Predicting the biomechanical behaviour of the arterial wall through a temporal analysis of the oscillometric signal during pressure measurement

Abstract

The present invention relates to a method for determining indicator parameters indicative of the biomechanical behaviour of the arterial wall of a patient, on the basis of a temporal analysis of an oscillometric signal acquired during the deflation phase of the cuff around the arm of the patient, the indicator parameters being extracted from the presence of a first maximum and the possible presence of a second maximum detected on the time-evolution plot of the foot-to-apex time intervals (FATI) identified as belonging to the same cycle for each of the cardiac cycles of the oscillometric signal. The first FATI maximum corresponds to the systolic blood pressure. The second maximum may be in correlation with the modification of the biomechanical properties of the arterial wall, such as increased arterial stiffness or pathological behaviour of the wall.

Description

Description Title: Prediction of the biomechanical behavior of the arterial wall through the temporal analysis of the oscillometric signal during pressure measurement Technical field [1] The present invention relates to the field of blood pressure (BP) measurement and more specifically to the determination of parameters indicating the biomechanical behavior of an arterial wall through the temporal analysis of the oscillometric signal acquired during the deflation phase of the cuff installed on a subject. The present invention also relates to a method for screening and predicting the occurrence of pre-eclampsia by specific evaluation of the parameters indicating changes in the biomechanical behavior of the arterial wall associated with clinical parameters. Prior art [2] Non-invasive blood pressure measurement is an essential clinical routine examination to assess cardiovascular function and monitor its disturbances in a patient. [3] The clinical reference method is the auscultatory method, based on the detection of sounds produced by the brachial artery when the pressure of the pneumatic cuff is between the systolic pressure and the diastolic pressure. However, this manual technique is difficult to implement, particularly by non-medical or inexperienced personnel. [4] The auscultatory method is gradually being replaced by the oscillometric method, which is simpler to use, since it is automated, requiring no specific skills or the rigor required in the case of the auscultatory technique. Oscillometrics consists of detecting the pressure oscillations generated in the pneumatic cuff of the blood pressure monitor by the expansion of the arteries under the effect of the arterial wave (or "pulse wave") and measuring the amplitude relative to the different pressures of the cuff during its deflation. The envelope of the oscillometric curve representing the amplitude of these oscillations as a function of the cuff pressure is analyzed to identify, in particular, the pressure at which the maximum amplitude (peak of the oscillometric envelope) of the oscillations is observed, which corresponds to the mean arterial pressure (MAP).From this MAP pressure, various algorithms are then used to derive an estimate, rather than an actual measurement, of systolic and diastolic pressure. This results in significant differences between measurements taken with different devices, resulting in a lack of reproducibility and standardization, which can lead to errors in patient monitoring, raising the issue of reliability not only at the individual level, but also in clinical studies, particularly epidemiological ones. [5] Furthermore, the currently available measuring devices are therefore not always reliable for the exact determination of blood pressure measurement in certain populations. This is particularly the case for patients with arterial stiffness, including patients with severe renal failure or in elderly subjects. This is also the case in cases of arrhythmia. In these situations, The validity limits of standard oscillometric techniques are reached, this is mainly explained by the use of algorithms that analyze only the envelope of the oscillometric curve and are therefore dependent on any pathological deformation of this curve. The causes of deformations are frequent: age of the subject, arrhythmia such as atrial fibrillation, arterial stiffness modifying the compliance of the wall and the transmission of the pulse wave. In addition, pulmonary ventilation movements also affect the accuracy of the oscillometric results. Finally, the increase in pulse pressure corresponding to the difference between systolic and diastolic pressure amplifies the difference between the oscillometric measurement and the auscultatory measurement of systolic blood pressure.Differences between auscultatory and oscillometric blood pressure measurements can be clinically significant, particularly in patients with diabetes and / or increased arterial wall stiffness, resulting in, most often, overestimation of systolic and diastolic blood pressures. This can therefore lead to inappropriate or delayed management and clinical risk. [6] The inventors propose a new approach which consists of analyzing the oscillometric curve cycle by cycle, thus making it possible to overcome the limits linked to the shape of the oscillometric curve, and in particular to its variability according to populations. This approach allows a true measurement of the PAS and no longer an estimation as proposed by known devices. [7] The invention also relates to a new method for screening and predicting arterial pathologies linked to increased rigidity of the arterial wall by determining the parameters indicating the biomechanical behavior of the arterial wall through the temporal analysis of the oscillometric curve from cycle to cycle. [8] Current methods for determining arterial wall stiffness require a sometimes lengthy protocol, which is generally indirect. For example, the indirect reference method is known, which consists of determining the carotid-femoral pulse wave propagation velocity (PWVcf) (in m / sec) along the thoracoabdominal aorta, with recording of the arterial wave at the level of the common carotid and femoral arteries. This requires prior determination of the blood pressure measurement. Other indirect measurement methods have also been proposed, with recording of blood pressure over several cycles and estimation of stiffness using the delay between the ECG Q wave and the Korotkoff noise corresponding to the PAD. However, these techniques require simultaneous electrocardiogram (ECG) recording. [9] On the other hand, there is currently no reliable and effective method to screen for and predict the risk of pre-eclampsia (PE), linked to changes in the biomechanical behavior of the arterial wall that precedes the onset of hypertension. Currently, screening and risk prediction methods are based on various composite scores, integrating more or less completely, maternal history, certain biological markers and hemodynamic data of the uterine arteries. However, this approach, although effective, remains insufficiently applied to all pregnancies, particularly due to its cost and lack of accessibility. However, it is essential to predict the risk of PE occurring early so that appropriate management can be initiated as soon as possible, including in particular the introduction of aspirin treatment from the first trimester, which significantly reduces the risk of PE occurring.

