Classifications

• • G—PHYSICS
  • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
  • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
  • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
  • G16H20/30—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to physical therapies or activities, e.g. physiotherapy, acupressure or exercising
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/25—Bioelectric electrodes therefor
  • A61B5/279—Bioelectric electrodes therefor specially adapted for particular uses
  • A61B5/28—Bioelectric electrodes therefor specially adapted for particular uses for electrocardiography [ECG]
  • A61B5/282—Holders for multiple electrodes
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/30—Input circuits therefor
  • A61B5/304—Switching circuits
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/30—Input circuits therefor
  • A61B5/307—Input circuits therefor specially adapted for particular uses
  • A61B5/308—Input circuits therefor specially adapted for particular uses for electrocardiography [ECG]
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/24—Detecting, measuring or recording bioelectric or biomagnetic signals of the body or parts thereof
  • A61B5/316—Modalities, i.e. specific diagnostic methods
  • A61B5/318—Heart-related electrical modalities, e.g. electrocardiography [ECG]
  • A61B5/346—Analysis of electrocardiograms
  • A61B5/349—Detecting specific parameters of the electrocardiograph cycle
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
  • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
  • A61B5/6802—Sensor mounted on worn items
  • A61B5/6804—Garments; Clothes
  • A61B5/6805—Vests, e.g. shirts or gowns
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
  • A61B5/7235—Details of waveform analysis
  • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
  • A61B5/7267—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems involving training the classification device
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B2562/00—Details of sensors; Constructional details of sensor housings or probes; Accessories for sensors
  • A61B2562/04—Arrangements of multiple sensors of the same type
  • A61B2562/046—Arrangements of multiple sensors of the same type in a matrix array
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/08—Measuring devices for evaluating the respiratory organs
  • A61B5/0806—Measuring devices for evaluating the respiratory organs by whole-body plethysmography
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
  • A61B5/7235—Details of waveform analysis
  • A61B5/7253—Details of waveform analysis characterised by using transforms
  • A61B5/726—Details of waveform analysis characterised by using transforms using Wavelet transforms
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
  • A61B5/7271—Specific aspects of physiological measurement analysis
  • A61B5/7285—Specific aspects of physiological measurement analysis for synchronizing or triggering a physiological measurement or image acquisition with a physiological event or waveform, e.g. an ECG signal
• • G—PHYSICS
  • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
  • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
  • G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
  • G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients

Definitions

• the field of the invention relates to methods and devices for selecting a set of electrophysiological descriptors linked to an individual's cardiac activity. More particularly, the field of the invention relates to the methods implemented by means of surface electrodes recording signals used to detect a representative cardiac electrical activity. State of the art
• the duration of the QRS designating the duration of depolarization of the ventricles, is frequently used as an indicator of cardiac pathologies.
• This measurement is a very good indicator to detect certain singular activities of the heart, but this measurement alone is only an overview of certain cardiac characteristics. It is for example possible, for a patient suffering from a given pathology, that the measurement of the duration of the QRS can be representative of an electrophysiological singularity. However, this measurement may prove to be insufficient to characterize certain electrophysiological activities of an individual, in particular an electrophysiological activity that can be used or corroborated with other variables to anticipate a cardiac risk.
• the invention therefore aims to propose a method for selecting parameters for measuring the cardiac activity of a patient which overcomes the aforementioned drawbacks.
• the invention relates to a method for selecting a subset of first electrophysiological descriptors characteristic of a characteristic cardiac electrical activity from a set of predefined first descriptors.
• Each first electrophysiological descriptor is associated with at least one channel, a signal type, a signal marker and a statistical calculation modality.
• the method includes recording a plurality of electrical activities defining said pathways and includes:
• Second quantification o of a second proximity factor between the values of the components of each descriptor not selected during the first selection step and the values of the components of the descriptor vector(s) selected during the first selection step; and o a third proximity factor between the component values of each descriptor not selected during the first selection step and the component values of the characteristic vector;
• An advantage of the invention is to propose a method for selecting a subset of different electrophysiological descriptors to characterize cardiac activity.
• the selection method according to the invention makes it possible to select a subset of electrophysiological descriptors which is relevant for discriminating two different populations, from among a set of physiological descriptors.
• the method according to the invention comprises a step of second selection taking into account the third proximity factor, a set of electrophysiological descriptors is then obtained which are not redundant with each other to characterize said cardiac activity.
• the selection method according to the invention therefore makes it possible to obtain an effective and non-redundant subset for the discrimination of two populations.
• At least one first descriptor of the subset is associated with a statistical modality different from that of another first descriptor of the subset and with a signal marker different from that of the other first descriptor.
• the different steps of the method of the invention can be implemented by means of calculations such as computers. These can be those of an electronic board of dedicated equipment or those of a computer or a remote data server.
• the second quantification and second selection steps are reproduced from descriptors not previously selected until a predefined number of selected descriptors is obtained.
• This arrangement makes it possible to build the subset of descriptors in an iterative way by taking each time descriptors carrying information different from those carried by the descriptors already selected.
• the quantification of the first proximity factor and/or of the second proximity factor, and/or of the third proximity factor is a correlation calculation.
• a correlation calculation is an efficient way to estimate the proximity between two vectors. For example, the correlation is performed between vectors defined in the metric space.
