurement.

In application W00141638 (A1) of Wallace Arthur W; Shmulewitz Ascher proposes an apparatus and method to monitor heart rate using bioelectrical impedance techniques, in which three orthogonal pairs of sensor electrodes (18a-18e) are placed in the trachea or esophagus in the vicinity of the aorta. Meanwhile, an excitation current is injected into the mass of the tissue intervened through a current electrode (20), so that the bioelectrical impedance measurements based on the voltage drop measured by the sensor electrodes reflect the voltage changes induced primarily by the dynamics of blood flow. The method and apparatus are directed to the control of the administration of intravenous fluids or medication, as well as to the optimization of the heart rate controlled by a pacemaker. Another document by the same authors, document US6095987 (A) by Shmulewitz Ascher and Wallace Arthur A, is related to the previous document because it also claims an apparatus and method to monitor the amount of blood ejected by the heart using bioelectric impedance techniques in which the first and second electrodes are placed in the trachea or bronchi in the vicinity of the ascending aorta, while an excitation current is also injected into the thorax via the first and second current electrodes, so that the impedance measurements based on the voltage drop measured by the sensor electrodes reflect the voltage changes induced primarily by the dynamics of blood flow, rather than by the physiological effects related to breathing or non-cardiac. The method and apparatus, by means of the supervision of the blood ejected by the heart, are directed to the control of the administration of intravenous fluids or medication in the organism, as well as to the optimization of the heart rate controlled by a pacemaker. In both cases it involves the introduction of an implant in the body.

Ross B GB2367896 (A); Bayford Richard, describes a device for monitoring the changes in blood impedance caused by the heart's pumping cycle. The device comprises means of signal generation to be applied to a subject whose heart rate should be monitored; means for detecting an output signal that results from said signal when circulating through a part of the body; means for monitoring the variation of the output signal to monitor the changes caused in the signal applied by the changes in impedance in the signal path due to blood flow through said body part; and means for determining the heart rate from the variation of the output signal. The proposed method is punctual and the subject is aware of its use.

Document US5879308 (A) by Raesaenen Taisto refers to a procedure for measuring impedance in a patient. By measuring the impedance, it is possible to monitor the patient's breathing and blood circulation. In the procedure, a plurality of electrodes (1 a, 1b; 1d, 2; 2a) is connected to the patient and changes in the interrelation between the impedances between the electrodes (1a, 1b; 1d, 2; 2a) are measured. The invention allows using the same electrodes and cables used to obtain the electrocardiogram (EGG). Likewise, document EP0747005 (A1) by Raesaenen Taisto is related to the previous document as it also refers to a procedure to measure the impedance in a patient, and from it to monitor the breathing and blood circulation of the patient.

There are some works dedicated to the measurement of cardiovascular signals at a distance. Harland et al. [see, "Electric potential probes-new directions in the remote sensing of the human body," Measurement Science and Technology, vol. 13, pp. 163-9, 2002] developed an amplifier with a very high input impedance to measure bioelectric signals without any physical contact with the body. They recorded the heartbeat with an electrode placed 1 m from the chest, as well as the electroencephalogram (EEG) [see, CJ Harland, TD Clark and RJ Prance, "Remote detection of human electroencephalograms using ultrahigh input impedance electric potential sensors," Appl . Phys. Lett., Vol. 81, pp. 3284-6, 2002] with a separation of 3 mm of air between the hair and the electrodes.

The origin of the real signal detected when the electrical potential is measured at a distance from the surface of the body and without any contact is a controversial issue because variations in the distance between the body and the electrodes can contribute to the potential differences detected. [see, O. Casas and R. PallasAreny, "Electrostatic interference in contactless biopotential measurements," in 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2007, pp. 2655-8].

In another work [see, C. J. Harland, T. D. Clark and R. J. Prance, "High resolution ambulatory electrocardiographic monitoring using wrist-mounted electric potential sensors," Measurement Science and Technology, vol. 14, pp. 923-8, 2003] ECG was obtained a few millimeters from the wrist with capacitive electrodes (which implies a physical contact but not a conductor), using a wristwatch type interface. Therefore, it is a liable technology, which causes minimal disturbance, but cannot yet be considered truly unconscious.

