Classifications

• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
  • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
  • A61B5/1102—Ballistocardiography
• • 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
  • A61B5/7221—Determining signal validity, reliability or quality
• • 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
  • 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
  • A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• • 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
  • G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
• • 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
  • A61B2562/02—Details of sensors specially adapted for in-vivo measurements
  • A61B2562/0204—Acoustic sensors
• • 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
  • A61B2562/02—Details of sensors specially adapted for in-vivo measurements
  • A61B2562/0247—Pressure sensors
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/0002—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network
  • A61B5/0015—Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network characterised by features of the telemetry system
  • A61B5/002—Monitoring the patient using a local or closed circuit, e.g. in a room or building
• • A—HUMAN NECESSITIES
  • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
  • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
  • A61B5/00—Measuring for diagnostic purposes; Identification of persons
  • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
  • A61B5/024—Detecting, measuring or recording pulse rate or heart rate
• • 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
  • A61B5/6887—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient mounted on external non-worn devices, e.g. non-medical devices
  • A61B5/6892—Mats
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N3/00—Computing arrangements based on biological models
  • G06N3/02—Neural networks
  • G06N3/04—Architecture, e.g. interconnection topology
  • G06N3/045—Combinations of networks

Definitions

• the present disclosure relates to the field of methods for monitoring the cardiac state of an individual during sleep.
• JP2013500757 A An example of a method for calculating the frequency of the heartbeat is illustrated by EP3456256.
• a method for determining an individual's cardiac state for supplying a characteristic of a cardiac signal comprising the steps following:
• lel provide a characteristic of the signal representative of cardiac activity, based on steps /a/ to /d/.
• the method comprises the following step: /b/ determining a relevance of at least a part of the at least one signal obtained, in order to determine whether the signal obtained can be used to determine a characteristic of the cardiac signal. Step /b/ can be done at any time.
• step /b/ comprises an automatic learning step, implemented by an artificial intelligence.
• An automatic learning step makes it possible to implement the principle of learning by experience, that is to say to determine, on the basis of signals obtained previously, whether a signal or part of a signal is relevant and should be considered in determining an individual's cardiac condition.
• the at least one signal representative of cardiac activity is a pressure signal and the step Here comprises the determination of a ballistocardiogram.
• Ballistocardiography is an exploratory technique of minute movements of the body caused by cardiac contraction.
• attention and analysis are focused on the minute movements of the body caused by a cardiac contraction.
• Characteristics of the cardiac state can be found in the ballistocardiogram, and be analyzed during the implementation of step ldi.
• a ballistocardiogram can be obtained in a completely non-invasive way, even without the knowledge of the individual.
• a ballistocardiogram can be obtained without direct contact with the individual.
• the step Here comprises the calculation of a self-similarity.
• the self-similarity makes it possible to analyze and classify the at least part of the at least one signal obtained, and this without the use of an external reference signal.
• the self-similarity is interested in all or part of the extrema of the signal obtained, and one calculates, according to a time lag between a signal obtained and its copy (also called reference signal hereafter) shifted in time, a score representative of the sum of the distances between each extremum of the signal obtained and the extremum closest to its copy.
• the step here comprises: calculating, for a portion of the representative signal, a set of similarity scores, said set of similarity scores associating similarity score values with respective time offsets.
• This set of similarity scores can take the form of a score function, whose variable is the time lag.
• similarity scores are assigned to the different time offsets.
• the set of similarity scores can be calculated by comparing the portion with a plurality of recopies of the portion to which a respective time shift has been applied. A similarity score value is then assigned for each time lag, which generates the set of scores.
• This embodiment by shifting a copy is an implementation of self-similarity. Nevertheless, the calculation of similarity scores can be done with other methods.
• the step here comprises: stacking a plurality of sets of similarity scores calculated for a plurality of respective portions, in order to form a stack of self-similarity scores of the representative signal.
• the stack therefore associates similarity score values of the representative signal with time shifts and portions of the representative signal. Data relating to the similarity of the representative signal are thus obtained over time.
• This stack is then analyzed in step /d/ (it can in particular serve as input to a processing algorithm).
• step /d/ is implemented at least in part by an artificial intelligence.
• the artificial intelligence includes an artificial neural network.