[0010] The present invention aims to overcome the drawbacks of the prior art by proposing an apparatus and method, simple to implement while being reliable, which allows reliable and precise measurement of blood pressure.

[0011] Another object of the present invention is a diagnostic tool which allows early and reliable detection of changes in the biomechanical behavior of the arterial wall.

[0012] Another object of the present invention is a diagnostic tool which contributes to the early and reliable detection of pre-eclampsia. Summary

[0013] The present invention improves the situation.

[0014] A method is proposed for determining the parameters indicating the risk of the occurrence of pre-eclampsia and for screening for pre-eclampsia in a subject, through the temporal analysis of an acquired oscillometric signal (ACQ_SIG) during the deflation phase of the cuff installed on the subject, the method comprising the following steps: calculating the second derivative with respect to time of the oscillometric signal (CALC_DERIV_SEC); locating the peaks of the second derivative of the oscillometric signal by applying adaptive thresholding to eliminate non-significant peaks, said peaks corresponding to peaks having an amplitude greater than that of the previous point and that of the following point (LOC_PICS); identifying the peaks of negative values ​​corresponding to the peaks of each cycle of the oscillometric signal (IDENT_SOM); identifying the peaks of positive values ​​corresponding to the feet of each cycle of the oscillometric signal (IDENT_PIED);calculating the identified foot-to-peak delay (DPS) belonging to the same cycle for each of the cardiac cycles of the oscillometric signal and determining a time evolution of the DPS delays (CALC_DPS); extracting a first maximum of the DPS delay and a second maximum of the DPS delay from the time evolution of the DPS delays (EXT_MAX); calculating at least one parameter associated with the first maximum and the second maximum, said at least one parameter being an indicator of the biomechanical behavior of the arterial wall in the subject (CALCJND).;

[0015] According to one embodiment, the method may further comprise, after the steps of identifying peaks, a step of eliminating the peaks detected at a heart rate greater than 40 bpm compared to the previous frequency in order to eliminate peaks linked to the subject's movements.

[0016] According to another embodiment, the method may further comprise, after the step of calculating the DPS delay, a step of eliminating the pulses having a DPS delay twice that of the previous pulse and that of the following pulse so as to eliminate the pulses linked to the movement of the subject.

[0017] The features set out in the following paragraphs may, optionally, be implemented, independently of each other or in combination with each other:

[0018] The DPS delay can be defined as a maximum when it is greater than the previous DPS delays and greater than 5% of the average of a number n of subsequent DPS delays, n being an integer defined as a function of the heart rate and the cuff deflation speed.

[0019] The first maximum of the DPS delay corresponds to the systolic pressure.

[0020] Said at least one indicator parameter can be chosen from the parameters of the following list: amplitude of the first maximum; amplitude of the first maximum greater than 120 ms; presence of a second maximum; amplitude of the second maximum; value of the blood pressure when the second maximum occurs.

[0021] According to one embodiment, the method may further comprise a step of analyzing the amplitude of the first maximum of the DPS delay at a threshold value greater than 120 ms, to determine the risk of the occurrence of pre-eclampsia.

[0022] According to another embodiment, the method may further comprise a step of comparing the amplitude of the second maximum of the DPS delay to a threshold value corresponding to 10% above the average of two previous DPS vectors, to determine the risk of the occurrence of pre-eclampsia.

[0023] According to one embodiment, the method may comprise a step of comparing the pressure at which the second maximum of the DPS delay occurs to the mean arterial pressure, to determine the risk of the occurrence of pre-eclampsia.

[0024] The method may further comprise a step of calculating a score for predicting the risk of the occurrence of pre-eclampsia by combining at least one indicator parameter associated with the first maximum and / or the second parameter with at least one clinical parameter specific to the pregnant woman, and in which said at least one clinical parameter associated with the pregnant woman is chosen from the clinical parameters of the following list: history of pre-eclampsia, high blood pressure (HBP) when measuring the oscillometric signal.

