Signal processing method for determining a respiratory parameter

Abstract

Title: Signal processing method for determining a respiratory parameter. Signal processing method in which a respiratory parameter (32) is determined as a function of an accelerometer signal (18) acquired on three distinct axes and an electrocardiogram signal (10), these signals being intended to be processed by signal processing and computing means (8, 9) to define respectively cardiorespiratory data (20, 22, 24) and cardiac electrical activity data (12, 14), these data being combined in a step of determining at least one respiratory parameter (32), said method being characterized in that a quality parameter (16, 26) is determined by analysis of the electrocardiogram signal (10) and / or by analysis of the accelerometer signal (18),The data defined on the basis of said electrocardiogram signal and said accelerometer signal respectively are considered in the step of determining at least one respiratory parameter (32) only if a value of the associated quality parameter (16, 26) is greater than a defined threshold value (34, 42). Figure For the Summary: Figure 2,

Description

Title of the invention: Method for processing signals to determine a respiratory parameter

[0001] The present invention relates to the field of medical devices and systems for monitoring the cardiac health of a living being. The present invention relates more particularly to a signal processing method for determining a respiratory parameter.

[0002] Heart failure is a chronic condition affecting a large part of the population, particularly those over 60, and this condition can in particular generate episodes of decompensation, the frequency of which increases with the deterioration of cardiac function.

[0003] Episodes of decompensation of heart failure are often accompanied by emergency hospitalization of the patient suffering from this heart condition because they are not detected sufficiently early, the episode of decompensation of heart failure regularly appearing asymptomatic at its beginnings. This absence of symptoms detectable by the patient implies that the episode of decompensation of heart failure evolves without any preventive action being taken. However, as soon as the patient begins to feel the first symptoms of decompensation of heart failure, such as fatigue, palpitations or even shortness of breath, hospitalization is difficult to avoid.

[0004] It is understood from the above that the earlier the detection of the episode of decompensation of heart failure occurs, the lower the risk of hospitalization, and the costs that it can generate, is, the early detection of decompensation of heart failure in the patient making it possible to intervene on the latter by prescribing drug treatment or by reminding him of the hygiene and dietary rules making it possible to stabilize the patient's cardiac function.

[0005] Early detection of decompensation of heart failure can be based in particular on an in-depth analysis of various subclinical cardiorespiratory parameters.

[0006] In this context, it is interesting to consider the patient's breathing, which is important information to process because heart failure has an effect on breathing. For this, data acquired by implants, for example electrocardiogram data or cardiorespiratory data, can be used to deduce information on breathing, and in particular information on the respiratory wave profile and / or on the number of respiratory cycles per minute.

[0007] For example, an individual in a normal situation experiences approximately 6 to 18 respiratory cycles per minute, whereas 20 to 30 respiratory cycles per minute may occur when an individual is having difficulty breathing.

[0008] The present invention falls within this context and proposes to provide a signal processing device which makes it possible to recover reliable information on the respiratory wave profile as well as on the number of cycles per minute.

[0009] Thus, the main object of the present invention is a signal processing method in which a respiratory parameter is determined as a function of an accelerometric signal acquired on three distinct axes and of an electrocardiogram signal, these signals being intended to be processed by signal processing means to respectively define cardiorespiratory data and cardiac electrical activity data, these data being combined in a step of determining at least one respiratory parameter, said method being characterized in that a quality parameter is determined by analysis of the electrocardiogram signal and / or by analysis of the accelerometric signal,the data defined on the basis of said electrocardiogram signal and said accelerometric signal respectively being considered in the step of determining at least one respiratory parameter only if a value of the associated quality parameter is greater than a defined threshold value.

[0010] The at least one respiratory parameter may in particular be a number of breaths per minute and / or a respiratory wave profile.