Measures based on direct contact between bare skin and the sensor do not necessarily disrupt daily activities. Tanaka et al. in: "Fully automatic system for monitoring blood pressure from a toilet-seat using the volumeoscillometric method," in 27th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2005, pp. 3339-3941, measured blood pressure (TA) in a person sitting in a sink. They used the oscillometric method with a contact area of 40 mm in the thigh. Changes in blood volume were detected with an optical sensor.

The above measures, however, only last a short time interval, so that they are occasional rather than continuous, and also the subject is aware of them. Measures that go unnoticed to the subject, based on sensors hidden in the furniture, can provide longer periodic records, and even continuous records if artifacts produced by the movements of the subject are reduced. Using conductive electrodes covered by the upholstery of a chair, Lim et al. [see "ECG measurement on a chair without conductive contact," IEEE Transactions on Biomedical Engineering, vol. 53, pp. 956-9, 2006] obtained the heart rate (HR) from the ECG of a person sitting and dressed in normal clothes. The amplitude of the detected ECG depended strongly on the proximity of the body to the electrodes since the capacitance of the electrodes decreases when the body is separated from the seat and instead the amplitude of the superficial ECG is constant. However, by varying the physical parameters and the geometry of the system, a method could be achieved to determine FC that was really effective. For example, measurements that involve an excitation signal can compensate for an increased separation between the body and the electrodes using a larger amplitude excitation signal.

The analysis of the existing information to date shows that there is no continuous measurement method that can measure the heart rate while the subject is performing another activity and without being aware of the performance of such measures. Such a method may be the determination of heart rate by impedance plethysmography with hidden electrodes when the subject is seated. Surprisingly we have found that by applying a sufficiently large excitation signal, the electrodes can detect impedance changes in the subject, e.g. eg, of the thigh, through clothing and upholstery, without the latter being aware of the measurement.

DESCRIPTION OF THE INVENTION

We have developed a new method and device to measure the heart rate of a seated subject that is based on detecting blood flow in the thigh using impedance plethysmography with hidden electrodes.

The principle of operation of the apparatus consists of a set of four parallel strip electrodes, placed under the thigh, with transverse direction with respect to it, and hidden under the seat upholstery. The apparatus has means for injecting an alternating current of between 10kHz and 250kHz through two external electrodes and for detecting changes in the voltage drop between the pair of internal electrodes.

Surprisingly we have found that by applying a sufficiently large excitation signal, the electrodes can detect impedance changes at a certain distance, e.g. eg, on the thigh, through clothing and upholstery. The subject is not aware of the measure taken. Qualitative conclusions regarding the relationship between electrode separation and sensitivity to deep vessels in conventional impedance plethysmography (i.e., with electrodes in contact with the skin) offer a guide to custom design a series of hidden electrodes under the upholstery of different seats, such as normal chair, sofa, cushion, etc.

Impedance plethysmography detects changes in volume by measuring the electrical impedance of the tissues of interest in an area of the body. In the extremities, these volume changes are mainly due to blood flow. The impedance plethysmography measurements common in a member use four band electrodes around it to produce a homogeneous electric field in the target volume. The IPG can also be obtained by round electrodes (used in electrocardiography), but then the detected signal depends more on the position of the electrodes, a fact that may be important to estimate blood flow, but which is not essential in the detection of heart rate.

We have used a set of four flat electrodes (see Fig. 1) placed under the thigh and transverse direction to it. The two external electrodes are used to inject an alternating current and the two internal electrodes measure the voltage drop across the biological tissues located above them. If the electrical conductivity of those tissues changes, for example due to blood flow, the detected voltage, which will be of the same frequency as the applied alternating current, will have an amplitude that will depend on the baseline impedance and those changes in conductivity. We have built electrodes with adhesive tape and external electrodes to detect changes in the voltage drop between the pair of inner electrodes. conductive that is thin enough to go unnoticed when placed under the seat upholstery. Since the excitation and detection are performed by different pairs of electrodes, the contact impedance between the skin and the electrode will have no influence, provided that the amplifier has a sufficiently high input impedance [J. G. Webster, Medicallnstrumentation: Application and Design. John Wiley & Sons, 1998].

Let us now see what role the position and size parameters of the electrodes play, as well as the characteristics of the signal detection, in the impedance plethysmography heart rate detection system of the present invention.

