Method and equipment for analysing biological signals representing intracranial and blood pressure fluctuations

Abstract

A method of analyzing biological pressure signals in a patient including obtaining arterial blood pressure, cerebral blood flow rate and intracranial pressure signals from the patient, and analyzing frequencies of the signals in relation to a range of target frequencies corresponding to at least one of type-B slow waves, infra-B waves and ultra-B waves; and an apparatus for analyzing biological pressure signals in a patient including an acquisition module, located adjacent to the patient and having input channels that receive signals from an arterial blood pressure (ABP) sensor, a cerebral blood flow rate (CBF) sensor, and an intracranial pressure (ICP) sensor; an analysis module that performs frequency analysis of the signals from a target frequency selected from the group consisting of type B slow waves corresponding to a frequency between 8×10−3 hertz and 50×10−3 hertz, infra-B waves corresponding to a frequency lower than 8×10−3 hertz, and ultra-B waves corresponding to a frequency including between 50×10−3 hertz and 200×10−3 hertz; and an exploitation and display module optionally located adjacent the patient in the form of an equipment unit or a local work station connected to the analysis module by an internal network or be constituted by a remote working station connected to the analysis module by a communication network.

Description

[0001] The present invention pertains to the field of methods and equipment for the analysis of biological signals representative of variations in intracranial pressure and blood pressure, and notably for the blood pressure in the cranium.

[0002] The purpose of such analyses is to assist the clinician in the interpretation of data provided by sensors providing signals representative of the intracranial pressure (ICP) and linked signals, the blood pressure, the arterial or venous circulation rate and the gas concentrations. They enable the clinician to deduce from them the suitable treatments for the pathology deduced from these information elements. These analyses have been the object of various scientific studies, e.g.:

[0003] Lemaire et al., “A computer software for frequency analysis of slow intracranial pressure waves”, Comput. Methods Programs Biomed., 1984, 42, 1-14.

[0004] Daley et al., “Fluctuations of intracranial pressure associated with the cardiac cycle”, Journal of Neurosurgery, vol. 11, no. 5, 11-1982, pp. 617-621.

[0005] Avezaat et al., “Cerebrospinal fluid pulse pressure and intracranial volume-pressure relationships”, J. Neurol., Neurosurg. and Psych., 1979, 42, 00. 687-700.

[0006] Portnoy et al., “Cerebrospinal fluid pulse wave as an indicator of cerebral neuroregulation”, J. Neurosur., vol 56, 5-1982, pp. 666-678.

[0007] Equipment for the acquisition and processing of pressure signals in order to perform such analyses have been proposed in the state of the art. For example, PCT patent WO 132076 describes a surveillance device that can determine a physiological parameter in a patient. This device comprises a calibration device configured to provide a calibration signal that is representative of the physiological parameter. A noninvasive sensor is placed on the vessel, with this noninvasive sensor being configured to detect a blood parameter and to produce a signal that is representative of the blood parameter. Thus, there is defined in this context a blood parameter such as the pressure, the flow rate, the volume, the velocity, the movement and the position of the vessel wall and other related parameters. A processor is configured to determine the relationship existing between a characteristic of the excitatory wave received and a characteristic of the physiological parameter.

[0008] PCT patent WO 98/49934 describes a device and a noninvasive method for the measurement of intracranial pressure. The measurement system emits acoustic signals traversing the cranium by means of transmitters and provides an indication of the intracranial pressure as a function of the acoustic signal received after interaction with the brain. Properties such as the impedance of the acoustic transmission, the resonance frequency, the resonance characteristics, the sound velocity and other can be measured and correlated with the intracranial pressure. For example, the acoustic signals present characteristic frequencies of at least 100 kHz, in audible and infrasonic fields. The intensity of the acoustic signal used to measure the intracranial pressure is relatively weak from which derives the possibility of health risks during short or long examinations.

[0009] PCT patent WO 068647 pertains to a method enabling surveillance in a noninvasive manner of the intracranial pressure of a patient. One obtains at least one oscillogram representing a pulsation of an anatomical characteristic of the patient's head, preferable integrating ultrasound reflection traces in a temporal gate corresponding to the reflections of said characteristic. Said anatomical characteristic is preferably the third cerebral ventricle. One infers a quantitative measurement of the intracranial pressure from at least two diagnostic characteristics, such as the diagnosis times, associated with the oscillogram. In a variant, one obtains a qualitative measurement of the intracranial pressure from the shape of the respiratory curve imposed on the wave path by the patient's respiration.

[0010] PCT patent WO 99/26529 describes a fast Fourier transform processing unit applied to waveform frequency analysis (MHj) without corporeal movement and provides waveform analysis data (MKD). In parallel, a descendent wave extraction unit and a wave extraction unit linked to an incisure provide respectively descendent wave data (tide wave data, TWD) and incisure-linked dicrotic wave data (dicrotic wave data, DWD), which represent respectively a descendent wave and a incisure-linked dicrotic wave. A pulse evaluation unit then provides data relative to the pulse status (ZD) on the basis of TWD and DWD data, by means of which a notification unit establishes a report on the status of the pulse of the subject under consideration.