[0011] The signal processing method according to the invention is particularly advantageous in that it allows a fusion of known data for the calculation of a respiratory parameter, on the basis of signals conventionally obtained by an implant or a non-invasive acquisition device, and in that it introduces the analysis of at least one quality parameter to decide whether to take into account such or such data among these known data.This makes it possible to provide a signal processing method that is particularly reliable since there is no risk of introducing unreliable data into the data fusion, particularly robust since it implements signal acquisition techniques that are proven and conventionally used, and particularly economical since it can make it possible to avoid performing calculations to determine specific data whose fusion with other data could result in the determination of a respiratory parameter that is less accurate than if it had been determined without taking these specific data into account.

[0012] According to an optional characteristic of the invention, a first cardiac electrical activity data item resulting from the processing of the electrocardiogram signal is respiratory sinusoidal arrhythmia data item calculated on the basis of intervals between consecutive RR peaks of the electrocardiogram signal.

[0013] Respiratory sinusoidal arrhythmia data, or RSA data, is in other words terms calculated by averaging, for a given cycle, the difference from near to near between the R peaks, visible on the electrocardiogram signal in each QRS complex.

[0014] According to an optional characteristic of the invention, a second cardiac electrical activity data item resulting from the processing of the electrocardiogram signal is respiration data item derived from the electrocardiogram (ECG) calculated on the basis of amplitudes of the RR Peaks of the electrocardiogram signal.

[0015] The ECG-derived respiration data, or EDR data, is in other words calculated by averaging, for a given cycle, the amplitude of each of the R peaks, visible on the electrocardiogram signal in each QRS complex.

[0016] According to an optional characteristic of the invention, a first cardiorespiratory data item resulting from the processing of the accelerometric signal is an ADR type data item for which the characteristic value is a derivative of the accelerometric signal on the three axes.

[0017] According to an optional characteristic of the invention, a second cardiorespiratory data item resulting from the processing of the accelerometric signal is an MDR type data item, for which the characteristic value is an amplitude of the accelerometric signal on the three axes.

[0018] According to an optional characteristic of the invention, a third cardiorespiratory data item resulting from the processing of the accelerometric signal is PCR type data, for which the characteristic value is the value, among the representative values ​​along each of the three axes, which has the greatest amplitude.

[0019] According to an optional characteristic of the invention, said cardiorespiratory data of the ADR, MDR and PCR type are obtained by filtering the accelerometric signal over a frequency range from 0 to 1 Hz, then by calculating energy expenditure and adapting the filter as a function of the energy expenditure.

[0020] According to an optional characteristic of the invention, the quality parameter associated with the electrocardiogram signal is a signal / noise ratio parameter obtained by calculating the ratio between the amplitude of the peaks R and the amplitude of the frame of the electrocardiogram signal.

[0021] According to an optional characteristic of the invention, the threshold value to which the value of the quality parameter associated with the electrocardiogram signal is compared is equal to 3.4.

[0022] According to an optional characteristic of the invention, a quality coefficient is calculated for each of the cardiac or accelerometric electrical activity data resulting from the signal processing and whose quality parameter is greater than the threshold value.

[0023] Other characteristics, details and advantages of the invention will emerge more clearly on reading the description which follows on the one hand, and examples of embodiment given for informational and non-limiting purposes with reference to the schematic drawings annexed on the other hand, in which:

[0024] [Fig.l] very schematically represents the operation of a system for monitoring at least one parameter representative of an episode of decompensation of heart failure in a patient, said patient being equipped with a subcutaneous implant forming part of the monitoring system according to the invention;

[0025] [Fig.2] is a diagram illustrating the different building blocks of the signal processing method according to the invention making it possible to determine a respiratory parameter, within the framework of the monitoring system of [Fig.l];

[0026] [Fig.3] is a schematic representation of signal processing data of a respiratory parameter of the RSA type;

[0027] [Fig.4] is a schematic representation of a signal processing data of an EDR type respiratory parameter;

[0028] [Fig.5] is a schematic representation of different data resulting from the signal processing of an accelerometric parameter;

[0029] [Fig.6] is a flowchart illustrating the signal processing method according to the invention, and in particular the taking into account of a quality parameter;

[0030] [Fig.7] is a diagram similar to that of [Fig.2], illustrating the process in a standard operating mode, when the quality parameters are correct;

[0031] [Fig.8] is a diagram similar to that of [Fig.2], illustrating the process in a first degraded operating mode, when a first quality parameter is deemed non-compliant;

[0032] [Fig.9] is a diagram similar to that of [Fig.2], illustrating the process in a second degraded operating mode, when a second quality parameter is deemed non-compliant.