• An artificial neural network implements the principle of learning by experience, which makes it possible to optimize the results obtained in the analysis according to step /d/ according to results obtained previously.
• the artificial neural network comprises a two-dimensional convolutional neural network, i.e. a neural network with at least one two-dimensional convolutional layer.
• This type of neural network is specially adapted for image analysis, in particular for the detection and analysis of certain patterns and shapes in images.
• an image is made from a juxtaposition of several self-similarity results and the image in question constitutes the input to the neural network.
• this type of network uses reduced memory and computing power, allowing it to be embedded in a device of reduced size and inexpensive design.
• At least a portion of the steps 1 to 1 are performed repeatedly, and a feature is provided every 1 minute to 5 minutes.
• the signals are monitored over a medium or long term with regard to the cardiac period.
• step /b/ comprises:
• This step makes it possible to check whether the conditions necessary to provide a characteristic are fulfilled, for example whether the individual is present and positioned correctly.
• the characterization of stage le/ includes an irregular heartbeat or atrial fibrillation.
• the characterization of stage le/ may include an absence of irregularity.
• the characteristic of the representative signal may relate to the regularity or irregularity of the signal over time.
• the method makes it possible to provide a relevant and easy-to-understand characterization for an individual without medical training, of the "No particularity detected” or “Atrial fibrillation detected” type: Please contact your attending physician. ".
• signals can be measured continuously and in a state of rest of the individual, while minimizing disturbances of the measurements, that is to say by avoiding for example tensions or stress that the individual might feel during an examination in a doctor's office.
• Another aspect of the disclosure comprises a processing unit (known as a first processing unit) for determining a characteristic of an individual's cardiac condition.
• the processing unit is configured to implement a method for determining a characteristic of a cardiac state of an individual as described previously.
• the processing unit may comprise a memory storing instructions corresponding to the methods described above and a control circuit capable of executing said instructions.
• the unit can thus in particular implement the following steps:
• /d/ analyzing (204) a temporal evolution of at least a part of the at least one processed signal; lel providing (205) a signal characteristic representative of cardiac activity based on steps /a/ through /d/.
• Another aspect of the disclosure comprises a device for determining a characteristic of a cardiac state of an individual, the device comprising at least one sensor configured to measure at least one signal representative of a cardiac activity of the individual repeatedly, and the processing unit configured to implement the method for determining a characteristic, the device being configured to determine the cardiac state of the individual without being in physical contact with the individual .
• the device comprising at least one sensor configured to measure at least one signal representative of a cardiac activity of the individual repeatedly, and the processing unit configured to implement the method for determining a characteristic, the device being configured to determine the cardiac state of the individual without being in physical contact with the individual .
• This device makes it possible to implement the method for determining a characteristic of a cardiac state of an individual.
• the individual does not wear electrodes or other sensors on his body during the implementation of the method. Totally non-invasive, the device is transparent to use for an individual, and can even be used in certain pathological cases even without the knowledge of the individual. Useful signals can be obtained without direct contact with the individual.
• the device comprises a second processing unit remote from the device and configured to receive by the first processing unit, after implementation of the method by the first processing unit, data representative of at least a part of the at least one measured signal and/or of the at least one processed signal and/or of the characteristic.
• At least part of the measured and/or processed data and/or any other data relating to the characteristic can be transmitted to the second processing unit, for saving and/or viewing and/or later use. .
• the second processing unit comprises a “smartphone” or a tablet.
• a "smartphone” or a tablet makes it possible to save data, for example on the "smartphone” or tablet, a cloud or any other storage unit, to view data or to share data with a qualified person such as than a doctor.
• Another aspect of the disclosure comprises a computer program product comprising instructions for implementing the method of the present invention, when the computer program product is executed by a processor.
• This program can use any programming language (for example, an object language or other), and be in the form of an interpretable source code, a partially compiled code or a completely compiled.
• FIG. 7 described in detail below can form the first processing unit configured for the implementation of the method for determining a characteristic of a cardiac state of the individual.
• FIG. 1 shows a device for implementing a method for determining a characteristic of a heart condition of an individual.
• FIG. 2 shows a flowchart of a method of determining a characteristic of a heart condition.
• FIG. 3 shows the result of a two-dimensional self-similarity calculation based on a signal measured and representative of an individual's cardiac activity, which indicates sinus rhythm or atrial fibrillation respectively.