[0025] Based on the indicator parameters determined by the method described above and the clinical parameters, the method may comprise a step of calculating the risk prediction score for the occurrence of pre-eclampsia according to one of the relationships chosen from the relationships in the following list: - Y1 = 2.931 (if history of PE) + 1.486 (if occurrence of a second maximum in mean arterial pressure); - Y2 = 2.725 (if history of PE) + 2.141 (if occurrence of a second maximum in mean arterial pressure); - Y3 = 3.457 (if PE antecedent) + 0.022 * amplitude in ms of the first maximum; - Y4 = 3.118 (if history of PE) + 0.017 (amplitude of the first maximum) + 1.644 (occurrence of a second maximum at mean arterial pressure); - Y5 = 2.526 (if history of PE) + 0.023*amplitude in ms of the first maximum + 1.590 (if taking aspirin during pregnancy); - Y6 = 2.182 (if history of PE) + 0.018*amplitude in ms of the first maximum + 1.857 (if occurrence of a second maximum at mean arterial pressure) + 1.808 (if taking aspirin during pregnancy); - Y7 = 0.092*value of blood pressure at the time of measurement of systolic blood pressure measurement (mmHg) + 0.621 (if amplitude of the first maximum > 120 ms); - Y8 = 3.152 (if PE history) + 0847 (if amplitude of the first maximum > 120 ms); - Y9 = 2.879 (if history of PE) + 1.399 (if amplitude of the first maximum > 120 ms) + 1.796 (occurrence of a second maximum at mean arterial pressure); - Y10 = 3.031 (if history of PE) + 1.731 (occurrence of a second maximum at mean arterial pressure) + 2.660 (if amplitude of the first maximum > 150 ms).

[0026] The invention also relates to an apparatus for determining the parameters indicating the risk of the occurrence of pre-eclampsia and for screening for pre-eclampsia in a subject, from a temporal analysis of an oscillometric signal acquired during the deflation phase of a pneumatic cuff (BRASS) installed around the subject's arm, the apparatus comprising: - a device for acquiring oscillometric signals, said device comprising a pressure sensor (CAP_P) configured to measure the blood pressure coming from the pneumatic cuff and a data processing module (MOD_TRAIT); - a calculation module (MOD_CALC) configured to receive signals from said acquisition device and to determine the parameters indicating the biomechanical behavior of the arterial wall by implementing the steps of the method as described above; - a display unit (AFFICH) to display the results obtained. Brief description of the drawings

[0027] Other features, details and advantages will become apparent upon reading the detailed description below, and upon analyzing the attached drawings, in which: Fig. 1

[0028] [Fig. 1] Figure 1 shows a functional diagram of a measuring device implementing the method for determining the parameters indicating the behavior of the arterial wall in a subject according to one embodiment. Fig. 2

[0029] [Fig. 2] Figure 2 represents the main steps of a method performed by the apparatus of Figure 1 for determining the parameters indicative of the behavior of the arterial wall in a subject according to one embodiment. Fig. 3

[0030] [Fig. 3] Figure 3 shows an oscillometric signal acquired during cuff deflation. Fig. 4

[0031] [Fig. 4] Figure 4 shows the characteristic points of an oscillometric pulsation. Fig. 5

[0032] [Fig. 5] Figure 5 represents on the upper part of the diagram three possible forms of an oscillometric pulsation and on the lower part the maxima of the second derivative corresponding to the three forms of oscillometric pulsation. Fig. 6

[0033] [Fig. 6] Figure 6 represents the temporal evolution of the value of the DPS delay during the deflation of the pneumatic cuff and the projection of the vector of the DPS delays on the Korotkoff noises which allows to show that the first maximum of the DPS delay coincides with the first Korotkoff noise. Fig. 7

[0034] [Fig. 7] Figure 7 shows a ROC (Receiver Operating Characteristic) curve plotted according to a score for predicting the risk of developing preeclampsia according to the following relationship: Y1 = 2.931 (if history of PE) + 1.486 (if presence of a second maximum), the area under the ROC curve or AUC (Area Under the Curve) being equal to 0.809. Fig. 8

[0035] [Fig. 8] Figure 8 shows a ROC curve plotted according to a risk prediction score for the occurrence of pre-eclampsia according to the following relationship: Y2 = 2.725 (if history of PE) + 2.141 (if occurrence of a second maximum in mean arterial pressure (MAP)), the AUC being equal to 0.865. Fig. 9

[0036] [Fig. 9] Figure 9 shows a ROC curve plotted according to a risk prediction score for the occurrence of pre-eclampsia according to the following relationship: Y3 = 3.457 (if history of PE) + 0.022*amplitude of the first maximum in ms, the AUC being equal to 0.832. Fig. 10

[0037] [Fig. 10] Figure 10 shows a ROC curve plotted according to a risk prediction score for the occurrence of pre-eclampsia according to the following relationship: Y4 = 3.118 (if history of PE) + 0.017*amplitude of the first maximum in ms + 1.644 (if occurrence of a second maximum at mean arterial pressure (MAP)), the AUC being equal to 0.925. Fig. 11

[0038] [Fig. 11] Figure 11 shows a ROC curve plotted according to a score for predicting the risk of the occurrence of pre-eclampsia according to the following relationship: Y5 = 2.526 (if history of PE) + 0.023*amplitude of the first maximum in ms +1.590 (if aspirin during pregnancy), the AUC being equal to 0.825. Fig. 12

[0039] [Fig. 12] Figure 12 shows a ROC curve plotted according to a score for predicting the risk of the occurrence of pre-eclampsia according to the following relationship: Y6 = 2.182 (if history of PE) + 0.018 * amplitude of the first maximum in ms + 1.857 (if occurrence of the second maximum at MAP) + 1.808 (if aspirin during pregnancy, the AUC being equal to 0.920. Fig. 13