[0033] As a reminder, the invention consists of a signal processing method in which a respiratory parameter is determined as a function of an accelerometric signal and an electrocardiogram signal, each of these signals being considered only if a value of a quality parameter associated with this signal is greater than a defined threshold value.

[0034] [Fig.l] illustrates a monitoring system 1 for an episode of decompensation of heart failure. Within this monitoring system 1, a detection device, here a subcutaneous implant 2, is intended to collect measurements relating to the functioning of the heart of a patient 4, both via the acquisition of an accelerometric signal along three distinct axes and an electrocardiogram signal.

[0035] The detection device 2 is configured to communicate with a computer server 6, where appropriate via a communication relay, such that the information containing the different measurements collected by the detection device can be transmitted from the detection device 2 to the computer server 6.

[0036] The signal processing means embedded here in the computer server 6 could be implemented in other media, and for example mobile media, as soon as the computing power necessary for carrying out these calculations is reached. Furthermore, mention is made here of an electrocardiogram signal and an accelerometric signal acquired by the same acquisition device and in particular the same implant, but it will be possible without departing from the context of the invention to have several implants or other acquisition devices, for example non-invasive, to obtain these two signals.

[0037] At the level of the computer server 6, the information collected by the detection device 2 is processed by signal processing means and appropriate calculation means.

[0038] First means 8 dedicated to the processing of the electrocardiogram signal are in particular capable in a first stage of detecting in the acquired electrocardiogram signal the R peaks of the QRS complex for a given cycle, and in a second stage of calculating values ​​on the duration of the RR interval between these R peaks or the amplitude of these R peaks.

[0039] Second means 9 dedicated to processing the accelerometric signal are in particular capable of considering each of the detection axes of the accelerometric signal to deduce therefrom a derived value, an amplitude value or a value with high variability.

[0040] Further details will be given below in the description with reference in particular to Figures 2 to 4 which will follow.

[0041] [Fig.2] is a diagram illustrating the data processing based on the two acquired signals as discussed above.

[0042] In particular, the acquired electrocardiogram signal 10 is processed by the first image processing and calculation means 8 to define cardiac activity data, including a first cardiac electrical activity data item 12 and a second cardiac electrical activity data item 14. These first means 8 also make it possible to calculate a quality parameter 16 associated with the electrocardiogram signal.

[0043] At the same time, the acquired accelerometric signal 18 is processed by the second image processing and calculation means 9 to define cardiorespiratory data, or cardiorespiratory data, including a first cardiorespiratory data item 20, a second cardiorespiratory data item 22 and a third cardiorespiratory data item 24. These second means 9 also make it possible to calculate a quality parameter 26 associated with the accelerometric signal.

[0044] Each cardiac or cardiorespiratory electrical activity data is capable of being processed by a calculation module 27 to determine intermediate respiratory parameters 28, i.e. parameters which depend only on the data considered. Each intermediate respiratory parameter 28 may in particular be an extrapolation of the number of breaths per minute BR based on the data for a defined cycle, and / or be a respiratory profile BW based on this same data.

[0045] Each intermediate respiratory parameter 28 may be accompanied by a quality index QI, which is determined by autocorrelation of the electrocardiogram signal or the accelerometric signal, depending on the intermediate respiratory parameter considered, from cycle to cycle.