2.1. POSITION AND SIZE OF THE ELECTRODES

Figure 1 shows a set of four parallel ribbon electrodes to obtain the IPG in the thigh without touching the skin. Both pairs of electrodes are symmetrically positioned with respect to the center line of the assembly (dashed line). The distance D between the current injection electrodes should be as large as possible, to involve the largest possible volume of the measured leg segment. The optimal separation d between the voltage sensing electrodes, however, is not known in advance.

Some authors have studied the sensitivity to local conductivity changes in bipolar and tetrapolar measurements of transfer impedance in models, whether using circumferential electrodes (bands) around a uniform anisotropic cylinder [see articles by FA Anderson Jr, "Impedance plethysmography in the diagnosis of arterial and venous disease, "Annals of Biomedical Enginneirng, vol. 12, pp. 79-102, 1984; And that of B. C. Penney, L. M. Narducci, R. A. Peura, F. A. Anderson and H. B. Wheeler, "The Impedance Pletismographic Sampling Field in the Human Calf," IEEE Transactions on Biomedical Engineering, vol. BME-26, pp. 193-8, 1979], either using round electrodes in a semi-infinite homogeneous medium [see P. Bertemes-Filho, B. H. Brown, R. H. Smallwood and A. J. Wilson, "Tetrapolar Probe Measurements: Can the Sensitivity Distribution be Improved?" Proc. 11th Int. Conf Electrical Bio-Impedance (ICEBI) 2001, pp. 561-4; as well as B. Brown, A. Wilson and P. Bertemes-Filho, "Bipolar and tetra polar transfer impedance measurements from volume conductor," Electronics LeUers, vol. 36, pp. 2060-2, 2000]. These studies have shown that tetrapolar measurements are less sensitive than bipolar measurements, but are more robust to artifacts due to impedance variations in the surface layers near the electrodes. Moreover, the sensitivity of bipolar measurements decreases with tissue depth, but for tetrapolar measurements in a semi-infinite homogeneous medium it is maximum at a depth that depends on the relationship between the distance between the current injection electrodes and the distance between the voltage sensing electrodes.

For circumferential electrodes, the closer the two pairs are, the greater the sensitivity to the vessels located near the surface. If the distance between the current injection torque increases and the distance between the voltage sensing torque remains the same, the sensitivity for the deeper vessels relative to the superficial vessels increases, although not proportionally [see

B. C. Penney et al., Ibid.]. Furthermore, the relationship between electrode separation and sensitivity for deep vessels depends on the diameter of the member. Our set of four electrodes of parallel strips, however, should fit any seated adult, and therefore should be designed so that the current, or part of it, is always injected into the thigh whatever its dimensions.

It has been reported that round electrodes (used in electrocardiography) can replace circumferential electrodes in blood flow measurements based on impedance plethysmography if they are in direct contact with the skin [see 1. G. Schraibman, ibid.]. The best results are obtained when the electrodes are on opposite faces of the member (current injection electrodes on the front face and voltage sensing electrodes on the rear face). But for a hidden measure, the electrodes must be placed under the thigh, all on the same side. In addition, our electrodes are on a flat surface while each pair of round electrodes on the member can be along a different line, even if they are placed on the same face of the member, which can be advantageous for aiming at voxels ( a volume element, the three-dimensional version of a specific pixel). On the other hand, the area of the round electrodes is fixed to those of commercially available models, while in our set of four electrodes we can design different areas, in addition to selecting the distance D between the current injection electrodes and the spacing between The voltage sensing electrodes. As for the length b of the electrodes, it should be longer than the diameter of the thigh for a seated person, so that the measurement is as independent as possible of the position of the thigh with respect to the particular seat for which They are designed. The width A of the current injection electrodes must be as large enough to obtain a uniform distribution of the electric current lines along the thigh. The effect of the width a of the voltage sensing electrodes on the detected signal is a variable that must be analyzed.

2.2. SIGNAL DETECTION

Figure 2 shows the measurement system we have developed and its components. It shows that the impedance changes due to blood flow in a body segment are synchronously demodulated, using the carrier signal c (t) as a reference; the time-invariant bioimpedance component (baseline impedance) is rejected by FPB1; G2 amplifies small changes in the impedance signal; and FPB2 eliminates the component due to breathing and reduces noise bandwidth.