[0011] Also known is the American patent U.S. Pat. No. 4,893,630 describing equipment and a method for the analysis of the pressure in a living organ by a sensor delivering an analog signal, comprising an analog-digital converter and an analysis by Fourier transform to emit a distribution of the signal frequencies provided by the pressure sensors.

[0012] These different solutions are not completely suitable for the provision of particular information useful for the clinician.

[0013] The intention of the invention is to resolve this problem by proposing a system for the representation and analysis of pressure variables comprising pressure sensors [arterial pressure, intracardiac pressure], means for the processing of the signals emitted by said sensors and means for displaying the variations in said signals, characterized in that the processing means comprise a multiplicity of inputs for receiving analog signals stemming from the different sensors, each input being connected to a sampling circuit providing a digital signal used by a calculator for performing processing steps comprising:

[0014] the resampling of the signals for the expansion of the displayed or recorded signal,

[0015] the frequency analysis of said sampled signal in relation to a range of target frequencies recording in a memory, for the display and recording of the temporal variations of the signals corresponding to the slow waves,

[0016] the determination of the temporal shift between the signals corresponding to two distinct inputs.

[0017] The frequency analysis for the extraction of the information relative to the slow waves constitutes an essential improvement of the equipment of the prior art because it offers the clinician new interpretation possibilities.

[0018] According to a first variation, the frequency analysis means are constituted by a calculator applying Fast Fourier Transform (FFT).

[0019] According to a second variant, the frequency analysis means are constituted by a calculator applying wavelet analysis.

[0020] Said range of recorded target frequencies advantageously comprises frequencies comprised between 8·10−3 hertz and 50·10−3 hertz.

[0021] According to a preferred mode of implementation, the system comprises a sampling circuit performing a sampling of each of the input signals at a first frequency and a resampling circuit at a sampling frequency lower than the first frequency for the recording of the time-stamped signals in a memory.

[0022] According to another variant, it comprises a module for the acquisition and processing of the local signals and at last one remote monitoring station connecting to said local acquisition module by a telecommunication network.

[0023] The invention also pertains to a method for the analysis of biological pressure signals comprising a sampling step, characterized in that it also comprises a step of frequency analysis in relation to a range of target frequencies corresponding to type B and UB slow waves.

[0024] According to a variant, the method moreover comprises a step for the determination of the temporal shift between the signals corresponding to the two distinct inputs.

[0025] Better understanding of the invention will be obtained from the description below with reference to a nonlimitative example of implementation in which:

[0026] FIG. 1 represents a schematic view of a system according to the invention;

[0027] FIG. 2 represents a view of a screen for the display of the information displayed by the system.

[0028] The equipment according to the invention is composed of three principal modules:

[0029] an acquisition module (1), generally located close to the patient,

[0030] a processing module (2) generally located with the clinician,

[0031] an exploitation and display module which can be located close to the patient in the form of an equipment unit (3) or in the form of a local work station (4) connected to the processing network by an internal network or constituted by a remote work station (5) connected to the processing module (2) by a telecommunication network, e.g., the internet.

[0032] The first module (1) presents a multiplicity of input channels for receiving the signals originating from various pressure sensors:

[0033] arterial blood pressure (ABP) sensor,

[0034] cerebral blood flow (CBF) sensor,

[0035] intracranial blood pressure (ICP) sensor.

[0036] In the example described, module (1) presents 8 channels. It comprises an 8-path input-output interface (6) emitting a reference signal and analog signals corresponding to the various channels in parallel or multiplexed form.

[0037] The analog signals are then sampled by a circuit (7) controlled by a clock (8). Each signal is preferably sampled at 100 samples by second. It can be displayed directly on the screen of a monitor (9) or be the object of an extraction of one sample out of 8 for recording in a memory (10) in the form of time-stamped digital sequences, forming tables associated with general information (patient's name, practitioner's name, etc.). The module optionally also comprises means for capturing free information in maker form. This information is associated with the recorded information for describing, e.g., an event, or for annotating the curve of an observed signal.

[0038] The analysis module (2) performs a frequency analysis of the sampled signals from a target frequency selected from among:

[0039] the type “B” slow waves corresponding to a frequency comprised between 8·10−3 and 50·10−3 hertz,

[0040] the infra-B waves corresponding to a frequency lower than 8·10−3 hertz,

[0041] the ultra-B waves corresponding to a frequency comprised between 50·10−3 hertz and 200·10−3 hertz.

[0042] This analysis is performed on blocks of N sampled points (power of two), for example, by blocks of 256 points. The result is recorded in a memory (12). A sampled sequence of a signal comprising M samples is thus translated into a file of M/N points corresponding to the variation of the frequency analysis transform.

[0043] The frequency analysis can be performed by Fast Fourier Transform (FFT).