[0046] If the quality index is greater than or equal to 60%, the signal is good and the intermediate respiratory parameter can be kept to be combined with the other intermediate respiratory parameters to determine an overall respiratory parameter.

[0047] If the quality index is less than 60%, the intermediate respiratory parameter is not retained.

[0048] These intermediate respiratory parameters can then be combined in a calculation system 30 to determine at least one global respiratory parameter 32, here two global respiratory parameters which are a number of breaths per minute and a respiratory wave profile.

[0049] According to the invention, the various data are processed effectively, to obtain intermediate respiratory parameters and participate in the determination of a global respiratory parameter 32, if a processing authorization is given following the analysis of the corresponding quality parameter. In other words, the cardiac activity data 12, 14 are processed if the quality parameter 16 associated with the electrocardiogram signal 10 is satisfactory and the cardiorespiratory data 20, 22, 24 are processed if the quality parameter 26 associated with the accelerometric signal 18 is satisfactory.

[0050] It should be noted here that according to the invention the quality parameter 16, 26 is distinct from the quality index QI. The signal processing method according to the invention thus comprises two levels of qualitative sorting of the signals, in both cases to be certain of only considering reliable information for the calculation of an overall respiratory parameter.

[0051] According to the invention, to the calculation of the quality index QI which makes it possible to discard the intermediate respiratory parameters which are judged to be less reliable than the others before their combination in the calculation system 30, the calculation of the quality parameters has been added which make it possible to avoid the calculation of the quality index if the signal is not judged to be correct from the start of the method. This avoids launching tedious calculations which have a strong chance of resulting in a quality index below the threshold, since the signal is immediately judged to be poor.

[0052] The calculations implemented in the signal processing method of the invention will be described in detail, with reference to Figures 7 to 9, depending on whether the quality parameters are deemed satisfactory or not.

[0053] As mentioned, the acquired electrocardiogram signal 10 is processed by the first image processing and calculation means 8 to define a first cardiac electrical activity datum 12, a second cardiac electrical activity datum 14 and a quality parameter 16 associated with the electrocardiogram signal.

[0054] The first cardiac electrical activity data 12 and the second cardiac electrical activity data 14 are here obtained by electrocardiogram signal processing means which are configured to detect the R peaks.

[0055] The first cardiac electrical activity data 12, as illustrated in [Fig. 3], is respiratory sinusoidal arrhythmia data RSA which is calculated by determining each interval between two successive R peaks and deducing therefrom a respiratory curve based on an average, for a given cycle, of all these intervals.

[0056] The second cardiac electrical activity data 14, as illustrated in [Fig. 4], is ECG-derived respiration data, or EDR data, which is calculated by determining the amplitude of the R-peaks of a given cycle and deriving a respiratory curve based on the average of these amplitudes.

[0057] The quality parameter 16 is a signal-to-noise ratio parameter that links the desired signal power to the unwanted noise. This signal-to-noise ratio parameter is obtained by calculating the ratio between the amplitude of the R peaks as just mentioned and the amplitude of frame 33 of the electrocardiogram signal. Frame 33 of the electrocardiogram signal is defined as the signal band outside the QRS complexes and is illustrated as an example in [Fig.4].

[0058] As mentioned, the quality parameter 16 is intended to be compared by control means associated with a threshold value 34. The threshold value may in particular be equal to 3.4. The control means are configured to send a control instruction II for processing the cardiac electrical activity data if the value of the quality parameter 16 is greater than or equal to the threshold value 34 and a specific control instruction 12 to prevent this processing of the data if the value of the quality parameter 16 is less than the threshold value 34. The control means may be combined with the first image processing and calculation means 8.

[0059] Simultaneously, and as already mentioned, the acquired accelerometric signal 18 is processed by the second image processing and calculation means 9 to define a first cardiorespiratory data item 20, a second cardiorespiratory data item 22, a third cardiorespiratory data item 24 and a quality parameter 26 associated with the accelerometric signal.