• FIG. 4 shows the result of a three-dimensional self-similarity calculation based on a signal measured and representative of an individual's cardiac activity, which indicates a sinus rhythm.
• FIG. 5 shows the result of a three-dimensional self-similarity calculation from a signal measured and representative of an individual's cardiac activity, which indicates atrial fibrillation.
• FIG. 6 shows an illustration of an artificial neural network intended for the implementation of part of the method for determining a characteristic of a cardiac state of an individual.
• FIG. 7 is a schematic view of a processing unit configured to implement the method for determining a characteristic of a heart condition of an individual.
• Figure 1 describes a device 101 for determining a characteristic of an individual's heart condition.
• the device 101 can be installed in a bed, for example under a mattress, in order to monitor the individual present on the mattress during his sleep, and to identify heart problems such as atrial fibrillation.
• the device 101 may comprise a strip of fabric 102 forming a sheath.
• the device 101 can comprise at least one sensor 107, 108 configured to measure at least one signal representative of cardiac activity, such as a pressure sensor 107 or an acoustic sensor 108.
• the strip of fabric 102 may contain a pneumatic chamber acting as a pressure sensor 107.
• the pressure sensor 107 may comprise at least one piezoelectric sensor, enclosed by the strip of fabric 102.
• the pressure sensor 107 comprises a pneumatic chamber
• the weight and the movements of the individual present on the bed act on the pneumatic chamber, which causes pressure variations.
• the pressure signal may for example be a micro-vibration of the individual's body, generated by the beating of the heart.
• the strip of fabric 102 can enclose a casing 104.
• the box 104 may include electronic means connected to a connection cable 103 comprising a wire and a USB connector in order to be able to power the device 101 electrically.
• the device may include a plug intended to be plugged into a socket current.
• the device may also include a rechargeable battery.
• the housing 104 may include a pressure converter, configured to transform a pressure signal, detected using the pressure sensor 107, into a first voltage.
• the housing 104 may include a microphone, acting as an acoustic sensor 108.
• the acoustic sensor 108 can measure sounds from the environment, in particular breathing or snoring coming from the individual, and convert them into a second voltage.
• the acoustic sensor 108 and the pressure converter can be electronically connected to a first processing unit 106 which is located in the housing 104.
• the first processing unit 106 can be configured to implement all or part of the method for determining a characteristic of a heart condition of the individual. This process is described in detail in Figure 2.
• the device comprises a screen.
• This screen can be configured to display characteristic data such as “No features detected” or “Atrial fibrillation detected: Please contact your treating physician. ".
• the first processing unit can be further configured to communicate, as represented by the dotted arrow 109 in FIG. 1, with a second processing unit.
• the second processing unit can be configured to receive, by the first processing unit, at least part of the measured data (which represent the at least one measured signal) and/or processed and/or any other data relating to the feature, for backup and/or viewing and/or other future use.
• the second processing unit can be remote from the device, that is to say not be physically linked to the device.
• the second processing unit can be or comprise a “smartphone” 105 or a tablet, which makes it possible to save data, for example on the “smartphone” or the tablet or in a cloud or another external device, view data or share data with a qualified person such as a doctor.
• a suitable application can be installed on the “smartphone”/tablet.
• the sharing of data with a doctor can be done at the customer's initiative, or in an automated way, in particular if a characteristic relating to an irregular heartbeat is identified by the device.
• the pneumatic chamber can be inflated before its use.
• the device 101 can either be stored, rolled up, or folded so that it is compact when not in use.
• the device 101 can have a rectangular shape, with a length LX between 50 mm and 800 mm, a width LY between 10 mm and 400 mm, and a thickness TZ of less than 20 mm.
• the shape of the pneumatic chamber can be substantially the same as the shape of the strip of fabric 102.
• FIG. 2 shows the flowchart of a method for determining a characteristic of an individual's heart condition. This method can be implemented by the device 101 shown in FIG. 1.
• This device can comprise a first processing unit 106.
• At least part of the method can be implemented during the individual's sleep.
• the individual can be a human being, and the device 101 can be installed in a bed, for example under the mattress, in order to monitor the human being during his sleep.
• the individual can be an animal, for example a dog, and the method can be implemented while the dog is sleeping in a basket.
• a first step 201 at least one signal representative of the individual's cardiac activity can be measured. This measurement can be made repeatedly, for example continuously for a certain period of time, between several minutes and several hours, or even a day.
• This signal can be a pressure signal and/or an acoustic signal, detected by at least one pressure sensor 107 and/or an acoustic sensor 108 included in the device 101.
• the acoustic signal can be representative of breathing or snoring of the individual.
• the pressure signal can be representative of minute movements of the body caused by a cardiac contraction made by the individual, but also of other movements when the individual moves in his sleep.