[0040] [Fig. 13] Figure 13 shows a ROC curve plotted according to a score for predicting the risk of developing preeclampsia according to the following relationship: Y7 = 0.092 * PAS value + 0.621 (if amplitude of the first maximum > 120 ms), the AUC being equal to 0.923. Fig. 14

[0041] [Fig. 14] Figure 14 shows a ROC curve plotted according to a risk prediction score for the occurrence of pre-eclampsia according to the following relationship: Y8 = 3.152 (if history of PE) + 0.847 (if amplitude of the first maximum > 120 ms), the AUC being equal to 0.811. Fig. 15

[0042] [Fig. 15] Figure 15 shows a ROC curve plotted according to a score for predicting the risk of the occurrence of pre-eclampsia according to the following relationship: Y9 = 2.879 (if history of PE) + 1.399 (if amplitude of the first maximum > 120 ms + 1.796 (if occurrence of the second maximum at MAP), the AUC being equal to 0.900. Fig. 16

[0043] [Fig. 16] Figure 16 shows a ROC curve plotted according to a risk prediction score for the occurrence of pre-eclampsia according to the following relationship: Y10 = 3.031 (if history of PE) + 1.731 (if occurrence of the second maximum at MAP + 2.660 (if amplitude of the first maximum > 150 ms), the AUC being equal to 0.928. Description of the embodiments

[0044] Reference is now made to Figure 1 which represents a functional diagram of an apparatus 10 according to an embodiment for determining the parameters indicating the behavior of the arterial wall of a subject 11, through the temporal analysis of an oscillometric signal acquired during the deflation phase of a cuff 12 (BRASS) installed on the subject.

[0045] The cuff 12 is placed at the brachial level and is inflated by a mini-compressor, and its deflation is controlled by a solenoid valve. The mini-compressor and the solenoid valve are not shown in Figure 1.

[0046] The apparatus 10 comprises an oscillometric signal acquisition device 13 configured to measure the pressure prevailing in the cuff and generate an oscillometric signal representative of the pressure variations and a calculation module (MOD_CALC) 17 configured to implement the steps of the method for determining the parameters indicating the biomechanical behavior of the arterial wall through the temporal analysis of the oscillometric signal according to one embodiment.

[0047] The acquisition device 13 comprises a pressure sensor (CAP_P) 14 configured to detect the pressure in the cuff. For example, this sensor is a piezoelectric sensor and a data processing module (MOD_TRAIT) 15 configured to extract the oscillometric signal from the signal from the pressure sensor.

[0048] The acquisition device 13 further comprises a wireless or wired transmission module (MOD_TRANS) 16 configured to transmit the acquired data to the calculation module 17.

[0049] The calculation module 17 also comprises a reception module (MOD_REC) 18 configured to receive the processed data coming from the acquisition device 13.

[0050] The apparatus comprises a display unit 19 for displaying the results from the calculation module. For example, it can display the oscillometric curves, the heart rate, the value of the systolic pressure obtained from the temporal analysis of the oscillometric curve, the parameters indicating the biomechanical behavior of the patient's arterial wall following the temporal analysis of the oscillometric signal. The display unit can thus display, for example, the amplitude of the first maximum, the presence of a second maximum, the amplitude of the second maximum, the value of the average arterial pressure when the second maximum occurs.

[0051] Figure 2 represents the main steps of a method 100 for determining the parameters indicating the biomechanical behavior of the arterial wall through the temporal analysis of the oscillometric signal. The method is executed by the calculation module 17 of the apparatus 10 of Figure 1.

[0052] Advantageously, all the steps of the process are automated.

[0053] The method 100 comprises a step 101 of acquiring an oscillometric signal (ACQ_SIG) acquired during the deflation phase of the cuff installed on the subject. The oscillometric signal presented in the form of a curve consisting of a succession of oscillations corresponding to cardiac cycles. Figure 3 illustrates an example of an oscillometric signal 200 with time on the abscissa and pressure on the ordinate.

[0054] Figure 4 schematically represents a possible form of oscillometric pulsation or cardiac cycle 301, on which the characteristic points are marked which are the foot 302, the summit 301 and the notch 303. On this form, it is noted that the notch 303 occurs after the summit 301. One of the steps of the method therefore consists of locating these characteristic points (LOC_PICS) on the oscillometric signal to determine the delay between the foot 302 and the summit 301 (DPS) which therefore corresponds to the time elapsing between the foot and the summit.

[0055] The method comprises a step 102 of calculating the second derivative (CALC_DERIV_SEC) of the oscillometric signal with respect to time to enable locating the peaks which correspond to the peaks of its negative values ​​and the feet which correspond to the peaks of its positive values. This step consists first of all in calculating the first derivative of the oscillometric signal with respect to time and then applying a low-pass filter at 5 Hz of the Butterworth type of order 2 to the signal resulting from the first derivative to obtain the derivative of the filtered signal which represents the second derivative of the oscillometric signal. The signal of the second derivative is in the form of peaks of positive and negative values ​​which correspond to the peaks, feet and notches.