• the second quantization step includes a step of projecting, in a first plane orthogonal to the descriptor vector(s) associated with the last selected descriptors, of each non-selected descriptor vector and of the characteristic vector; in that the quantification of the third proximity factor is performed from the components of the projections of the projected vectors; and in that the second selection is made as a function of the third proximity factor uniquely.
• the projection in the plane orthogonal to the selected descriptor vector makes it possible to remove the descriptor vectors very close to the selected descriptor vector. This arrangement therefore makes it possible to select one or more descriptor vectors which carry different information.
• the second quantization step can be implemented by an operator other than the projection.
• An operator for defining a distance in the metric space of vectors can be used to calculate any of the proximity factors, including the first proximity factor, the second proximity factor, and the third proximity factor.
• different operators can be configured so as to calculate the different proximity factors. It is understood in the present invention that a proximity factor is understood with respect to a distance defined in the metric space. Thus, certain distances may have advantageous properties depending on the proximity factor that is measured. Some operators favor certain dimensions of the vectors during the quantification, others allow favoring a measure of proximity or proximity of vectors between them, others favor the distance, the orthogonalisation or the difference of the vectors between them.
• the projection of two vectors makes it possible to obtain a quantification of a proximity factor in a subspace, however other methods, such as the calculation of quadratic error between two vectors, can be used to define a distance for the purpose of quantifying a proximity factor.
• the predefined number of selected descriptors is determined by testing the effectiveness of a set of “n” first selected descriptors in characterizing a given electrical activity, “n” corresponding to the number of descriptors taken from the selected descriptors , starting from the first to the nth selected descriptor, and incrementing the value of n in steps of 1 .
• This arrangement makes it possible to calibrate the size of the subset of descriptors in order to have the subset with the greatest efficiency.
• the set comprises at least one second geographical descriptor associated with several routes and several geographical groups, each geographical group being formed by a central channel and the at least four channels close to the central channel, the value of the electrophysiological descriptor being determined:
• second descriptors are used which take geographical groups into account.
• the addition of this other type of descriptors makes it possible to take into account the concentration of singular measured values and enriches the subset of selected descriptors.
• the at least one channel comes from a predefined zone on the patient's body is chosen between:
• the type of signal analyzed is chosen between:
• a Laplacian signal estimated by subtracting from the potential of a central electrode the average of the voltage of the eight electrodes directly near said central electrode.
• This arrangement makes it possible to take into account several types of signals and therefore to have descriptors which take into account several types of signal measurement.
• the descriptors take into account several types of information and reflect more faithfully the state of said patient.
• the reference electrode is an electrode arranged on the surface of an upper limb or a lower limb of the patient. This arrangement allows the measurement of a reference potential.
• a plurality of reference electrodes is arranged on the surface of the lower or upper limb(s) of the patient. This arrangement makes it possible to obtain a reference potential with great precision.
• the signal marker is the measurement of the voltage of an averaged signal.
• the averaging of the signal makes it possible to obtain a stable measurement of said voltage.
• the signal marker is the measurement, on the signal averaged and filtered between 40 and 250 Hertz, of the duration of depolarization of the ventricles or of the fragmentation of the signal during the depolarization of the ventricles .
• Ventricle depolarization time is a very representative measure of cardiac activity.
• the signal marker is the measurement on the discrete wavelet decomposition of the signal:
• the signal marker is the measurement on the decomposition into continuous wavelets of the signal of the number of chains of local maxima. This measurement is a measurement that makes it possible to account for a singular cardiac activity.
• the signal marker is the measurement on a roundel taken between 256 and 512 hertz of the signal:
• the signal marker is the measurement on a roundel taken between 128 and 256 hertz of the signal:
• the signal marker is the measurement on a roundel taken between 64 and 128 hertz of the RMS (Root Mean Square). This measurement makes it possible to report the cardiac activity of the patient.
• the statistical modality is chosen from:
• the method of the present invention can be succeeded by a method for generating an electrophysiological parameter to characterize cardiac activity.
• Such a step can be implemented by a method for generating an electrophysiological parameter which comprises:
• each electrophysiological descriptor of the subset being associated with at least one channel, with a type of signal , a signal marker and a statistical calculation modality;
• each electrophysiological descriptor being calculated from the statistical modality which is applied to the signal marker of the acquired signal according to the type of signal on a selected channel associated with said electrophysiological descriptor; • comparison of the value of the set of electrophysiological descriptors with at least one threshold value specific to the set of electrophysiological descriptors, said at least one threshold value being defined by a statistical distribution of said descriptors of a set of healthy patients;
• the invention relates to a device or a system comprising means for implementing the method of the invention.
• the means may include computers, memories, electronic cards, electrodes and electrode supports.
• the device or system of the invention may include computers or servers when computing resources are required. The invention is described below in such a way that the features described may relate to the method of the invention or to the device or system of the invention.
• the invention also relates to a device for selecting a subset of electrophysiological descriptors.
• the device according to the invention is capable of implementing the method mentioned above. Subsequently, the elements described in this description will be applicable both to the method according to the invention and to the device according to the invention.
