RU2565804C1 - Method for computer-aided remote estimation of pulse wave velocity - Google Patents

Abstract

FIELD: medicine.

SUBSTANCE: pulsograms of two pulsating body surface areas above the examined arteries are recorded; a distance of these two areas and a wave travel time are measured; a pulse wave velocity is calculated. The pulsating body surface areas are recorded simultaneously by a video camera detecting video images in the visible or infrared electromagnetic wavelength band. The pulse wave travel time is determined by a cardiac cycle diagram shift of the recorded pulsograms.

EFFECT: method enables increasing the reliability of the pulse wave velocity estimation that is ensured by avoiding the use of a contact sensor affecting a vascular wall and providing the proper positioning of the sensors with respect to the examined arteries.

3 cl, 5 dwg, 1 ex

Description

1. The technical field

The invention relates to medical equipment, and in particular to methods and devices for functional non-contact diagnosis of the state of the human cardiovascular system by assessing the propagation speed of a pulse wave.

2. The level of technology

The pulse wave velocity is an important assessment of the state of the arteries of the cardiovascular system. As is known, a number of vascular diseases are characterized by changes in the morphology of the vascular wall (stiffness of the vascular wall). An increase in this parameter due to age-related changes or due to various pathologies (atherosclerosis, muscle hypertrophy) leads to an increased risk of major cardiovascular complications (Francesco US Mattace-Raso, Tischa JM van der Cammen, Albert Hofman, et al. Arterial Stiffness and Risk of Coronary Heart Disease and Stroke: The Rotterdam Study. Epidemiology, Circulation. 2006; 113: 657-663). Thus, evaluating the propagation velocity of the pulse wave in the main arteries, it is possible to conclude that the morphological changes in the test vessel are severe and take preventive measures to prevent the patient from deteriorating.

A known method for estimating the propagation velocity of a pulse wave based on the registration of heart sounds with a microphone mounted over the region of the heart and a pulsogram (patent EP 1273262). To register the pulsogram, a photoelectric sensor is used, fixed on the finger of the subject. In this case, the pulse wave propagation speed is determined by the distance between the region of registration of heart sounds by the microphone and the photoelectric sensor and the time interval between the first heart tone recorded by the microphone and the leading edge of the pulse wave recorded by the photoplethysmograph. The disadvantage of this method is the need to use additional registration of heart sounds using a microphone.

Australian company AtCor Medical Pty. Ltd. A method for estimating the propagation velocity of a pulse wave, which is recorded using a plethysmographic cuff placed on the patient’s thigh, and a sphygmographic sensor detecting the pulse on the carotid artery, is proposed. In this case, the pulse wave propagation speed is determined by the distance between the pulse wave registration point by the sphygmograph on the carotid artery and the plethysmographic sensor located on the hip and the time interval between the leading edges of the pulse waves recorded by both sensors. The disadvantage of this method is the need to hold the operator's hand with a sphygmographic sensor over the area of the carotid artery.

Also known are devices for measuring the speed of a pulse wave using two microwave sensors superimposed on the surface area of the arteries. For example, it is proposed to use Doppler sensors to measure the propagation velocity of a pulse wave (patent EP 1921987). The device consists of two sensors, with which the pulse wave signals are recorded at two points in the body, after which the pulse wave propagation velocity is calculated. Each of the sensors emits a signal at a frequency from the range from 400 MHz to 5 GHz, and in the received signal evaluates the frequency change caused by the Doppler shift, which is caused by pulsation of the artery walls. The advantage of the device is the absence of the need for direct tight contact of the sensor with the patient’s body. However, accurate positioning of the sensor with respect to the areas of occurrence of arterial vessels and relative immobility of the test subject are required. The disadvantage of such devices is the influence of external noise on the measurement results. Known methods for increasing the noise immunity of such devices due to the use of asymmetric strip transmission lines as part of the sensors (patent RU 126257), and also due to the fact that each sensor is made in the form of an electrically conductive screen mounted on an electrically conductive grounded layer and forming a closed structure around the printing track of the worker sensor channel (patent RU 137720).

Known methods for determining the propagation velocity of the pulse wave, for the implementation of which are used plethysmographic sensors of various types, recording oscillatory changes in the volume of tissues adjacent to the sensor, which are caused by pulsations of arterial blood flow. In this case, the pulse wave propagation velocity can be estimated in several ways.