The detection is based on injecting a sinusoidal current c (t) of between 10kHz and 250kHz, which are frequencies high enough to reduce the impedance of the Cee and Cev capacitive electrodes, but not so high as to require amplifiers with a gain product by very large bandwidth. The amplitude modulated s (t) signal, detected by the Cev electrodes, is the product of c (t) times the baseline impedance Zo plus the signal related to the blood flow z (t) . As a result, the frequency of this signal fz is transferred to higher frequencies and gives two lateral bands around the carrier frequency fe. We recover Zo + z (t) by coherent (or synchronous) demodulation using the excitation signal as a reference. Before demodulating, the detected differential voltage is amplified by an instrumentation amplifier (Al). Since the equivalent signal source is capacitive, the front of the amplifier includes a network of T-shaped resistors (see R1, R'1 and R2 below in Fig. 3) to provide a ground path for the currents of input polarization. Figure 3 shows the equivalent input circuit for the impedance measurements, as well as the capacitive input electrodes Cev that form a voltage divider with the input capacitance C¡ and C¡ 'of the Al. These resistors R1, R '1 and R2 can be selected of a sufficiently large value so that the impedance between each signal line and ground is determined by the capacitance of the corresponding amplifier input (C¡, C'¡). Therefore, each voltage detected will be attenuated according to

Once amplified by G1, the signal is filtered high pass to block the offset voltage (FPA1).

The synchronous amplitude demodulator can be implemented by means of a

switching gain amplifier (± 1), controlled by a square signal

symmetric obtained from the carrier signal by means of a voltage comparator. After

of demodulation, a pass-band filter (FPB1) rejects the continuous component

5

(due to baseline impedance) and any harmonics that result from demodulation.

As the impedance variations produced by blood circulation are in

the range of 0.1% to 1% of the total impedance [R. Patterson, "Bioelectric

impedance measurements, "in The Biomedical Engineering Handbook, J. D. Bronzino,

1st

Ed. CRC Press, IEEE Press, 1995, pp. 1223-1230], the signal is further amplified by an amplifier with G2 gain before being filtered again with a pass filter

band to attenuate the respiratory components and reduce the bandwidth of the

noise (FPB2).

fifteen

BRIEF DESCRIPTION OF THE DRAWINGS

A preferred embodiment of the invention of

agreement with the accompanying figures, in which:

Fig. 1 presents a set of four electrodes of parallel strips for

measure the IPG in the thigh, where both pairs of electrodes are symmetrically

twenty

placed with respect to the center line (dashed line) and arranged

transversely to the thigh.

Fig. 2 presents the IPG measurement system, where changes in

impedance due to blood flow in a segment of the body are synchronously

demodulated using the carrier signal, c (t}, as a reference.

25

Fig. 3 presents the equivalent input circuit to measure eIIPG.

Fig. 4 presents the results of experiment 1, where the upper curve is

the IPG obtained with the ratios diO = 0.2 and alA = 1, and the lower one is the

photoplethysmogram, used as reference.

Fig. 5 presents the results of experiment 2, where the upper curve is

30

the IPG obtained with the ratio diO = 0.6 and alA = 1, and the lower is the

photoplethysmogram, used as reference.

Fig. 6 presents the results of experiment 3, where the upper curve is

the IPG obtained with the ratios diO = 0.2 and alA = 0.25, and the lower one is the

photoplethysmogram, used as reference.

35

Fig. 7 presents the results of experiment 4, where the upper curve is

the IPG obtained with the ratios diO = 0.6 and alA = 0.25 And the lower one is the

photoplethysmogram, used as reference. Fig. 8 presents the spectral power density dellPG obtained with dlD = · 0.2 Ya) alA = 1, Yb) alA = 0.25

Fig. 9 presents the results of experiment 5, performed with electrodes

hidden, where the upper curve is the IPG obtained with the dlD = 0.2 Y alA ratios

= 0.25, and the lower one is the photoplethysmogram, used as a reference.

Fig. 10 presents the spectral content of the IPG measured with the electrodes: a)

visible and b) hidden, where it is appreciated that the information related to the

breathing is clearer with visible electrodes.

Fig. 11 presents the heartbeat (vertical lines) detected from

dellPG using an algorithm based on the slope of the signal.

Fig. 12 presents the results of experiment 6, in which it was obtained

the IPG with hidden electrodes, in successive times. The dimensions of the

electrodes were D = 27 cm, b = 47 cm, A = 5 cm, d = 0.2 x D, yen a) alA = 0.5, yen b)

alA = 0.25.