• step 201 can include obtaining a signal representative of the cardiac activity that is being measured; in particular, the device which acquires the signal may be different from the device which implements steps 202, 203, 204, 205 of the method of the description.
• the measured signal can be used to determine the relevance of at least part of the measured signal.
• Step 202 can comprise an automatic learning step, implemented by an artificial intelligence such as an artificial neural network.
• At least one measured signal representative of cardiac activity of the individual can be analyzed.
• At least one additional signal representative of a movement performed by the individual, of breathing by the individual or of pressure exerted by the individual can be measured and analyzed.
• the evolution of the amplitude of a pressure signal can be observed, in particular the evolution from one maximum of the pressure signal to another.
• a macroscopic movement made by the individual can saturate the reading of the signal, that is to say that if the individual moves during his sleep, the signal measured by the pressure sensor configured to detect minute pressure signals can saturate.
• the result of step 202 for this part of the signal may be an error message, and a signal representative of cardiac activity cannot be extracted. No characteristic of the signal representative of cardiac activity can be provided in this case.
• a result of step 202 may be that the individual is not on the bed or is not positioned correctly, and the characteristic cannot be provided accordingly.
• a third step 203 at least part of the measured signal is processed.
• the third step 203 can comprise the determination of a ballistocardiogram.
• a ballistocardiogram is a written record of minute body movements caused by a heart contraction. Characteristics of the heart condition can be encoded in a ballistocardiogram.
• the third step 203 can include signal similarity calculations, which can make it possible to obtain information as to the rhythm and the regularity of the cardiac signal.
• the third step 203 comprises the calculation of a self-similarity.
• the implementation and results of a self-similarity calculation are described in detail in Figures 3 to 5.
• the third step can comprise the calculation of an autocorrelation.
• a temporal evolution of at least part of the at least one processed signal is analyzed.
• similarity score calculations can serve as inputs to a processing algorithm.
• the fourth step 204 is implemented at least in part by an artificial intelligence, which allows great flexibility in the implementation of the method.
• the method can succeed and provide a feature for different shapes or patterns in the signals/data, thanks to the ability of the data classification artificial intelligence to be analyzed.
• the artificial intelligence comprises an artificial neural network such as a two-dimensional convolutional neural network 301 which comprises at least one hidden layer with two-dimensional convolution, as explained in the description relating to FIG. 6.
• a two-dimensional convolutional neural network is specially adapted for image analysis, in particular for the detection and analysis of certain patterns in images.
• An artificial neural network is an algorithm that allows a processing unit 106 executing this algorithm to learn from new data.
• An artificial neural network can learn by feeding the algorithm with data including tagged processed records of pressure variation and/or sounds.
• a reference database can be created and improved over time.
• a qualified person such as a doctor can label the recordings of sounds and pressure variation in association with the profile.
• step 204 is described in detail in Figure 6.
• a characteristic based on the first 201 to fourth 204 steps can be provided.
• the characteristic may relate to a temporal regularity or a temporal irregularity of the signal.
• the fifth step 205 may comprise the identification of atrial fibrillation or the identification of an absence of abnormality relating to the cardiac activity.
• the second step 202 can be implemented at any time during the implementation of the method, for example directly after the signal measurement, or at the same time as the third or the fourth step.
• a characteristic such as “No particularities detected” or “Atrial fibrillation detected: Please contact your treating physician. is provided only if it is determined during the second step 202 that the measured signal is relevant. Otherwise, the process may provide an error message.
• the method notifies the user of the characteristic of the measured signal only if the measured signal is usable.
• first step 201 it is not necessary for the first step 201 to be completed to trigger the implementation of the second step 202.
• the second 202, third 203 or fourth 204 steps can be implemented at least in part when measuring signals in the first step 201.
• part of the method in particular any one of steps 202, 203, 204 and 205, can be implemented by a second processing unit remote from the device, such as a smartphone.
• At least a portion of the steps I to Ic is performed repeatedly, and a feature is provided every 1 minute to 5 minutes.
• Figure 3 shows the two-dimensional result of the calculation of a self-similarity.
• the self-similarity calculation can be performed during the implementation of the third step 203 of the method described in FIG. 2, from the signal measured during the first step 201.
• a band-pass filter can be applied to the at least one measured signal to eliminate high-frequency background noise. Additionally, the band pass filter can be configured to remove low frequency components representative of the individual's breathing.
• a measured signal can be used to create mirrors (i.e. reference signals) of the measured signal, which in turn can be used to classify subsequently measured signals, as explained in detail below.