• the invention also relates to a device for selecting a subset of electrophysiological descriptors which comprises:
• a plurality of surface electrodes configured to be deposited on the body of a patient and to measure an electric potential of the surface of the body of the patient, each surface electrode defining a path;
• a means of calculation configured for: i. Estimate for a first set of patients not affected by a predefined state, values of each electrophysiological descriptor of a set of first predefined descriptors, each first descriptor being associated with at least one pathway, with a type of signal, to a signal marker and to a statistical mode of calculation; at least one first descriptor of the subset being associated with a statistical modality different from that of another first descriptor of the subset and with a signal marker different from that of the other first descriptor; ii. Estimating for a second set of patients affected by the predefined condition, values of each descriptor of the set of predefined descriptors; iii.
• An advantage of the invention is to provide a device for selecting a subset of different electrophysiological descriptors to characterize cardiac activity.
• the selection device according to the invention makes it possible to select a subset of electrophysiological descriptors which is relevant for discriminating between two different populations, among a set of physiological descriptors.
• a set of electrophysiological descriptors is then obtained which are not redundant with each other to characterize said cardiac activity.
• the selection device according to the invention therefore makes it possible to obtain an efficient and non-redundant subset for the discrimination of two populations.
• the electrodes are arranged on the patient's surface using adhesive strips. This characteristic represents a practical means to be implemented for arranging the electrodes on the patient.
• the device comprises a device for detecting the breathing phases of a patient, preferably a plethysmography belt. This arrangement makes it possible to carry out the measurements during a breathing phase of the patient which does not disturb said measurements.
• the device according to the invention is able to implement the method according to the invention.
• Fig. 1 a schematic flowchart of the process according to the invention
• Fig. 2 a front view of a patient's torso on which a plurality of measurement electrodes are placed to implement a method according to the invention
• Fig. 3 a view of a plurality of measurement electrodes in a predefined area of the patient's body
• Fig. 4 two curves illustrating the method for calculating an asymmetry index
• Fig. 5 a graph illustrating the method for calculating a Kurtosis index
• Fig. 6 a view of a curve illustrating the method for calculating the number of zones of reduced amplitude.
• the invention relates to a method for selecting a subset of electrophysiological descriptors characteristic of a characteristic cardiac activity.
• the invention also relates to a device for selecting a subset of electrophysiological descriptors characteristic of a characteristic cardiac activity.
• the device according to the invention is capable of implementing the invention mentioned above. Subsequently, the elements described in this description will be applicable both to the method according to the invention and to the device according to the invention.
• the invention relates to a method for selecting a subset ⁇ Dk ⁇ of first electrophysiological descriptors Di characteristic of an electrical activity representative of a characteristic cardiac activity.
• characteristic cardiac activity is meant a cardiac activity of a patient which may be characteristic of a state of the latter.
• the subset ⁇ Dk ⁇ of first electrophysiological descriptors Di is selected from a set ⁇ DN ⁇ of electrophysiological descriptors.
• the method according to the invention relates to the selection, from a set ⁇ DN ⁇ of electrophysiological descriptors, of a more restricted subset ⁇ Dk ⁇ of descriptors.
• the aim of the method according to the invention is to carry out a relevant and reduced selection of a few descriptors Di from the set ⁇ DN ⁇ .
• each first electrophysiological descriptor Di is associated with the recording of at least one pathway Vi.
• the recording of a Vi channel corresponds to the recording of the electrical activity picked up by at least one EL electrode placed on the surface of a patient's body.
• the invention can be implemented in particular by means of a memory allowing the recording of the data acquired and/or the data processed by one of the steps of the method of the invention.
• Each first descriptor Di is associated with at least one pathway Vi on a predefined zone Zi on the patient's body.
• predefined zone Zi is meant the zone of the patient's body on which the measuring electrode(s) EL are deposited, the electrical activities of which are recorded to obtain the channel(s) Vi.
• the surface of a patient's body can be segmented into different functional and/or geometric and/or physiological zones. These zones may therefore correspond to geographical zones on the patient's body, to functional zones with respect to the cardiac activity of the patient, or to zones corresponding to the physiology and/or the physiology of the latter.
• Each first descriptor D i is associated with a type of measured signal Ti on the selected channels V.
• type of signal Ti is meant, for example, the measurement of a voltage between two electrodes. The different types of Ti signals that can be selected will be described later.
• Each first descriptor Di is associated with a signal marker Mi.
• Signal marker Mi means the characteristic of the type of signal Ti which will be measured. According to examples of signal markers Mi used, some comprise the measurement of frequency and energy characteristics of the signal. According to other examples, markers can include the measurement of voltage or electrical voltage. The various selectable Mi signal markers are described below.
• Each first descriptor Di is associated with a statistical modality MSi.
• Statistical modality MSi means a statistical measurement modality applied to the measurements performed on the signals. A statistical modality MSi can for example be the calculation of the average of the signal marker Mi measured on several electrodes. The different MSi statistical modalities that can be selected will be described later.
• Each first electrophysiological descriptor Di is therefore defined both by a selection of the predefined zone Z, of the patient's body, of the type of signal Ti, of the signal marker Mi and of the statistical modality MSi.
• At least two first descriptors Di, Dj of the subset ⁇ Dk ⁇ are associated with a different statistical modality MSi.
• two of the first descriptors Di, Dj are not associated with the same statistical modality MSi.
• At least two first descriptors Di, Dj of the subset ⁇ Dk ⁇ are associated with a different signal marker Mi.
• two of the first descriptors Di, Dj are not associated with the same signal marker Mi.
• a first step of the method according to the invention corresponds to an estimation ESTi for a first set of patients ENSi of values of each first electrophysiological descriptor Di.