DESCRIPTION OF A PREFERRED EMBODIMENT

The determination of the influence of the position and size of the electrodes is the way to get the best results. For this purpose we carry out a series of experiments varying different dlD relationships, the areas of the voltage sensing electrodes and the overall dimension. First, we vary the distance d for a fixed width a for visible electrodes placed in the seat, so that the only separation between the electrodes and the thigh was the subject's pants. Then we reduced to and changed dlD again. Finally, we recorded the IPG when the electrodes were covered by the synthetic fabric upholstery. Table 1 summarizes the experiments. The fixed dimensions were D = 27 cm and A = 5 cm. D and b were selected so that the area covered by the electrodes was larger than the area covered by the thigh of a seated person. For b two values were tested: b = 22 cm and b = 47 cm.

Table 1. Experiments to measure ellPG with capacitive electrodes. G, = 3000 in all experiments.

Experiment

b = dlD = alA = G1 = Figure

one

22 cm 0.2  one 16 4 (8)

2

22 cm 0.6 one 6.7 5

3

22 cm 0.2 1/4 3. 4 6 (8)

4

22 cm 0.6 1/4 7.5 7

5 (hidden)

22 cm 0.2 1/4 500 9, (10), 11

6 (hidden)

47 cm 0.2 1/4 49 12th

1/2

82 12b

5 The electronic circuits were fed at ± 12 V, the applied peak voltage was 10 V, the frequency of the current 50 kHz, which is a usual value in the bioimpedance measurement, and G1 was adjusted for each subject and experiment to obtain the maximum output in Al. This ensured that the amplitude of s (t) after G1 was the same for all experiments. Al (INA111) has: an impedance of

10 common mode input R¡ = 1 TO in parallel with C¡ = 3 pF And an input polarization currents (typical) of 2 pA. We select R1 = R'1 = 1 MO and R2 = 10 MO. At 50 kHz, the impedance of C¡ is 1 MO, much lower, then, than the common mode input resistance. If the R¡ were too small, Cev and R¡ would form a high pass filter that could attenuate the modulated signal.

15 The cutoff frequency of the filters in fig. 2 was fL1 = 1.6 kHz for FPA 1, fL2 = 0.5 Hz and fH2 = 10Hz for FPB1, and fL3 = 0.5 Hz and fH3 = 10 Hz for FPB2. All filters were first-order liabilities. The demodulated signal was amplified by G2 = 3000. To obtain a reference signal, the photoplethysmogram (PPG) was recorded.

20 simultaneously with the IPG. Each signal was digitized by a 12-bit resolution recorder that took 2000 samples every second. A personal computer processed the digitized signals. All the measures shown here in each case were obtained in the same person.

EXPERIMENT N ° 1 Figure 4 shows the IPG (upper table) and the PPG (lower table) obtained in the example of a volunteer. The IPG was obtained with voltage sensing electrodes of width a = A separated by a distance d = 0.2 x D. An ascending peak in the IPG is clearly visible in each cardiac cycle, which means a decrease in impedance.

EXPERIMENT No2 Next, in the example of the same volunteer, we increased the separation between the voltage sensing electrodes to a d = 0.6 x D, while maintaining the value of a = A. Figure 5 shows that ellPG does not show a clear peak in each beat

cardiac, which means reduced sensitivity to blood flow. The heart rate is not distinguishable by the naked eye from the IPG. EXPERIMENT N ° 3

To reduce the effect of the width of the voltage sensing electrodes, we reduced it to a ratio a = A / 4, and set the distance d = 0.2 x D, the same as in experiment 1. The result for this The experiment is shown in Figure 6. The synchronous rising peaks with each heartbeat are clearly present in the IPG (upper frame), and their amplitude is approximately eight times larger than that obtained in Figure 4 for wider electrodes, where dlD It was the same but the relationship to A was four times greater.