• sets of reference points can be obtained from the measured signal or from the signal processed in the optional preliminary sub-step.
• the signal is sampled, and extrema can be identified in the signal.
• the extrema can be maxima and/or minima of the signal.
• the reference points can be obtained by applying a time shift to the extrema identified to represent different frequencies of the signal.
• the time shift is on the order of one or more seconds.
• the sets of reference points may include data that correspond to expected positions for a signal having a certain frequency.
• the set of reference points can cover a range between 35 and 110 beats per minute.
• the different sets of reference points can be saved for the analysis of the measured signal.
• only the identified extrema are saved, and the remaining part of the signal is erased.
• a plurality of extrema are identified in the measured signal.
• the self-similarity is interested in all or part of the extrema of the measured signal, and one calculates, in a third sub-step, according to a time shift between the measured signal and its copy, a score representative of distances between each extremum of the measured signal and the extremum closest to its copy.
• the two-dimensional result of the calculation of a self-similarity is a score that can be assigned to each reference point, to indicate how close the reference point is to an extremum, so that the score closest high is assigned to the benchmarks with the smallest deviation.
• a self-similarity is an inverse function of distances between each extremum of the measured signal and the closest extremum of its copy.
• the time lag which constitutes a variable of the self-similarity function can be bounded, that is to say vary within a determined range.
• the comparison of the extrema to the reference points can include a frequency comparison and/or an amplitude comparison.
• the total score for a set of reference points can be obtained by calculating the sum of the scores assigned to each reference point in a set. [0140] In another variant, the total score for a set of reference points can be the maximum of the scores for a set of reference points.
• any other suitable method can be used to compare the set of reference points which correspond to the identified extrema.
• the set of reference points which correspond the most to the extrema is identified to determine the frequency of the signal to be analyzed.
• the time duration of the signal to be analyzed must be long enough to include at least two extrema. For example, if the signal to be analyzed represents cardiac activity with 75 beats per minute, the difference between two maxima is 0.8 seconds and consequently the sampling time should be at least 1.6 seconds.
• a curve of self-similarity as a function of the time lag (also called delay or also called “lag") is thus obtained every 1.6 seconds.
• This similarity curve thus forms a set of similarity scores, in which each offset is associated with a similarity score (similarity score as a function of the offset therefore).
• FIG. 3 a plurality of curves such as shown in FIG. 3 are juxtaposed or stacked, so as to form a three-dimensional representation of the temporal evolution of the sets of scores of FIG. 2.
• This representation three-dimensional thus forms a stack of similarity scores of the representative signal, in which a score is associated with each portion and each offset (score as a function of the portion and of the offset therefore).
• the abscissa of figure 3 corresponds to the ordinate of figures 4 and 5.
• Figure 4 shows the case of sinus rhythm
• Figure 5 the case of atrial fibrillation
• Such images typically comprise between 50 and 200 points on each axis, therefore between 0.0025 and 0.04 megapixels. This relatively small image size allows to minimize the computational costs for the analysis of such images.
• the maximum of each of the two-dimensional curves forms the dark curve (FIG. 4). More generally, we are interested in the dark areas in the image and the continuity of these dark areas. In other words, in a case of a normal heartbeat, there is a roughly horizontal dark band, not necessarily rectilinear or even not necessarily continuous.
• the periodicity of the heartbeat mainly varies between 0.8 seconds and 1.1 seconds.
• the similarity sets and stacking can be calculated with self-similarity, as described above, or else with autocorrelation, a sliding Fourier transform or even with a wavelet transform.
• the images can be processed by a “conventional” algorithm and not by a neural network.
• a “conventional” algorithm can be based for example on the recognition of the dark zone in the images. The analysis of the images shown in Figures 4 and 5 is explained below with the aid of Figure 6.
• FIG. 6 shows a two-dimensional convolutional neural network (called CNN hereafter), intended for the implementation of step 204 of the method described in FIG. 2.
• CNN convolutional neural network
• a CNN is an algorithm which allows a processing unit 106 executing this algorithm to implement the principle of learning by experience, that is to say to learn by analyzing examples, in this case, images according to FIGS. 4 or 5, calculated from signals representative of cardiac activity of an individual in a repeated manner.
• a large number of images of the same category can be presented to the CNN, for example self-similarity calculation results, calculated from signals representative of cardiac activity of an adult who suffers from atrial fibrillation.
• the processing unit 106 thus learns to recognize this type of recurring pattern when an unknown image is presented to it.
• the CNN can continue its learning during the implementation of the method.
• step 204 of the method according to FIG. 2 can be adapted and optimized according to the analysis of previously determined images.
• the operation of a CNN is based on a large number of processors operating in parallel and organized in layers.
• a CNN can include an input layer inL, one or more hidden layers hidL, and an output layer outL, arranged one after the other.