[0060] The second means 9 are in particular configured to process the signals of each of the three axes x, y, z, with a first filtering operation 36 over an appropriate frequency range, and for example a frequency range with frequencies less than or equal to 1 Hz, then by calculating an energy expenditure EE in a calculation phase 38 and by filtering the signals again in a second filtering operation 40 with an adaptive filter depending on the calculated energy expenditure EE. The energy expenditure EE is in particular a function of the sum of the squares of the characteristic values ​​for each axis of the accelerometric signal, according to the following formula: 100611 EE = 2 + ( W y, ) 2 + (s +1 -s) 2

[0062] The first cardiorespiratory data 20, the second cardiorespiratory data 22 and the third cardiorespiratory data 24 are then obtained by calculation means using the filtered values ​​to define main components of the accelerometric signal, namely respectively an average ADR derived value of the filtered signal on the three axes, an average MDR amplitude of the filtered signal on the three axes and a main PCR amplitude of the signal chosen as being the largest amplitude among the three axes.

[0063] The quality parameter 26 associated with the accelerometric signal can also be a signal / noise ratio parameter.

[0064] This quality parameter 26, also called cardiopulmonary activity index, can in particular be calculated from the raw acceleration data using the energy expenditure EE. If the energy expenditure EE is less than 10, the quality parameter takes a value equal to 0 and it takes a value equal to 1 otherwise. In other words, this index determines the presence or absence of cardiopulmonary activity in the accelerometric data, being equal to 0 for the absence of such activity, and the processing of the respiration derived from the accelerometric signal stops, and being equal to 1 for the presence of cardiopulmonary activity, the processing of the respiration derived from the accelerometric signal continues with the following steps.

[0065] As mentioned, the quality parameter 26 is intended to be compared by control means associated with a threshold value 42. The threshold value may in particular be equal to 3.4. The control means are configured to send a control instruction 13 for the processing of the cardiorespiratory data if the value of the quality parameter 26 is greater than or equal to the threshold value 42 and to send a specific control instruction 14 to prevent this processing of the data if the value of the quality parameter 26 is less than the threshold value 42.

[0066] When a control instruction II, 13 is sent, the calculation module 27 cor respondent is able to calculate in particular a QI quality index as mentioned previously, namely an index calculation by autocorrelation.

[0067] More particularly, the autocorrelation for calculating the quality index QI is performed for each of the cardiac electrical activity data 12, 14 taking into account the characteristic of the signal corresponding to this activity data. The quality index calculated on the basis of the first cardiac electrical activity data 12 is thus calculated on the coherence of the intervals between successive respiratory wave peaks, from one given respiratory cycle to another. For example, a reference cycle is defined and all the other cycles are compared to this reference cycle to define the degree of coherence of the interval values ​​of a cycle with the corresponding values ​​of the reference cycle. An index, in the form of a percentage, is defined after comparing all the cycles to the reference cycle.Equivalently, the quality index calculated on the basis of the second cardiac electrical activity data 14 is thus calculated on the consistency of the amplitude of the R peaks from one given cycle to another.

[0068] The different steps of the signal processing method according to the invention as just detailed are illustrated in the flowchart of [Fig.6].

[0069] The flowchart here illustrates the flow of the method for processing an electrocardiogram signal, from acquisition to the data combination step, but it should be noted that what will be described with reference to the flowchart for the electrocardiogram signal can be applied to the accelerometric signal.

[0070] The first step S1 therefore consists of a signal acquisition step, by an appropriate detection device, here a subcutaneous implant but which could just as well be a non-invasive acquisition device. The electrocardiogram signal is sent to the first means 8 dedicated to the processing of the electrocardiogram signal for a second step S2, during which the calculation of the quality parameter 16 associated with the electrocardiogram signal is carried out, and for a third step S3, during which the calculation of the cardiac electrical activity data is carried out, here the first RSA data and the second EDR data.