EXPERIMENT No. 4 To assess the extent to which the results of Experiment No. 3 were due to an increased sensitivity for the narrower electrodes, in this experiment we increased the distance of 0.6 x D, the same as in experiment 2. The Figure 7 shows the recorded signals. Although the IPG is smaller than that obtained in fig. 6, obtained with the same alA ratio but with a dlD ratio three times smaller, the heart rate is clearly distinguishable. The amplitude of the IPG decreased by a factor of 5, but it was still larger than that obtained with the closest and widest electrodes (compare with Fig. 4), and the heart rate was clearer than in the graphs obtained with more electrodes wide and placed at the same distance (compare with fig. 5). Comparing ellPG in figures 4 and 6, we observe that in addition to the greater amplitude of the peak of the IPG in fig. 6, the fluctuations of the peaks due to the

Breathing also seems to be larger when we use narrower electrodes. Fig. 8 shows the spectrum of both signals; The spectral power density of the IPG was obtained with a ratio dlD = 0.2, and the ratios: a) alA = 1, and b) alA =%. The results confirm that narrower voltage sensing electrodes better detect the impedance changes related to breathing, since the low frequency components (which are due to breathing) are greater when the electrodes are narrower.

It has been reported that for circumferential and round electrodes, sensitivity to deeper blood vessels is reduced when the d / D ratio increases. We here have observed the same effect for our set of four electrodes (see Figures 4 and 5). However, for any d / D ratio, the amplitude of the IPG can be increased by reducing the width of the voltage sensing electrodes (see Figures 6 and 7).

EXPERIMENT N ° 5 Taking into account that the highest corresponding signal allPG was the one obtained with the narrowest electrodes, separated a short distance d, we selected the values a = A / 4 and d = 0.2 x D to register the IPG with the electrodes hidden (experiment 5). Figure 9 corresponds to the present experiment 5 where the IPG is presented in the upper frame, measured by hidden electrodes with the ratios d / D = 0.2 and alA = 1/4, and in the lower frame the PPG is presented. Figure 9 shows that the amplitude of the IPG decreases by 50% compared to that obtained with visible electrodes (see fig. 6) and is louder. Figure 10 shows the spectral content of the IPG measured with electrodes a) visible, and b) hidden. It is appreciated that the information related to breathing is clearer with the visible electrodes (closer to the leg). That is, Figure 10b confirms that the signal-to-noise ratio (SNR) decreased to the point that the spectral contents in the respiratory rate are almost buried in the noise, while in Figure 10a, obtained with the visible electrodes, Frequency bands attributable to breathing were evident. However, the amplitude and waveform of the IPG obtained with hidden electrodes are good enough so that the heart rate can be easily extracted by identifying the peak in each cardiac cycle, for example with an algorithm based on a detection of the slope of the signal at each point. Figure 11 identifies heartbeats by vertical lines, detected from the IPG (see the upper signal) by an algorithm based on the slope.

EXPERIMENT N ° 6 To ensure that the electrodes are always under the thigh, its dimension b can be extended until it is almost equal to the width of the seat. Thus, at least one of the thighs will remain on the electrodes. Fig. 12 shows that indeed ellPG continues to show peaks coinciding with cardiac activity, and that when the detection electrodes are narrower (lower frame), respiratory activity is more evident, as was also observed when comparing Figures 4 and 6. Table 1 above shows that Zo was larger (ie, G1 had to be smaller) for separate voltage sensing electrodes. On the contrary, by increasing the separation between the electrodes and the leg (for example, when the electrodes are hidden), a much larger G1 needs to be arranged. But the heart rate remains clearly distinguishable. The decrease in capacitance due to the greater separation between the electrodes and the leg is compensated if it is measured simultaneously on both legs with the same electrodes (electrodes with a higher value of b since the contact area is increased. You can see in Table 1 above, where it is observed that experiment 6 required a much smaller G1 gain than experiment No. 5. The IPG recorded with electrodes without direct contact with the body is very susceptible to motion artifacts, and these they will undoubtedly be present in the prolonged records, however, if the main objective is to monitor the status of the subject, the movement is unavoidable because it implies its activity, if there were no activity, there would be no movement and the IPG would reveal the absence beats, moreover, records during a person's rest periods could be used to periodically monitor their health, knowing that even during In a nap, personal health is being monitored, it helps to feel safe.