[0056] In Figure 5, the upper part represents in addition to the oscillometric pulsation shape of Figure 4, two other possible shapes of oscillometric pulsation 400, 500. In the second shape, the notch 403 occurs between the foot 402 and the apex 403. In the third shape, the oscillometric pulsation comprises only the foot 502 and the apex 501. The location of the apexes and feet must take into account the shape of the oscillometric pulsations to actually locate the foot and the apex in order to determine the DPS.

[0057] The characteristic points are obtained by calculating the second derivative with respect to time of the oscillometric signal. In Figure 5, the lower part represents the second derivative corresponding to the three forms of oscillometric pulsation. The three characteristic points present on the cardiac cycle are transformed into positive and negative peaks. We note in Figure 5 that the negative and positive peaks which correspond to the notches have a lower amplitude than the peaks which correspond to the feet and the peaks. It is therefore necessary to locate the peaks which actually correspond to the peaks and feet to calculate the DPS and not to retain the peaks which correspond to the notches.

[0058] After the step of calculating the second derivative, the method comprises a step 103 of locating the pulsation peaks (LOC_PICS) of the second derivative by applying adaptive thresholding to eliminate non-significant peaks. For example, adaptive thresholding between 21% of the maximum peak and the average of the two following peaks can be adopted to eliminate non-significant peaks. The localization of negative and positive peaks is carried out according to the following definition: the localized peaks which are the characteristic points correspond to peaks having an amplitude greater than that of the previous point and that of the following point. The other peaks correspond to noise and will not be taken into account in the calculation of the DPS.

[0059] The method comprises a step 104 of locating the peaks (IDENT_SOM) among the peaks of negative values ​​located in step 103. As indicated above, the negative peaks contain the peaks which correspond to the peaks and possibly to the notches according to the shape of the oscillometric pulse. The peaks indicating the peaks have amplitudes greater than those of the other negative peaks. Thus, an adaptive threshold at 80% of the average of the negative peaks is adopted to locate the negative peaks indicating the peaks by retaining only the negative peaks lower than the calculated threshold. The negative peaks corresponding to the peaks therefore have an amplitude greater than the threshold in absolute value.

[0060] Referring to Figure 5, the second derivative 400D of the second form has two peaks of negative values. The negative peak 401 D which has a greater amplitude than the first negative peak corresponds to the peak 401 of the cycle.

[0061] According to one embodiment, the method may further comprise, after the peak identification steps, a step in which the peaks detected at a heart rate greater than 40 bpm compared to the previous frequency are eliminated so as to eliminate the peaks linked to the movement of the subject. Indeed, on the vector containing the peaks of the oscillometric signal, it is considered that the variability of the frequency cannot be greater than 40 bpm and any peak detected at a heart rate greater than 40 bpm compared to the previous frequency implies the elimination of this peak which is linked to the movement.

[0062] The method then comprises a step 105 of identifying the positive peaks (IDENT_PIED) corresponding to the feet among the positive peaks located in step 103 for each cycle. In this step, the cardiac cycles framed by the peaks located in the previous step are analyzed to determine the foot of each beat, which makes it possible to distinguish one of the three possible forms of oscillometric beats illustrated in Figure 5. In other words, the positive peaks are located on the signal of the second derivative between two consecutive peaks identified in step 104. In the case where two positive peaks are located, the maximum of these peaks corresponds to the foot of the beat.

[0063] Referring to Figure 5, the second derivative of the second form has two peaks of positive values ​​which are located between two peaks and the maximum of these two peaks corresponds to the foot of the pulsation.

[0064] The method comprises a step 106 of calculating the delay between the foot and the summit (DPS) belonging to the same cycle for each of the cycles (CALC_DPS). This is the time elapsing between the foot and the summit located on the second derivative as illustrated in Figure 5.

[0065] According to one embodiment, the method further comprises, after the step of calculating the DPS, a step of eliminating the pulsations having a DPS twice that of the previous pulsation and that of the following pulsation so as to eliminate the pulsations linked to the movement of the subject. Indeed, the deformed pulsation presents not only an increase in amplitude which does not preserve the shape of the oscillometric signal but also an increase in the DPS delay. Also, an adaptive thresholding of the DPS delay was applied to the DPS delays calculated in step 106 keeping only the pulses having a delay half that of the previous pulse and that of the following pulse.

[0066] In step 106, the delays being calculated for all the cycles of the oscillometric signal, it is therefore possible to determine a temporal evolution of the DPS and to plot the successive values ​​of DPS.

[0067] Figure 6 shows the plot or vector of DPS 600 delays in an elderly subject, the evolution of which over time allows several variations to be highlighted.

[0068] The method comprises a step 107 of extracting a first maximum of the DPS delay and a second maximum of the DPS delay from the plot established in the previous step (EXT_MAX).

[0069] According to one embodiment, a DPS delay is defined as a maximum when it is greater than the previous DPS delays and greater than 5% of the average of a number n of subsequent DPS delays, n being an integer defined as a function of the heart rate and the cuff deflation speed. Preferably, n is between 1 and 3. For example, if the number of cycles is greater than or equal to 40, n = 3, if the number of cycles is between 10 and 40, n = 2, if the number of cycles is equal to or less than 10, n = 1. In Figure 6, the DPS plot comprises a first maximum 601 and a second maximum 602.