[0071] In the illustrated example, the two steps S2, S3 are carried out simultaneously but it should be noted that the second step could be carried out in a staggered manner to the third step, and for example take place before the latter to avoid carrying out the third step if the quality parameter is deemed unsatisfactory. It should however be noted that to obtain the quality parameter, the first means 8 must analyze the signal and determine the peaks R, so that the computational load to additionally obtain the cardiac electrical activity data RSA and EDR is minimal. It is thus advantageous to carry out the two steps S2, S3 simultaneously and to have cardiac data already ready when the control means have finished. to determine whether the quality parameter is satisfactory or not.

[0072] For this purpose, a fourth step S4 occurs after the second step S2, with the comparison of the value of the quality parameter with a threshold value 34, here equal to 3.4.

[0073] If the value of the quality parameter 16 is greater than the threshold value 34, a control instruction II is sent to a calculation module 27 to launch a fifth step S5, namely the determination, for each activity data item, here two in number, of one or more intermediate respiratory parameters 28 and a quality index QI. This calculation module 27 is capable of transmitting to a calculation system 30 the intermediate respiratory parameters 28 if the quality index QI is greater than the given threshold, here 60% with reference to the above.

[0074] The calculation system 30 is configured to receive all the intermediate respiratory parameters transmitted by a calculation module 27, whether for the accelerometric signal or for the electrocardiogram signal, and to combine, in a sixth step S6, all these intermediate parameters into a global respiratory parameter 32.

[0075] It is understood that the fifth and sixth steps S5, S6 are only implemented if the quality parameter 16, 26 associated with the electrocardiogram signal, respectively accelerometric, is judged to be satisfactory, i.e. greater than the threshold value 34. This avoids the implementation of time-consuming and costly calculations to determine intermediate respiratory parameters and a quality index QI by autocorrelation, if it is judged that the signals do not have an adequate form to define clear intermediate respiratory parameters, i.e. a form with clearly distinguishable peaks R.

[0076] Figures 7 to 9 illustrate the progress of the method according to different configurations depending on whether a quality parameter 16, 26 is judged satisfactory (dotted lines) or not (solid lines), with a first configuration in which the quality parameters are judged satisfactory for both the accelerometric signal and the electrocardiogram signal ([Fig.7]), with a second configuration in which only the quality parameter associated with the electrocardiogram signal is judged satisfactory ([Fig.8]) and with a third configuration in which only the quality parameter associated with the accelerometric signal is judged satisfactory ([Fig.9]).

[0077] This results in the first configuration in a determination, by the calculation module 27, of intermediate respiratory parameters and of a quality index, for each of the cardiac electrical activity data 12, 14 and for each of the cardiorespiratory data 20, 22, 24. All these respiratory parameters are then transmitted to the calculation system 30 to determine at least one, here two global respiratory parameters 32.

[0078] In the second configuration, the method is unchanged for the electrocar signal diogram, since the quality parameter 16 associated with this electrocardiogram signal is deemed satisfactory in the same way as in the first configuration. For the accelerometric signal, the associated quality parameter 26 is deemed unsatisfactory, i.e. with a value lower than the threshold value 42. This results in a specific control instruction generated by the control means in the direction of the calculation module 27 associated with the accelerometric signal, to indicate that no calculation, and in particular no quality index calculation, must be carried out on the cardiorespiratory data. The calculation system 30 thus only receives intermediate respiratory parameters associated with the electrocardiogram signal to determine the global respiratory parameter(s) 32.

[0079] The third configuration is the reverse of what has just been described for the second configuration. For the accelerometric signal, the method is unchanged compared to the first configuration, since the quality parameter 26 associated with this accelerometric signal is judged satisfactory in the same way as in the first configuration. For the electrocardiogram signal, the associated quality parameter 16 is judged unsatisfactory, i.e. with a value lower than the threshold value 34, here equal to 3.4. This results in a specific control instruction generated by the control means in the direction of the calculation module 27 associated with the electrocardiogram signal, to indicate that no calculation, and in particular no quality index calculation, must be carried out on the cardiac electrical activity data.The calculation system 30 thus only receives intermediate respiratory parameters associated with the accelerometric signal to determine the global respiratory parameter(s) 32.