• the first set of patients only includes patients not affected by a predefined ETi state.
• predefined state is meant a state exhibiting particular characteristics at the cardiac level, whether this state is pathological or not.
• the predefined state ETi is a structural cardiac pathology.
• a second step of the method according to the invention corresponds to an estimation EST2 for a second set of patients ENS2 of values of each first electrophysiological descriptor Di.
• the second set of patients comprises only patients affected by the predefined state ETi.
• the estimation of the values of the first descriptors Di is performed for each first electrophysiological descriptor Di of the set of descriptors ⁇ DN ⁇ .
• a next step is the generation GENi of a first characteristic vector Vi of the state of each patient.
• This vector Vi comprises a component per patient of the first set ENSi and a component per patient of the second set ENS2.
• Each component of the characteristic vector V1 corresponds to a state relating to the predefined state ETi.
• the first vector comprises a first predefined value vpi for each patient of the first set ENS1.
• the first characteristic vector V1 comprises a second predefined value vp2 for each patient of the second set ENS2.
• the components of the first characteristic vector V1 are a function of the level of attack by the predefined state ET1 of the patients of the first and second sets ENS1, ENS2.
• the components of the first vector V1 can for example have a low value when the patient is weakly affected by the predefined state ET1 and a high value when the patient is strongly affected by the predefined state ET-i.
• the first characteristic vector V1 comprises a component equal to “zero” for each patient of the first set ENS1.
• the first characteristic vector V1 comprises a component equal to “one” for each patient of the second set ENS2.
• a next step of the method according to the invention is the generation GEN2 of a descriptor vector Vd for each first electrophysiological descriptor Di.
• This GEN2 generation is performed in metric space.
• Each component of the descriptor vector Vd corresponds to the value of said first electrophysiological descriptor Di.
• each descriptor vector Vd comprises one component per patient of the first and second sets ENS1, ENS2.
• a next step of the method according to the invention relates to the first quantification QUAN1 for each descriptor Di of a first proximity factor between the values of the components of the first characteristic vector V1 and the values of the components of the descriptor vector Vd.
• This step makes it possible to estimating whether there is a correlation relationship between the values of the first characteristic vector V1 and the values of the descriptor vector Vd for each descriptor Di considered.
• the first quantification step QUANi is performed by quantifying for each descriptor Di the statistical correlation between the values of the components of the first characteristic vector Vi and the values of the components of the descriptor vector Vd.
• this quantification can be performed by calculating a squared error.
• the first quantification step QUANi is carried out using a variable selection method such as the LASSO method.
• a RIDGE regression can be used.
• a next step of the method according to the invention is a first selection SELi of at least one descriptor Di.
• at least one descriptor Di is selected as a function of the previously quantified values of the first proximity factor of each descriptor Di.
• the first descriptor(s) Di selected are included in the subset ⁇ Dk ⁇ .
• the selection is made by taking the descriptor(s) Di which have optimal proximity factor values among all the descriptors Di of the set ⁇ DN ⁇ .
• optimal value we mean a value which accounts for the proximity of the descriptor vector Vd with the first vector Vi. In the case where the proximity factor is a correlation calculation, the optimal value is a maximum value. In a case where the proximity factor is a quadratic error, the optimal value is a minimum value. This step therefore makes it possible to select the descriptor or descriptors D, having the greatest proximity to the characteristic vector Vi.
• a next step of the method is a second QUAN2 quantification step.
• This step is broken down into two distinct steps. There is first the quantification of a second proximity factor between the components of the descriptor vectors Vd of the descriptors Di which were not selected during the first selection SEL1 and the values of the components of the descriptor vector(s) Vd which were selected during the first selection step SEL1. There is then the quantification of a third proximity factor between the components of the descriptor vectors Vd of the descriptors Di which were not selected during the first selection SEL1 and the values of the components of the characteristic vector V1. It can be noted that this second quantization step can be performed from the values of the components of the descriptor vectors Vd and of the characteristic vector or from a transformation in the narrow space of said components.
• the second quantization step QIIAN2 is performed by quantifying the statistical correlation between the values of the components of the different vectors involved.
• this quantification can be performed by calculating a quadratic error.
• the second QIIAN2 quantification step is performed using a variable selection method such as the LASSO method.
• a RIDGE regression can be used.
• a next step of the method according to the invention is a second selection step SEL2 of at least one new descriptor Di and its inclusion in the subset ⁇ Dk ⁇ .
• the selection of the new descriptor Di is performed as a function of the values of the second proximity factor and of the third proximity factor quantized during the second quantization QUAN2. More specifically, during this selection step, the descriptor(s) Di having a strong quantified proximity to the characteristic vector V1 and a weak quantified proximity to the descriptor vector(s) Vd selected previously are selected.
• the second proximity factor and the third proximity factor makes it possible to obtain a subset of descriptors D, which takes into account the redundancy of the information carried by several descriptors Di. According to these characteristics, a classification is obtained which no longer only takes into account the correlation between the descriptor vectors Vd and the first vector V1, but which also makes it possible to select descriptor vectors Vd which have a high correlation value, but are not too similar to the first descriptor vectors Vd selected. This arrangement makes it possible to select a subset ⁇ Dk ⁇ which comprises non-redundant descriptors Di.
• the method according to the invention allows the selection of a subset ⁇ Dk ⁇ of descriptors Di which make it possible to discriminate between two sets of patients in an efficient and non-redundant manner.