[0070] Advantageously, the inventors note that this first maximum of the DPS delay occurs precisely when the cuff pressure is equal to the systolic pressure and is perfectly concomitant with the first Korotkoff sound.

[0071] In Figure 6, the DPS delay vector is superimposed on the Korotkoff noises 700 recorded at the same time as the oscillometric signal in a subject. The first increase in the DPS delay 601 coincides exactly with the first Korotkoff noise 701 .

[0072] Thus, the comparison of the DPS delay calculation method of the present invention for measuring systolic blood pressure with Korotkoff noises shows a concordance. The DPS delay calculation technique therefore allows a direct measurement of systolic blood pressure from the oscillometric signal, thus allowing a more reliable measurement of systolic pressure. This technique of temporal analysis of the oscillometric signal can be easily implemented on existing oscillometric devices in addition to current analysis methods which determine the average pressure.

[0073] Advantageously, by applying the DPS technique to a sample of populations, the inventors observe that there is a correlation between the maximum DPS and elderly subjects and / or those with cardiovascular risk factors or pathologies. They observe in particular that in elderly subjects and / or those with cardiovascular risk factors, the DPS plot contains a second maximum, appearing in the form of a sudden and transient increase as illustrated in Figure 6. In young subjects, without cardiovascular risk factors or pathologies, the DPS plot contains a single maximum which coincides with the first Korotkoff noise.

[0074] The method comprises a step 108 of extracting at least one parameter indicating the biomechanical behavior of the arterial wall in the subject, associated with the first maximum and the second maximum (CALCJND).

[0075] The analysis of the biomechanical behavior of the arterial wall can be carried out using parameters chosen from the list indicated below.

[0076] This indicator parameter can be the amplitude of the first maximum, the amplitude of the first maximum greater than 120 ms, the presence of a second maximum, the amplitude of the second maximum and the value of the blood pressure when the second maximum occurs.

[0077] According to an exemplary embodiment, the indicator parameter may also be the amplitude of the first maximum in nominal value relative to the second maximum when the latter is present.

[0078] The inventors note that the DPS delay is directly impacted by the shape of each oscillation, itself determined by the biomechanical properties of the arterial wall and by the more or less rapid passage of the incident and reflected arterial waves. Therefore, it would be interesting to exploit this characteristic to have indicator parameters associated with the transmission of the wave, and therefore associated with the biomechanical properties of the wall. Indeed, in certain populations, the behavior of the arterial wall reflects a particular risk, for example, the increased cardiovascular risk in the case of arterial stiffness in the general population. In addition, in pregnant women at risk of pre-eclampsia (PE), changes in arterial function have been described, secondary to defective implantation of the placenta, with in particular generalized endothelial dysfunction and arterial stiffness, appearing before arterial hypertension.Thus, such indicator parameters could be used to screen or predict the risk of PE occurrence.

[0079] The invention relates to a method for screening for arterial wall stiffness, associated with cardiovascular risk in a subject, comprising determining at least one parameter associated with the first maximum and the second maximum and indicative of the behavior of the arterial wall according to the invention.

[0080] The invention also relates to a method for determining the risk of the occurrence of preeclampsia and for screening for preeclampsia comprising the determination of at least one parameter associated with the first maximum and the second maximum and indicative of the behavior of the arterial wall according to the invention.

[0081] The inventors conducted a study of arterial function in pregnant women, with or without blood pressure abnormalities, using the method of the invention. This study allowed them to identify indicator parameters associated with the occurrence of preeclampsia (PE) and to establish a list of scores associated with the risk of PE. This study was conducted at the Nîmes University Hospital, with pressure measurement and acquisition of the oscillometric curve by the method, in 132 women with normal or pathological pregnancies, seen in consultation or hospitalization, at various gestational ages. Monitoring of the pregnancy and delivery made it possible to identify eleven of them pre-eclampsia. Statistical analysis of clinical data and indices provided by the method identified markers associated with the occurrence of PE. Multivariate analysis of these indicator parameters made it possible to construct predictive scores, some of which proved to be very effective.

[0082] According to one embodiment, the method comprises a step of analyzing the amplitude of the first maximum of the DPS delay at a threshold value greater than 120 ms, to determine the risk of the occurrence of pre-eclampsia in a pregnant woman.

[0083] Alternatively, the method comprises a step of analyzing the amplitude of the first maximum of the DPS delay at a threshold value set at 150 ms, to determine the risk of the occurrence of pre-eclampsia.

[0084] Indeed, in a sample of 132 women, more than 9% of whom developed PE, a high amplitude of the first maximum was observed in those with a blood pressure abnormality. An analysis of all the DPS tracings obtained in the 132 patients confirms an association between crossing a threshold value equal to or greater than 120 ms and the occurrence of PE. This association is all the stronger when the amplitude is high.

[0085] According to another embodiment, the method further comprises a step of comparing the amplitude of the second maximum of the DPS delay to a threshold value corresponding to 10% above the average of two previous DPS vectors, to determine the risk of the occurrence of pre-eclampsia.

[0086] According to yet another embodiment, the method comprises a step of comparing the pressure at which the second maximum of the DPS delay occurs to the mean arterial pressure, to determine the risk of the occurrence of pre-eclampsia.