Using plethysmographic sensors, pulse wave signals for two sections of the body surface above the examined arterial arteries are recorded and the lag time (Δt) between the pulse waves corresponding to the two selected sections is estimated. Measure the distance between these sections (L). The propagation velocity of the pulse wave between the two sections under consideration is calculated as V = L / Δt. In the event that signals from plethysmographic sensors are recorded sequentially, to correctly determine the delay time Δt, an electrocardiogram signal recorded in parallel with the plethysmogram is used [US 6120456]. The propagation velocity of the pulse wave is determined by the distance between the ECG recording area and the plethysmographic sensor and the time interval between the R-wave of the ECG and the leading edge of the pulse wave detected by the plethysmographic sensor. The disadvantage of this method is the need to use additional registration of the ECG.

In the case when the registration of signals by two plethysmographic sensors is carried out in parallel, there is no need for an additional ECG sensor to synchronize the data on pulse waves recorded for two different parts of the subject's body. The propagation velocity of the pulse wave is determined by the distance between the two plethysmographic sensors and the delay time of the pulse waves, which is defined as the interval between the absolute positive extrema of the cardiac cycle graphs (patent RU 2511453). This method is the closest analogue (prototype), as it is based on the synchronous registration of pulse wave signals recorded in two different parts of the body of the subject, and does not require additional data synchronization. The disadvantage of the prototype is the contact nature of the information retrieval. In the case of using plethysmographic cuffs for registering pulse waves, it is necessary to select the cuff size taking into account the girth of the subject’s shoulder (or thigh), which also complicates the procedure and increases testing costs, since it requires additional measurements of the girth of the subject’s shoulder and the purchase of plethysmographic cuffs of several sizes.

3. The list of figures, drawings and other materials

FIG. 1 is a diagram of the implementation of the proposed method in assessing the propagation velocity of a pulse wave in a segment of the upper limb.

FIG. 2 is a diagram of an algorithm for estimating the propagation velocity of a pulse wave using a device that implements the method.

FIG. 3, 4, 5 - graphs of recorded signals.

4. The invention

4.1. A task

The technical problem is to eliminate this drawback, by simplifying the procedure for assessing the propagation speed and eliminating the influence of contact sensors on the recorded pulse wave profiles, namely: in the absence of the need to use any contact sensors that affect the underlying vessels, and their exact positioning relative to places of occurrence of the main arteries for which the pulse wave profile is recorded.

4.2. Features

The technical result is achieved in that, in contrast to the known method, which includes recording pulsograms from two (or more) pulsating sections of the body surface above the arteries under investigation, determining the distance L between these sections, the travel time Δt of the pulse wave between them and calculating the propagation velocity V a pulse wave according to the formula V = L / Δt, two (or more) pulsograms are simultaneously recorded from two (or more) pulsating parts of the body surface above the arteries being examined using one video camera, p recording video in the visible or infrared range of electromagnetic wavelengths, while the travel time of the pulse wave Δt is determined by the shift of the graphs of cardiocycles of two (or more) pulsograms.

A method characterized in that pulsograms for two (or more) parts of the human body are formed by processing a video sequence recorded using a video camera on which fragments corresponding to selected parts of the surface of the human body are highlighted. The pulsogram is formed by processing the video signal using the Euler algorithm, which consists in examining the time series of color values in any spatial position and amplifying the pixel color values in the frequency range corresponding to the frequency parameters of the pulse wave.

5. Information confirming the possibility of carrying out the invention

Images of the surface of body sections above the arteries lying close to the surface are recorded using a camera capturing a video image in the visible or infrared range. On the recorded video image, the user selects the areas of interest to him, corresponding to parts of the body surface above the main arteries. For example, in the case of estimating the propagation velocity of a pulse wave in the arteries of the upper limb, the characteristic sections of the video sequence allocated for further analysis are shown in FIG. 1. For the selected sections, the distance L between them is estimated. For the selected sections of the video sequence, the algorithm for estimating the pulse wave propagation velocity consists of the following steps (Fig. 2).

1. The decomposition of the video into components corresponding to different frequency ranges, by constructing the Laplace pyramid. At this stage, a spatial low-pass filter (Gaussian filter) is applied to frames from the video sequence and the image is decimated (construction of the Gauss pyramid); then, at each level of the Gauss pyramid, the blurred image is subtracted from the original one; the resulting residue is a new level of the Laplace pyramid.