• the method according to the invention makes it possible to sort among a large number of descriptors electrophysiological Di to select a reduced number capable of differentiating patients according to whether or not they have been reached by the predefined state ETi.
• the steps of second quantification QIIAN2 and second selection SEL2 are reproduced from descriptors Di not previously selected until a predefined number of descriptors Di is obtained.
• This method follows several iterations of the second quantification QIIAN2 and second selection SEL2 steps and thus makes it possible to obtain a subset ⁇ Dk ⁇ which includes the desired number of electrophysiological descriptors Di.
• the second quantization step QIIAN2 includes a projection step in a first plane Pi orthogonal to the descriptor vector(s) Vd associated with the last descriptors selected. During this projection step, all the descriptor vectors Vd which have not been previously selected are projected into the plane Pi. Similarly, the characteristic vector V1 is projected into the plane Pi.
• This projection step can be likened to the second quantification step QIIAN2 of the second proximity factor between the components of the descriptor vectors Vd not previously selected and the descriptor vectors Vd already included in the subset ⁇ Dk ⁇ .
• the projection is worth a quantification of the proximity factor, because the projection of a vector in a plane orthogonal to a vector which is very close to it gives a vector which is almost zero.
• the quantification of the third proximity factor is performed from the components of the projected descriptor vectors Vd.
• the quantification of the third proximity factor is performed from the components of the characteristic vector V1 which has been projected.
• the predefined number of descriptors Di selected is determined by testing the effectiveness of the subset ⁇ Dk ⁇ of descriptors Di in discriminating the first set ENS1 of patients from the second ENS2 set of patients. In this way, the number of descriptors Di which makes it possible to obtain the best performance for the subset ⁇ Dk ⁇ is sought.
• the predefined number of descriptors Di selected is determined by testing the effectiveness of a subset of descriptors, starting by testing the effectiveness of the first descriptor Di selected. Then, we test the efficiency of the first two selected descriptors Di. A number is then incremented in steps of "1", by testing the first three, then the first four descriptors. We continue this iteration until we reach a number “n” of descriptors tested. Finally, the predefined number of descriptors Di is kept for which the discrimination performance of the first set of patients ENS1 and of the second set of patients ENS2 is the greatest.
• each first electrophysiological descriptor Di is associated with at least one pathway V, of a predefined zone Zi of the patient's body, with a type of signal Ti, with a signal marker Mi, and with a statistical calculation modality MSi.
• a selection is made of a predefined zone Zi among a set of predefined zones Zi.
• a selection is made of a type of signal Ti from a set of types of signals Ti.
• a selection is made of a signal marker Mi from a set of signal markers Mi.
• a selection is made of a statistical modality MSi from a set of statistical modalities MSi.
• a plurality of electrodes EL is deposited on the surface of the patient's torso of each patient of the sets ENSi and ENS2 to implement the selection method according to the invention.
• the number of electrodes in the example represented can vary, one can for example have a number of electrodes much lower than that represented, or much higher.
• the plurality of EL electrodes covers a large portion of the patient's torso. As can be seen, this plurality of electrodes is separated into four distinct zones on the latter.
• a first part of the electrodes EL is located on an upper right zone Z1 of the patient's torso.
• a second part of the EL electrodes is located on an upper left zone Z2 of the patient's torso.
• a third part of the EL electrodes is located on a lower right zone Z3 of the patient's torso.
• a part of the EL electrodes is located on a lower left zone Z4 of the patient's torso.
• the plurality of electrodes EL is included in a zone comprising the entire Z5 of the patient's torso.
• the demarcation between the zones located on the left of the torso and those located on the right of the torso is a vertical line passing through the center of the torso, or substantially through the center of the torso.
• the demarcation between the areas located at the bottom of the torso and those located at the top of the torso is a horizontal line passing through the center of the torso.
• Each zone has a predefined number of EL electrodes.
• the number of electrodes arranged per zone may be around thirty. It is for example possible to have thirty EL electrodes per zone.
• each zone comprises the same number of EL electrodes.
• the zone comprising the entire torso Z5 comprises a different number of electrodes than the others, because this zone Z5 comprises the union of the electrodes of all the other zones Z-i, Z2, Z3 and Z .
• zone Z5 comprising the entire torso of the patient comprises thirty-six electrodes EL.
• Each Vi channel is obtained by recording the electrical activity of at least two EL electrodes.
• These at least two electrodes can be two electrodes of one or more areas of the patient's torso.
• These at least two electrodes can also be an electrode of a region of the patient's torso and a reference electrode.
• the set ⁇ DN ⁇ comprises at least one second geographic electrophysiological descriptor Di.
• the at least one geographical descriptor Di is associated with at least one channel V, and with several geographical groups.
• a geographical group is formed by an EL electrode and the four EL electrodes which are located directly near it.
• a geographic cluster is shown in Figure 3. This geographic cluster has a central electrode EL1, as well as the electrode located directly above it. It also includes the electrode located directly below the center electrode EL1, the electrode located directly to the left of the center electrode EL1, and the electrode located directly to the left of the center electrode EL1. These four electrodes are shown hatched in FIG. 3.
• the measurement of the value according to the type of signal and the signal marker is carried out on all the available geographical groupings of the plurality of electrodes arranged on the body. of the patient.