[0087] In the same study, the inventors found that women who developed PE had a modified oscillometric curve with the presence of a second DPS maximum.

[0088] Analysis of the DPS plots showed that the second maximum is significantly associated with the occurrence of pre-eclampsia when it is early, that is to say when it occurs for pressures close to the mean pressure.

[0089] This second maximum is identified when its amplitude is more than 10% greater than the average of the two previous vectors. Once this second maximum is identified, only the maxima occurring in the region of the average BP are associated with the occurrence of pre-eclampsia.

[0090] According to one embodiment, in order to analyze the performance of the prediction and screening model, the method comprises a step of calculating a score for predicting the risk of the occurrence of pre-eclampsia by combining at least one indicator parameter associated with the first maximum and / or the second parameter with at least one clinical parameter specific to the pregnant woman, from among the clinical parameters in the following list: history of pre-eclampsia, history of high blood pressure (HBP), HBP during measurement of the oscillometric signal.

[0091] Advantageously, the method comprises a step of calculating the score for predicting the risk of the occurrence of pre-eclampsia. The score can be defined according to one of the relationships below: below. These scores are the result of a statistical analysis made from data collected from a sample of 132 pregnant women.

[0092] Figures 7 to 16 respectively represent the ROC curve of each calculated score. The ROC (Receiver Operating Characteristic) curve represents on the abscissa the sensitivity as a function of 1 - specificity on the ordinate. For each curve, the area under the ROC curve (or Area Under the Curve, AUC) was calculated.

[0093] Figure 7 represents a ROC curve determined according to a score Y1 = 2.931 (if history of PE) + 1.486 (if occurrence of a second maximum in mean arterial pressure). The AUC is equal to 0.809.

[0094] Figure 8 represents a ROC curve determined according to a score Y2 = 2.725 (if history of PE) + 2.141 (if occurrence of a second maximum in mean arterial pressure). The AUC is equal to 0.865.

[0095] Figure 9 represents a ROC curve determined according to a score Y3 = 3.457 (if PE history) + 0.022 * amplitude in ms of the first maximum. The AUC is equal to 0.832.

[0096] Figure 10 represents a ROC curve determined according to a score Y4 = 3.118 (if history of PE) + 0.017 (amplitude of the first maximum) + 1.644 (occurrence of a second maximum at mean arterial pressure). The AUC is equal to 0.925.

[0097] Figure 11 represents a ROC curve determined according to a score Y5 = 2.526 (if history of PE) + 0.023*amplitude in ms of the first maximum) + 1.590 (if taking aspirin during pregnancy). The AUC is equal to 0.825.

[0098] Figure 12 represents a ROC curve determined according to a score Y6 = 2.182 (if history of PE) + 0.018*amplitude in ms of the first maximum + 1.857 (if occurrence of a second maximum at mean arterial pressure) + 1.808 (if taking aspirin during pregnancy). The AUC is equal to 0.920.

[0099] Figure 13 represents a ROC curve determined according to a score Y7 = 0.092*blood pressure value at the time of measurement (mmHg) + 0.621 (if amplitude of the first maximum > 120 ms). The AUC is equal to 0.923.

[0100] Figure 14 represents a ROC curve determined according to a score Y8 = 3.152 (if antecedent PE) + 0.847 (if amplitude of the first maximum > 120 ms). The AUC is equal to 0.811.

[0101] Figure 15 represents a ROC curve determined according to a score Y9 = 2.879 (if history of PE) + 1.399 (if amplitude of the first maximum > 120 ms) + 1.796 (occurrence of a second maximum at mean arterial pressure). The AUC is equal to 0.900.

[0102] Figure 16 represents a ROC curve determined according to a score Y10 = 3.031 (if history of PE) + 1.731 (occurrence of a second maximum at mean arterial pressure) + 2.660 (if amplitude of the first maximum > 150 ms). The AUC is equal to 0.928.

[0103] Among the ten established scores, the Y4 and Y10 scores, which combine the amplitude of the first maximum, the early occurrence of the second maximum and the clinical parameter of PE history, are particularly effective in predicting the occurrence of PE. Indeed, for Y4, the area under the ROC curve is 0.925 and for Y7 the calculated AUC is 0.928. These scores demonstrate that the DPS technique allows early and reliable detection of pre-eclampsia, so that appropriate management can be initiated as soon as possible.

Claims

Claims

1. Method (100) for determining the parameters indicating the risk of the occurrence of pre-eclampsia and for screening for pre-eclampsia in a subject, from a temporal analysis of an oscillometric signal acquired (101) during the deflation phase of the cuff installed on the subject, the method comprising the following steps: - calculate (102) the second derivative with respect to time of the oscillometric signal (CALC_DERIV_SEC); - locating (103) the peaks of the second derivative of the oscillometric signal by applying adaptive thresholding to eliminate non-significant peaks, said peaks corresponding to peaks having an amplitude greater than that of the previous point and that of the following point (LOC_PICS); - identify (104) the peaks of negative values ​​corresponding to the peaks of each cycle of the oscillometric signal (IDENT_SOM); - identify (105) the peaks of positive values ​​corresponding to the feet of each cycle of the oscillometric signal (IDENT_PIED); - calculate (106) the delay between the foot and the summit (DPS) identified belonging to the same cycle for each of the cardiac cycles of the oscillometric signal and determine a temporal evolution of the DPS delays (CALC_DPS); - extract (107) a first maximum of the DPS delay and a second maximum of the DPS delay from the temporal evolution of the DPS delays (EXT_MAX); - calculating (108) at least one parameter associated with the first maximum and the second maximum, said at least one parameter being an indicator of the biomechanical behavior of the arterial wall in the subject (CALCJND).