• a posterior layer receives as input the result of an earlier layer.
• Each layer can comprise several elements.
• the input element represents images to be analyzed.
• An artificial neuron represented by an arrow, represents a transfer function which transforms the activation of the elements of a layer into the activation of the elements of the next layer, according to rules that can change, following the principle of automatic learning.
• a CNN makes it possible to test functional hypotheses.
• Each arrow from a layer represents a tested hypothesis.
• each image can be analyzed with a sliding filter, and certain patterns can be sought in the image, such as a signal whose “lag” is between 0.8 seconds and 1.1 seconds.
• a plurality of hypotheses can be tested. The same assumptions are tested for each of the elements in a layer.
• Each of the elements of a layer receives all the results of a single hypothesis for all the elements of the previous layer.
• a first hidden layer hidL receives information outputs from the first layer, i.e. signals processed/analyzed according to a set of assumptions.
• the output of the second layer can be guided to a second hidden layer.
• the number of hidden layers, as well as the number of elements in each layer is not limited, and can be adapted according to the problem to be solved.
• the hidden layers in FIG. 6 comprise four elements each.
• the elements of a hidden layer can correspond for example to the different patterns that can be identified in an image.
• the last layer outL produces the results of the system.
• the result may be, for example, “No particularity detected” or “Atrial fibrillation detected: Please contact your treating physician. ".
• the last outL layer of the CNN can provide three different results: “atrial fibrillation” (AF), “sinus rhythm” (RS) or “error” (err).
• At least part of the steps /b/ to le/ are implemented repeatedly, and the characteristic of the signal representative of the cardiac activity is provided every 1 minute to 5 minutes.
• Figure 7 shows a processing unit 106 suitable for implementing the method described in Figure 2.
• This processing unit may be part of the device described in Figure 1.
• the processing unit comprises a memory 110 for storing instructions allowing the implementation of at least part of the method, the data received, and temporary data for carrying out the various steps and operations of the method as described above.
• the processing unit further comprises a control circuit 111.
• This circuit can be, for example:
• processor capable of interpreting instructions in the form of a computer program
• a programmable electronic chip such as an FPGA chip for "Field-Programmable Gate Array” in English, such as a SOC for "System On Chip” in English or as an ASIC for "Application Specifies Integrated Circuit” in English.
• SOCs or systems on a chip are embedded systems that integrate all the components of an electronic system into a single chip.
• An ASIC is a specialized electronic circuit that groups functionalities tailored to a given application. ASICs are usually configured during manufacture and can only be simulated by a user of the processing unit.
• FPGA-type programmable logic circuits are electronic circuits that can be reconfigured by the user of the processing unit.
• the processing unit comprises an input interface for receiving messages or instructions, and an output interface for communicating with the at least one sensor 107, 108.
• the processing unit is integrated in the device according to figure 1.
• the processing unit 106 can be a computer, a computer network, an electronic component, or another device comprising a processor operationally coupled to a memory, as well as, according to the chosen embodiment, a data storage unit, and other associated hardware such as a network interface and a media drive for reading removable storage media and writing to such media not shown in Figure 7.
• the removable storage medium can be, for example, a CD compact disc, a DVD digital video/versatile disc, a flash disc, a USB key, etc.
• the memory 110 contains instructions which, when executed by the control circuit 111, cause this control circuit to perform or control the input interfaces, output interface, data storage in the memory 110 and/or data processing and examples of implementation of the method described in FIG.
• the control circuit 111 can be a component implementing the control of the processing unit 106.
• processing unit 106 can be implemented in software form, in which case it takes the form of a program executable by a processor, or in hardware form, or "hardware", such as an integrated circuit. specific application ASIC, a system on chip SOC, or in the form of a combination of hardware and software elements, for example a software program intended to be loaded and executed on an electronic component described above such as FPGA, processor.
• the processing unit 106 can also use hybrid architectures, for example architectures based on a CPU+FPGA, a GPU for “Graphics Processing Unit” or an MPPA for “Multi-Purpose Processor Array”.
• the processing unit can control at least part of the components of the device shown in figure 1.
• the processing unit can control the pressure sensor 107 and the acoustic sensor 108.
• the processing unit can comprise storage hardware for storing at least part of the measured signal and/or of the processed signal and/or of the characteristic.
• the device according to FIG. 1 can be coupled 109 wirelessly, for example via bluetoothTM, Wi-Fi or a cellular network, to one or more devices, such as a second processing unit such as a "smartphone". 105 or a tablet or a remote server, without excluding the direct interfacing of a telemedicine in a remote server.
• a second processing unit such as a "smartphone”.
• the present disclosure makes it possible to determine a heart condition of an individual for the provision of a characteristic.