• Other arrangements of the electrodes can be envisaged. It is possible in particular to select the central electrode and the four electrodes located at the top left, at the top right, at the bottom left and at the bottom right of the central electrode EL1. These are the electrodes appearing without a pattern in FIG. 3. It is also possible to provide more electrodes, for example nine electrodes in the geographical group. These are the nine electrodes EL of FIG. 3 for example.
• the value obtained for each electrode EL of said grouping is compared with at least one geographical threshold value.
• the at least one geographic threshold value is obtained from a statistical distribution of the value of the considered EL electrode of the set of ENSi patients not affected by the predefined state ETi.
• the geographical group is considered significant.
• the value of the descriptor is the number of significant geographical groups counted. Alternatively, one can consider a geographical group as meaning from two electrodes exceeding their geographical threshold value, or else with four electrodes.
• the statistical modality is not taken into account, the value of the descriptor being the number of geographical groupings detected.
• the subset ⁇ Dk ⁇ of descriptors comprises at least one first descriptor Di and at least one second geographical descriptor. Additionally, the subset ⁇ Dk ⁇ comprises several first descriptors Di. According to this variant, the subset ⁇ Dk ⁇ comprises a second geographic descriptor per signal marker used in the first descriptors Di of the subset ⁇ Dk ⁇ .
• Each electrophysiological descriptor Di is associated with a type of signal Ti.
• Figure 3 is a schematic representation of nine contiguous EL electrodes on the patient's body.
• each circle represents an EL electrode.
• the dotted areas represent the different EL electrodes selected in the different signal types.
• the type of signal Ti can be chosen preferably between four different types of signals.
• a first type of signal T is a unipolar signal.
• a unipolar signal is a signal taken between an electrode EL of the predefined zone Zi and a reference electrode.
• the type of unipolar signal is the voltage measured between the electrode of the predefined zone and the reference electrode.
• Reference electrode means an electrode which is not located in one of the areas of the patient's torso defined previously.
• a reference electrode can be an electrode placed on a lower limb or an upper limb of a patient.
• a second type of signal Ti is a vertical bipolar signal.
• a vertical bipolar signal is a signal taken between an electrode in the predefined zone and the electrode located directly below it on the patient's torso.
• the type of signal acquired is the voltage between the two EL electrodes.
• a vertical bipolar signal is taken between two electrodes of the predefined zone Zi. These two electrodes form a vertical bipole B.
• a third type of signal Ti is a horizontal bipolar signal.
• a horizontal bipolar signal is a signal taken between an electrode of the predefined zone and an electrode located directly next to it along a horizontal line on the patient's torso.
• the type of signal acquired is the voltage between the two EL electrodes.
• the horizontal bipolar signal is taken between two electrodes of the predefined zone Zi. These two electrodes form a horizontal bipole Bh.
• a fourth type of signal Ti is a Laplacian signal.
• a Laplacian signal is estimated by subtracting from the potential of a central electrode ELi the average of the potentials of the eight electrodes which are directly close to said central electrode.
• the Laplacian signal is a compound voltage between the central electrode ELi and a set of EL electrodes peripheral to the central electrode ELi. These nine electrodes form an ELiap Laplacian electrode.
• Each electrophysiological descriptor Di is associated with a signal marker Mi.
• a signal marker Mi is a mode of measurement of a physical quantity associated with the types of signals measured by the EL electrodes.
• the signal marker Mi associated with a descriptor Di is preferably chosen from among fourteen signal markers Mi. These signal markers Mi are described below.
• a first signal marker Mi corresponds to the measurement of an averaged electrical signal. By averaged electrical signal is meant the calculation, performed on the measured voltage, of the average between the maximum peak and the minimum peak of the QRS. This measurement of QRS duration is generally fairly representative of cardiac activity.
• the signal is filtered using a bandpass filter.
• the bandpass filter is a bidirectional Butterworth filter.
• a bidirectional Butterworth filter has the advantage of limiting the oscillations due to the filtering, which makes the calculation of the values for certain signal markers Mi more precise.
• a signal marker Mi on the filtered signal is the duration of the QRS on the filtered signal.
• a mark is placed on the beginning of the QRS and a second mark is placed at the end of the QRS. The time between the two markers is measured.
• This operation can be performed automatically using an algorithm for detecting the beginning and end of QRS.
• this duration can be measured manually by an operator on an interface. It is also possible to provide an automatic measurement of the duration of the QRS and a manual control of said measurement by the operator on the interface.
• the duration of the QRS is detected by moving a sliding window measuring the energy of the filtered signal.
• a marker is placed that marks the start of the window.
• the end mark of QRS is placed in the same way.
• Another signal marker Mi is the fragmentation measurement of the filtered averaged signal between 40 hertz and 250 hertz.
• the number of QRS peaks on the filtered signal is measured.
• peak is meant a local maximum of the curve of the filtered signal.
• the number of peaks is measured on the section of the curve corresponding to the QRS.
• the QRS start and end markers are set in the same way as for the previous Mi marker, which as a reminder is the QRS duration marker on the filtered signal.
• the following Mi markers are calculated on the wavelet decomposition of the signal.
• these markers Mi it is possible to use the decomposition into continuous wavelets or the decomposition into discrete wavelets.
• a first marker Mi is the calculation of the energy on the discrete wavelet decomposition of the signal.