2. The method of claim 1, further comprising after the steps of identifying peaks (104, 105), a step of eliminating peaks detected at a heart rate greater than 40 bpm relative to the previous rate so as to eliminate peaks related to the subject's movement.

3. The method of claim 1 or 2, further comprising after the step of calculating the DPS delay, a step of eliminating pulses having a DPS delay twice that of the previous pulse and that of the next pulse so as to eliminate pulses related to the movement of the subject.

4. Method according to one of claims 1 to 3, in which a DPS delay is defined as a maximum when it is greater than the previous DPS delays and greater than 5% of the average of a number n of following DPS delays, n being an integer defined as a function of the heart rate and the speed of deflation of the cuff.

5. Method according to one of claims 1 to 4, in which the first maximum of the DPS delay corresponds to the systolic pressure.

6. Method according to one of claims 1 to 5, in which said at least one indicator parameter is chosen from the parameters of the following list: - amplitude of the first maximum; - amplitude of the first maximum greater than 120 ms; - presence of a second maximum; - amplitude of the second maximum; - blood pressure value at the time of the second maximum.

7. Method according to one of claims 1 to 6, in which the method further comprises a step of analyzing the amplitude of the first maximum of the DPS delay at a threshold value greater than 120 ms, to determine the risk of the occurrence of pre-eclampsia.

8. A method according to one of claims 1 to 6, wherein the method further comprises a step of comparing the amplitude of the second maximum of the DPS delay to a threshold value corresponding to 10% above the average of two previous DPS vectors, to determine the risk of the occurrence of pre-eclampsia.

9. A method according to any one of claims 1 to 6, wherein the method comprises a step of comparing the pressure at which the second maximum of the DPS delay occurs to the mean arterial pressure, to determine the risk of the occurrence of pre-eclampsia.

10. Method according to one of claims 1 to 9, further comprising a step of calculating a score for predicting the risk of the occurrence of pre-eclampsia by combining at least one indicator parameter associated with the first maximum and / or the second parameter with at least one clinical parameter specific to the pregnant woman, and in which said at least one clinical parameter associated with the pregnant woman is chosen from the clinical parameters of the following list: history of pre-eclampsia, high blood pressure (HBP) during the measurement of the oscillometric signal.

11. A method according to claim 10, comprising a step of calculating the risk prediction score for the occurrence of pre-eclampsia according to one of the relationships selected from the relationships in the following list: - Y1 = 2.931 (if history of PE) + 1.486 (if occurrence of a second maximum in mean arterial pressure); - Y2 = 2.725 (if history of PE) + 2.141 (if occurrence of a second maximum in mean arterial pressure); - Y3 = 3.457 (if PE antecedent) + 0.022 * amplitude in ms of the first maximum; - Y4 = 3.118 (if history of PE) + 0.017 (amplitude of the first maximum) + 1.644 (occurrence of a second maximum at mean arterial pressure); - Y5 = 2.526 (if history of PE) + 0.023*amplitude in ms of the first maximum) + 1.590 (if taking aspirin during pregnancy); - Y6 = 2.182 (if history of PE) + 0.018*amplitude in ms of the first maximum + 1.857 (if occurrence of a second maximum at mean arterial pressure) + 1.808 (if taking aspirin during pregnancy); - Y7 = 0.092*value of blood pressure at the time of measurement of systolic blood pressure measurement (mmHg) + 0.621 (if amplitude of the first maximum > 120 ms); - Y8 = 3.152 (if PE history) + 0847 (if amplitude of the first maximum > 120 ms); - Y9 = 2.879 (if history of PE) + 1.399 (if amplitude of the first maximum > 120 ms) + 1.796 (occurrence of a second maximum at mean arterial pressure); - Y10 = 3.031 (if history of PE) + 1.731 (occurrence of a second maximum at mean arterial pressure) + 2.660 (if amplitude of the first maximum > 150 ms).

12. Apparatus (10) for determining the parameters indicating the risk of the occurrence of pre-eclampsia and screening for pre-eclampsia in a subject (11), from a temporal analysis of an oscillometric signal acquired during the deflation phase of a pneumatic cuff (BRASS) (12) installed around the subject's arm, the apparatus comprising: - an oscillometric signal acquisition device (13), said device comprising a pressure sensor (CAP_P) (14) configured to measure the blood pressure coming from the pneumatic cuff and a data processing module (MOD_TRAIT) (15); - a calculation module (MOD_CALCUL) (17) configured to receive signals from said acquisition device (13) and to determine the parameters indicating the biomechanical behavior of the arterial wall by implementing the steps of the method according to one of claims 1 to 11; - a display unit (AFFICH) (17) to display the results obtained.

Images