[0044] It consists of performing a spatial frequency analysis by means of a transform enabling translation of a waveform into the frequency domain. The result of the transform is a succession of coefficients describing the amplitude of each frequency component present in the analyzed block.

[0045] The analysis by FFT or by Discrete Cosine Transform (DCT) characterizes each frequency by multiplying the input signal by an example of the target frequency of base function, selected from among the frequencies of the slow waves, and integrating the product obtained.

[0046] The analysis is performed with known electronic circuits (DSP) or by the application of known algorithms, with a sampling rate determined by the clinician, by selection of one of the observed slow-wave frequencies. The parameter N of the sample number in the analyzed recording is determined in advance, for example 256, or variable by the choice of the user. The discrete Fourier transform algorithms convert a time function of the sampled complex values into a function of complex frequency values, also sampled. They provide information such as:

[0047] The table of the coefficients of the cosines in the Fourier formula (real part of the result).

[0048] The table of the coefficients of the sines in the Fourier formula (imaginary part of the result.

[0049] The first output element of each of the tables contains the mean value of all of the inputs. The variant of this element is displayed on the monitor and is the object of comparative processing of one signal to another.

[0050] The extraction of the slow waves is performed from the amplitude (and power) spectrum and is performed by detection of the frequency (F) the amplitude (or power) of which is maximal (P) and this for the bands of B, UB and IB frequencies.

[0051] In the particular, the analysis module (2) comprises means for calculating the shift between the slow waves of two signals and representation on the display monitor of these shifts.

[0052] This analysis is by application of a coherence function in the frequency domain in which a Pearson coefficient is temporal.

[0053] As an example, FIG. 2 represents the view of a display screen from a system according to the invention.

[0054] The screen is subdivided into a multiplicity of display zones for the representation:

[0055] of the temporal variation of the intracranial pressure (zone (20)),

[0056] of the temporal variation of the arterial pressure (zone (21)),

[0057] of the variation of the component corresponding to the slow wave B with a curve (22) corresponding to the frequency in millihertz and a curve (23) corresponding to the amplitude of the pressure,

[0058] the correlation rate between the intracranial pressure signals (or the circulatory rate (20) and arterial pressure (21)) in the form of the curve (24),

[0059] the temporal shift between the slow waves of the intracranial pressure or circulatory rate ? (20) and arterial pressure (21), in the form of the curve (25).

Claims

1. Method for the analysis of biological pressure signals comprising a sampling step, comprising a frequency analysis step in relation to a range of target frequencies corresponding to type-B slow waves, characterized in that the frequency analysis in relation to a range of target frequencies also corresponds to the infra-B waves and/or the ultra-B waves (UB).

2. Method for the analysis of biological pressure signals according to claim 1, characterized in that the infra-B waves correspond to a frequency lower than 8·10−3 hertz.

3. Method for the analysis of biological pressure signals according to claim 1, characterized in that the ultra-B waves (UB) correspond to a frequency comprised between 50·10−3 hertz and 200·10−3 hertz.

4. Method for the analysis of biological pressure signals according to claim 1, characterized in that the frequency analysis step is constituted by a Fast Fourier Transform (FFT) of the sampled signal.

5. Method for the analysis of biological pressure signals according to claim 1, characterized in that the frequency analysis step is constituted by a wavelet analysis.

6. Method for the analysis of biological pressure signals according to claim 1, characterized in that it moreover comprises a step for the determination of the temporal shift between the slow waves of the signals corresponding to two distinct inputs.

7. Method for the analysis of biological pressure signals according to claim 1, characterized in that the type-B target frequency corresponds to frequencies comprised between 8·10−3 hertz and 50·10−3 hertz.

8. Method for the analysis of biological pressure signals according to claim 1, characterized in that it comprises a step of resampling at a sampling frequency lower than the first frequency for the recording of time-stamped signals in a memory (10).

9. Method for the analysis of biological pressure signals according to claim 1, characterized in that it comprises a step of signal processing by filtering in a rereading module.

10. Device for the implementation of the method for the analysis of biological pressure signals according to one of claims 1 to 9, characterized in that it comprises: an acquisition module, generally located close to the patient, presenting a multiplicity of input channels, for receiving signals stemming from different pressure sensors: arterial blood pressure (ABP) sensor, cerebral blood flow rate (CBF) sensor, intracranial pressure (ICP) sensor, an analysis module (2) performs a frequency analysis of the sampled signals from a target frequency selected from among: the type B slow waves corresponding to a frequency comprised between 8·10−3 hertz and 50·10−3 hertz, the infra-B waves corresponding to a frequency lower than 8·10−3 hertz, the ultra-B waves corresponding to a frequency comprised between 50·10−3 hertz and 200·10−3 hertz, an exploitation and display module which can be located close to the patient in the form of an equipment unit (3), or in the form of a local work station (4) connected to the processing module by an internal network or be constituted by a remote work station (5) connected to the processing module (2) by a telecommunication network, e.g., the internet.