• the energy is calculated on the sum of the coefficients on several levels. Typically, the sum of the coefficients is carried out between 64 hertz and 1024 hertz, i.e. on the four levels of this frequency band.
• the measured energy is normalized with respect to the duration of the QRS. Alternatively or additionally, the energy is normalized with respect to the maximum amplitude of the signal.
• a second marker Mi calculated on the discrete wavelet transform is the measure of the index called Kurtosis Sku.
• Kurtosis is meant an index making it possible to estimate the spread of a given curve.
• Figure 5 illustrates several measurements of curve spread on three example curves. For a flat curve P, the Kurtosis index is negative. For a slender curve E, the index is positive. Thus, the more the curve is spread out, the more the Kurtosis is negative. When the curve is narrow, the Kurtosis is positive. The Kurtosis of a curve representing a normal distribution N is equal to zero.
• the Kurtosis Sku is calculated on the sum of the coefficients on several levels of the discrete wavelet decomposition. Typically, the sum of the coefficients is carried out between 64 hertz and 1024 hertz, ie over the four levels of this frequency band.
• a third marker Mi calculated on the discrete wavelet transform is the measurement of the Fischer asymmetry coefficient.
• This coefficient can also be called “skewness”. This coefficient makes it possible to estimate the asymmetry of a given curve.
• Figure 4 illustrates two measurements of asymmetry on two curves given as examples. Curve 1 is a left-tending curve and curve 2 is a right-tending curve. Fischer's asymmetry coefficient has a positive value when the curve tends to the left. This is the case of curve 1. Fischer's asymmetry coefficient has a negative value when the curve tends to the right. This is the case of Figure 2. Concretely, the Fischer asymmetry coefficient is calculated on the sum of the coefficients on several levels of the discrete wavelet decomposition. Typically, the sum of the coefficients is carried out between 64 hertz and 1024 hertz, ie over the four levels of this frequency band.
• a fourth marker Mi calculated on the discrete wavelet transform is the measurement of the number of local minima chains of said decomposition.
• a chain of local minima is the presence on several levels of discrete wavelet decomposition of the same minimum.
• the measurement is made between 64 hertz and 1024 hertz, i.e. on the four levels of this frequency band.
• the measurement of the number of chains of local minima can be performed on the continuous wavelet decomposition.
• Another signal marker Mi that can be chosen is the measurement on the continuous wavelet decomposition of the number of chains of local maxima.
• a chain of local maxima is the presence on several levels of discrete wavelet decomposition of the same maximum.
• the measurement is made between 64 hertz and 1024 hertz, i.e. on the four levels of this frequency band.
• the measurement of the number of chains of local maxima can be performed on the discrete wavelet decomposition.
• the following two Mi signal markers are measured on the wavelet of the signal included in the frequency band from 256 hertz to 512 hertz of the signal.
• the first concerns the measurement of the Kurtosis Sku index on this wavelet.
• Kurtosis Sku is meant the same indicator as described previously in the application.
• the second signal marker Mi measured on this wavelet is the measurement of the number of areas of reduced amplitudes RED of Rondelle.
• the upper and lower envelopes of the signal are created.
• the number of reduced amplitude zones is calculated on the signal envelopes. This is illustrated by Figure 6 which shows a signal and the areas of reduced RED amplitudes detected.
• the following three signal markers Mi are measured on the wavelet of the signal included in the frequency band ranging from 128 hertz to 256 hertz of the signal.
• the first concerns the measurement of the Kurtosis Sku index on this wavelet.
• Kurtosis Sku is meant the same indicator as previously described in the application.
• the second signal marker Mi measured on this wavelet is the measurement of the number of zones of reduced amplitudes of the wavelet. The number of zones of reduced amplitude is calculated in the same way as for the marker relating to roundness of the signal comprised in the frequency band going from 256 hertz to 512 hertz of the signal.
• the third signal marker concerns the measurement of the RMS of rondeau in the frequency band 128 Hertz to 256 Hertz.
• RMS or “Root Mean Square”
• a signal marker Mi which can be selected is measured on a roundel of the signal comprised in the frequency band going from 64 to 128 hertz.
• This signal marker Mi concerns the measurement of the effective amplitude RMS (“Root Mean Square”) of the signal.
• Mi signal markers may be used beyond the fourteen Mi signal markers described. It is for example possible to use signal markers Mi which are combinations of signal markers Mi already described.
• Each first electrophysiological descriptor Di is associated with a statistical modality MSi.
• statistical modality MSi is meant a modality for processing the various quantities measured in order to calculate a value for each descriptor Di.
• a first statistical modality is the minimum at the fifth percentile of the measured values.
• This statistical modality has the advantage, by removing the five percent of the lowest values, of removing aberrant values which could distort the representativeness of the measurement.
• a second statistical modality that can be selected is the maximum at the 95th percentile .
• the five percent highest values are taken from all the values selected. The highest value of the remaining values is then selected.
• This statistical modality makes it possible not to take into account, for a measurement of a maximum value, the aberrant values which could appear in the highest measured values. In this way, an upper limit representative of all the measured values is available.
• a third statistical modality MSi is the average of the measured values.
• the mean is a classic indicator and representative of a distribution.
• a fourth statistical modality MSi is the standard deviation calculated on the set of measured values.
• the standard deviation is a representative value of the dispersion of values.
• the dispersion can be a significant value, a large variance in the measurements carried out being able to be the sign of a disorder in the cardiac activity of the patient.