Landscapes

• Health & Medical Sciences (AREA)
• Engineering & Computer Science (AREA)
• Life Sciences & Earth Sciences (AREA)
• Public Health (AREA)
• Medical Informatics (AREA)
• Biomedical Technology (AREA)
• Pathology (AREA)
• Artificial Intelligence (AREA)
• General Health & Medical Sciences (AREA)
• Physics & Mathematics (AREA)
• Animal Behavior & Ethology (AREA)
• Biophysics (AREA)
• Heart & Thoracic Surgery (AREA)
• Molecular Biology (AREA)
• Surgery (AREA)
• Physiology (AREA)
• Veterinary Medicine (AREA)
• Signal Processing (AREA)
• Psychiatry (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Dentistry (AREA)
• Cardiology (AREA)
• Oral & Maxillofacial Surgery (AREA)
• Evolutionary Computation (AREA)
• Fuzzy Systems (AREA)
• Mathematical Physics (AREA)
• Data Mining & Analysis (AREA)
• Databases & Information Systems (AREA)
• Epidemiology (AREA)
• Primary Health Care (AREA)
• Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
• Measuring And Recording Apparatus For Diagnosis (AREA)
• Investigating Or Analysing Biological Materials (AREA)

Applications Claiming Priority (2)

Publications (1)

Family

ID=76034682

Family Applications (1)

Country Status (3)

Family Cites Families (8)

• 2021
  • 2021-01-07 FR FR2100114A patent/FR3118574B1/fr active Active
• 2022
  • 2022-01-07 EP EP22704404.7A patent/EP4274477A1/de active Pending
  • 2022-01-07 WO PCT/FR2022/050045 patent/WO2022148938A1/fr unknown

Also Published As

Similar Documents

Legal Events