[0080] The invention as just described achieves the aims it set itself, in particular to provide a signal processing method which makes it possible to determine a reliable respiratory parameter and which does not operate calculation means if the operation of these is not necessary or counterproductive.

Claims

Claims

1. Signal processing method in which a respiratory parameter (32) is determined as a function of an accelerometric signal (18) acquired on three distinct axes and an electrocardiogram signal (10), these signals being intended to be processed by signal processing and calculation means (8, 9) to define respectively cardiorespiratory data (20, 22, 24) and cardiac electrical activity data (12, 14), these data being combined in a step of determining at least one respiratory parameter (32), said method being characterized in that a quality parameter (16, 26) is determined by analyzing the electrocardiogram signal (10) and / or by analyzing the accelerometric signal (18),the data defined on the basis of said electrocardiogram signal and said accelerometric signal respectively being considered in the step of determining at least one respiratory parameter (32) only if a value of the associated quality parameter (16, 26) is greater than a defined threshold value (34, 42).,

2. Signal processing method according to claim 1, characterized in that a first cardiac electrical activity data (12) resulting from the processing of the electrocardiogram signal (10) is a respiratory sinusoidal arrhythmia data calculated on the basis of intervals between consecutive RR peaks of the electrocardiogram signal.

3. Signal processing method according to claim 1 or 2, characterized in that a second cardiac electrical activity data (14) resulting from the processing of the electrocardiogram signal (10) is respiration data derived from the electrocardiogram calculated on the basis of amplitudes of the RR Peaks of the electrocardiogram signal.

4. Signal processing method according to one of the preceding claims, characterized in that a first cardiorespiratory data item (20) resulting from the processing of the accelerometric signal (18) is data item, called ADR type, for which the characteristic value is a derivative of the accelerometric signal on the three axes.

5. Signal processing method according to one of the preceding claims, characterized in that a second cardiorespiratory data item (22) resulting from the processing of the accelerometric signal (18) is data item, called MDR type, for which the characteristic value is an amplitude of the accelerometric signal on the three axes.

6. A method of processing signals according to one of the preceding claims- preceding, characterized in that a third cardiorespiratory data (24) resulting from the processing of the accelerometric signal is data, called PCR type, for which the characteristic value is the value, among the representative values along each of the three axes, which has the greatest amplitude.

7. Signal processing method according to claims 4, 5 and 6, characterized in that said cardiorespiratory data (20, 22, 24), called ADR, MDR and PCR type, for which the characteristic value is respectively a derivative of the accelerometric signal on the three axes, an amplitude of the accelerometric signal on the three axes, and the value, among the representative values along each of the three axes, which has the greatest amplitude, are obtained by filtering the accelerometric signal over a frequency range from 0 to 1 Hz, then by calculating energy expenditure (EE) and adapting the filter as a function of the energy expenditure (EE).

8. Signal processing method according to one of the preceding claims, characterized in that the quality parameter (16) associated with the electrocardiogram signal (10) is a signal / noise ratio parameter obtained by calculating the ratio between the amplitude of the R peaks and the amplitude of the frame (33) of the electrocardiogram signal (10), said frame (33) of the electrocardiogram signal being defined as the band of the signal excluding QRS complexes.

9. Signal processing method according to one of the preceding claims, characterized in that the threshold value (34) with which the value of the quality parameter (16) associated with the electrocardiogram signal (10) is compared is equal to 3.

4.

10. Signal processing method according to one of the preceding claims, characterized in that a quality coefficient (QI) is calculated for each of the cardiac electrical activity data (12, 14) or accelerometric data (20, 22, 24) resulting from the signal processing and whose quality parameter (16, 26) is greater than the threshold value (34, 42).

Images