2. Frequency filtering of movements in video signals (pixel-by-pixel temporal processing). Temporal processing is performed on each selected spatial range: a band-pass filter is used to extract a group of frequencies of interest (for the task of amplifying the vibrational movements of the body surface due to the pulse, frequencies are selected in the range 0.7-2.0 Hz, which corresponds to 42-120 beats per minute ) Temporal processing is the same for all spatial levels and for all pixels at each level. Then, the extracted signal is multiplied by a coefficient that regulates the amplification of the emitted vibrational movements. Next, the amplified signal is added to the original, after which the levels of the spatial pyramid are added to obtain the final result. Since natural video recordings are smooth in space and time, and filtering is performed uniformly in pixels, the method preserves the spatio-temporal smoothness of the video.

3. Reconstruction of the image of the pulse wave. At this stage of the algorithm, summing the values of pixel intensities for each of the selected areas of the video image is performed; then, the functional dependence of the luminous flux values of a given region in time is built.

4. Low-pass filtering of the obtained pulse wave realizations for selected sections of the body surface. For each of the obtained implementations of the pulse waves, a Fourier transform and a band-pass filter are used to select the main frequency response of the video stream corresponding to the fluctuation of the surface of the body over the artery. The result of this stage of the algorithm is to obtain a smooth pulse wave curve in a given section, corresponding to a change in the intensity of pixels in the video image over time.

5. The determination on the pulse wave curves of the positions of local maxima that correspond to the peak values of blood flow to areas of the body allocated for analysis.

6. Determination of the propagation delay time of the pulse wave Δt between two (or more) selected sections.

7. Calculation of the pulse wave propagation velocity V = L / Δt for all local maximums of pulse waves recorded during the time interval during which the video was recorded. Calculation of the average value of the pulse wave propagation velocity in the selected area and the standard deviation.

8. Display of results

An example implementation of the method

The subject was in a sitting position with his bare right hand from the shoulder joint to the hand, at a distance of 1 m from it was mounted on a tripod and aimed at the camera, which was not covered by clothing, recording the image in the visible range of wavelengths. The experimental design is shown in FIG. 1. The video image was recorded by the camera for 10-20 seconds, during this period of time, the test subject was not supposed to move.

In FIG. Figure 3 shows the pulse wave curve extracted using the proposed method before filtering for a portion of the body on the wrist above the radial artery. In FIG. 4 shows the corresponding curve of the pulse wave after filtering. In FIG. Figure 5 shows the pulse waves extracted using the proposed method for the sections on the wrist above the radial artery and the inner side of the elbow bend above the ulnar artery; for the selected pulse waves, the delay time was estimated as the interval between the nearest local maximums for the pulse wave curves for the selected sections (for the case of Fig. 5 Δt = 0.04 s). The measured distances between the selected sections for which pulse waves were recorded, L = 0.295 m. The propagation velocity of the pulse wave between the selected sections was V = L / Δt = 0.295 m / 0.04 s = 7.2 m / s.

Similarly, by recording with the camera an image of body sections above the main arteries of the lower extremities or the carotid artery, the propagation velocity of the pulse wave along the main arteries of the lower extremities and the aorta can be estimated.

The analysis carried out by the applicant according to the prior art, showed that the proposed invention, having novelty and industrial applicability, meets the requirements of the criterion of "inventive step" with respect to the totality of its essential features. The prior art also does not know the mechanism for achieving the technical result disclosed in the application materials.

Claims (3)

1. A method for determining the propagation velocity of a pulse wave of blood pressure, including recording pulsograms from two pulsating parts of the body surface above the arteries under investigation, determining the distance L between these sections, the travel time of the pulse wave Δt between them and calculating the velocity V of the pulse wave propagation using the formula V = L / Δt, characterized in that from two or more pulsating parts of the body surface above the examined arteries, two or more pulsograms are simultaneously recorded using the bottom of the video camera recording the video image in the visible or infrared range of electromagnetic wavelengths, while the travel time of the pulse wave Δt is determined by the shift of the graphs of cardiocycles of two or more pulsograms. 2. The method according to p. 1, characterized in that pulsograms of two or more parts of the human body are formed by processing a video sequence recorded using a video camera on which fragments corresponding to selected parts of the surface of the human body are highlighted. 3. The method according to claim 1, characterized in that the pulsogram is generated by processing the video signal using the Euler algorithm, which consists in examining the time series of color values in any spatial position and amplifying the pixel color values in the frequency range corresponding to the frequency parameters of the pulse wave.

Images