• a fifth MSi statistical modality that can be selected is the median.
• the median value of a set of values is the value that separates the set of values into two sets of the same size. This value gives information that can vary from that given by the average value, because the median makes it possible not to give too much importance to aberrant values close to the maximum and the minimum of the measured values.
• a sixth statistical modality MSi is the value of the interquartile. To calculate this value, we calculate the value of the 25th percentile and a value of the 75th percentile .
• the interquartile value represents the difference between the 75th percentile value and the 25th percentile value.
• the interquartile is an interesting statistical value to look at to characterize the distribution of the measured values.
• the invention also relates to a device for measuring the values of the electrophysiological descriptors previously described.
• the invention also relates to the means used to implement the method for selecting a subset of descriptors according to the invention.
• the characteristics described above concerning the method according to the invention also apply to the device according to the invention.
• the characteristics described below for the device also apply to the method according to the invention.
• the device for measuring the values of the electrophysiological descriptors comprises a plurality of electrodes which are arranged on the surface of the patient's body. Each surface electrode defines a channel Vi.
• the device comprises adhesive strips comprising the surface electrodes.
• the adhesive strips are intended to be stuck to the surface of the patient's body.
• each adhesive strip comprises several surface electrodes. This arrangement facilitates the installation of the electrodes on the patient, the installation of a strip comprising several electrodes being simpler than that of the electrodes one by one.
• the device comprises a vest or a jacket comprising the plurality of EL measurement electrodes.
• the vest is intended to be put on by the patient. This arrangement allows rapid installation of the device on the patient.
• the device comprises at least 14 electrodes.
• the device comprises means for measuring the signal of each channel Vi. More specifically, the measuring means is configured to measure one the electric potential of each of the channels Vi.
• the measuring means can for example be an acquisition card.
• the acquisition card may include an input for collecting an electrical signal, and an analog-digital converter for digitizing the acquired signal.
• the digitized signal is then transmitted to a computer.
• the digitized signal can for example be transmitted to a computer which performs the processing steps on the signal.
• the device comprises a calculation means.
• the calculating means records the measurements of the channels Vi provided by the measuring means.
• the computer then processes this data.
• the calculator calculates the value of each descriptor D, of the set ⁇ Dk ⁇ for each patient of each set ENSi and ENS2. This calculation is made from the measurements of the channels Vi. The calculation is carried out in accordance with the predefined zone Zi, the type of signal Ti, the signal marker MSi and the statistical modality STi selected for the descriptor Di in question.
• the computer then performs the generation steps GEN1 of the first vector V1 as described above. It also performs the steps of generation GEN2 of the descriptor vector Vd, of quantification QUAN of the correlation, of first classification CLAS1, of second classification CLAS2, and of selection SEL described in the method according to the invention. It is capable of carrying out all the other calculation steps described in this application.
• the device comprises a device for detecting the patient's breathing phases.
• a device for detecting the patient's breathing phases detects when the patient is in the exhalation phase or “flat” breathing phase. It also detects when the patient is in the inspiration phase. Breathing tends to interfere with measurements made at the EL electrodes. This is particularly the case during the inspiration phases during which heartbeats and their measurement can be affected.
• the measurement of the potential of each channel V is carried out during the expiration phase. This arrangement makes it possible to avoid the disturbances caused by a measurement during the inspiration phases.
• the breathing phase detection device can be connected to the computer. Alternatively, it is connected to the signal measurement means of each channel Vi.
• the breathing phase detection device is a plethysmography belt.
• the plethysmography belt is a convenient way to perform this type of detection. Nomenclature :
• DISPO arrangement of a plurality of electrodes
• ENR Recording of a plurality of electrical activities

Landscapes

• Health & Medical Sciences (AREA)
• Life Sciences & Earth Sciences (AREA)
• Engineering & Computer Science (AREA)
• Medical Informatics (AREA)
• Public Health (AREA)
• General Health & Medical Sciences (AREA)
• Biophysics (AREA)
• Cardiology (AREA)
• Pathology (AREA)
• Biomedical Technology (AREA)
• Physics & Mathematics (AREA)
• Veterinary Medicine (AREA)
• Animal Behavior & Ethology (AREA)
• Surgery (AREA)
• Molecular Biology (AREA)
• Heart & Thoracic Surgery (AREA)
• Primary Health Care (AREA)
• Epidemiology (AREA)
• Artificial Intelligence (AREA)
• Data Mining & Analysis (AREA)
• Physical Education & Sports Medicine (AREA)
• Databases & Information Systems (AREA)
• Physiology (AREA)
• Psychiatry (AREA)
• Signal Processing (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Mathematical Physics (AREA)
• Fuzzy Systems (AREA)
• Evolutionary Computation (AREA)
• Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
• Complex Calculations (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=76283892

Family Applications (1)

Country Status (4)

Family Cites Families (2)

• 2021
  • 2021-03-25 FR FR2103047A patent/FR3121029B1/fr active Active
• 2022
  • 2022-03-25 US US18/550,630 patent/US12569182B2/en active Active
  • 2022-03-25 WO PCT/EP2022/057991 patent/WO2022200602A1/fr not_active Ceased
  • 2022-03-25 EP EP22718205.2A patent/EP4315352A1/de active Pending

Also Published As

Similar Documents

Legal Events