SUMMARY

We have developed an apparatus based on a set of four electrodes of parallel strips, placed under the thigh in a transverse direction with respect to it. It can be interpreted that this set of electrodes is as if the band electrodes that in conventional impedance plethysmography are placed around one limb and in contact with the skin, would have been "open" until they stopped making contact with the skin. The two external electrodes are used to inject an alternating current and the two internal electrodes measure the voltage drop across the tissues above it; If the electrical conductivity of these tissues changes, for example due to blood flow, the voltage detected, which will be of the same frequency as the applied current, will have an amplitude that will depend not only on the baseline impedance but also on the corresponding changes in conductivity. Such electrodes may be constructed of conductive adhesive tape thin enough to go unnoticed when placed under the upholstery of seats, such as chairs, sofas or cushions. Thus, the qualitative conclusions regarding the relationship between electrode separation and sensitivity to deep vessels in conventional impedance plethysmography offer a guide to the custom design of a cluster of electrodes hidden under the seat upholstery.

We have developed a new method for non-conscious measurement of heart rate that is based on the interpretation of the results provided by the detection of blood flow in the thigh by means of impedance plethysmography implemented with hidden electrodes. Thus, the experiments carried out allow us to find the qualitative relationship between the waveform of the IPG and the amplitude of its peaks and the distance and width of the voltage sensing electrodes with respect to those of the current injection electrodes. Regarding the distance, the lower the dlD ratio, the greater the IPG (see fig. 4 compared to fig. 5; and fig. 6 compared to fig. 7). For any given dlD ratio, the IPG increases when the width of the voltage sensing electrodes decreases. The amplitude of the IPG decreases when the distance between electrodes and the thigh increases, for example, when the electrodes are under the upholstery. To ensure that the thigh is always on the electrodes, it is convenient that b is almost equal to the width of the seat. Thus, we have developed a method of determining, for the different specific cases, the optimal values of the dlD and alA relationships, where D is determined by the depth of the seat in question, both in terms of the amplitude of the IPG, and in terms of to the detection of impedance variations related to breathing. The amplitude and waveform of the IPG obtained with hidden electrodes are good enough so that the heart rate can be easily extracted by identifying the peak in each cardiac cycle with an algorithm based on detecting the slope of the IPG. This method constitutes the key to the design of heart rate determination devices according to the different specific cases.

Since the electrodes can be hidden under the upholstery of a normal chair, sofa or cushion, the proposed method can provide a useful means for continuous monitoring of heart rate in non-clinical settings, that is, in everyday life settings. normal. In the absence of motion artifacts, the heart rate can be extracted from the IPG by a simple algorithm based on slope detection.

Once the invention is sufficiently described as well. as of a preferred embodiment thereof, it should only be added that it is possible to make modifications in its constitution and materials used without departing from its scope, defined in the following claims.

Claims (4)

1.-An apparatus for continuous non-conscious measurement of heart rate based on impedance plethysmography characterized in that it comprises two pairs of electrodes with parallel strips, hidden inside the seat material and placed in a direction transverse to the thigh, in which two External electrodes constitute a first pair and two internal electrodes constitute the second pair, and because said electrodes are constructed of a very thin conductive material imperceptible to the eye, such as a tape or a sheet. 2. The apparatus for continuous measurement not aware of the heart rate according to claim 1, characterized in that said pair of external electrodes is used to inject an alternating current of a frequency between 10kHz and 250kHz; And said pair of internal electrodes is used to measure the voltage drop across the biological tissues located above said pair; and where, if the electrical conductivity of the tissues changes due to blood flow, the tension
detected will have an amplitude that will depend on the baseline impedance and the corresponding changes in conductivity at each beat. 3.-A method for continuous non-conscious heart rate measurement based on impedance plethysmography characterized in that the separation distance D between the current injection electrodes is determined by the seat depth and should be as large as feasible. ; the length b of the electrodes should be greater than the diameter of the thigh for the seated person and preferably equal to the width of the seat; the width A of the current injection electrodes should be large enough to obtain a uniform distribution of the current lines along the thigh; and because the optimum separation d between the voltage sensing electrodes, and the width a of the voltage sensing electrodes are determined by finding the optimal values of the dlD and alA ratios, both in terms of the amplitude of the plethysmogram (IPG), as for the detection of impedance variations related to breathing. 4. The method for continuous non-conscious measurement of the heart rate according to claim 3 characterized in that the amplitude and waveform of the IPG obtained with hidden electrodes are suitable so that the heart rate can be easily extracted by identifying the peak in each Cardiac cycle with an algorithm based on the detection of the dellPG slope at each point.