SYSTEM FOR PREVENTING VASCULAR PLAQUE RUPTURE OR DETACHMENT THAT LEADS TO A STROKE AND / OR IS CAPABLE OF PREVENTING VASCULAR THROMBOSIS

Description

Domaine technique

[0001] This disclosure relates to a system for predicting the rupture or detachment of vascular plaques in the carotid artery or supra-aortic trunks that could lead to a stroke.

[0002] This disclosure relates more specifically to a system and method for continuously monitoring the stability of vascular plaques in the carotid artery or supra-aortic trunks of a patient and predicting the occurrence of a stroke due to plaque rupture. The system is also adapted for continuously monitoring the biomechanical properties of the internal jugular vein crossing the carotid bifurcation to predict the occurrence of vascular thrombosis. Technique antérieure

[0003] A stroke (cerebrovascular accident, CVA) is linked to a sudden interruption of blood flow to all or part of the brain.

[0004] The brain is supplied with blood by the internal carotid artery and the network of supra-aortic trunks, providing the oxygen necessary for its function. A process known as atherosclerosis causes a thickening and hardening of the arteries through the accumulation of lipids, complex carbohydrates, blood, and calcium deposits on the inner walls of the arteries, leading to the progressive formation of one or more atherosclerotic plaques. There are also other causes of vascular plaque formation. This process is accelerated by identified risk factors, namely high blood pressure, diabetes, obesity, and smoking. The presence of atherosclerotic plaques thus causes a reduction in the diameter of the artery, a condition known as stenosis, a disease more commonly referred to as atherosclerosis. This plaque can also be of post-radiation or post-inflammatory origin, or due to a collagen disease.In all these cases, the risk of rupture exists. The presence of vascular plaque (due to atherosclerosis or other causes) can be detected by various techniques, such as vascular Doppler ultrasound, optical coherence tomography, computed tomography, or magnetic resonance imaging.

[0005] A dramatic consequence of plaque rupture in the carotid artery or supra-aortic trunks is the migration of plaque components into the bloodstream, potentially obstructing a cerebral artery and causing an ischemic stroke. Another, less common, potential risk associated with atherosclerotic plaque is plaque rupture, which can lead to occlusion of the internal carotid artery. This can cause a low-flow hemodynamic stroke if the circle of Willis is not compensating. In both cases, the dramatic consequences are correlated with the detachment or rupture of part or all of the plaque. Rupture of part or all of the plaque is most often followed by a more or less calcified fibrin-thrombosis embolus that can obstruct the cerebral artery.

[0006] Current assessment methods, based solely on the degree of stenosis, are far from sufficient and cannot predict plaque rupture or the true risk of stroke. Therefore, it is crucial to assess the risk of plaque rupture from atherosclerotic causes or other sources to accurately determine an individual's risk of stroke.

[0007] Currently, the assessment of the risk of vascular plaque rupture is primarily based on plaque morphology and tissue composition. Indeed, morphological parameters such as fibrous cap thickness, necrotic body size, and stenosis (obstruction of the arterial lumen) are considered factors influencing plaque rupture. Ruptured plaques are characterized by a significant lipid core, a thin fibrous cap, and intraplaque hemorrhage. However, these indicators do not reliably predict the risk of vascular plaque rupture.

[0008] US2017 / 340393-A1, US2020 / 279656-A1, and US2020 / 015758-A1 demonstrate vascular plaque rupture prediction systems using trained artificial intelligence. US2003 / 125637-A1, US2004 / 102722-A1, KIPS ET AL: "Identifying the vulnerable plaque: A review of invasive and non-invasive imaging modalities", vol. 2, no. 1, January 25, 2008, pages 21-34, as well as US2003 / 220556-A1, and US2012 / 065514-A1, demonstrate other vascular plaque rupture prediction systems.

[0009] One purpose of this disclosure is to propose a system that allows continuous monitoring of changes in the integrity of the vascular plaque in order to predict the occurrence of a stroke caused by plaque rupture or detachment.

[0010] Another purpose of this disclosure is to propose a system and method for predicting the occurrence of a stroke that is not very burdensome for the patient and easy for the practitioner to use.

[0011] Another purpose of this disclosure is to propose a system and method that allows continuous monitoring of the evolution of the biomechanical properties of an internal jugular vein in order to predict the risk of vascular thrombosis. Résumé

[0012] This disclosure improves the situation.

[0013] A system is proposed for predicting at least partial rupture or detachment of a vascular plaque that could lead to a stroke, said vascular plaque being present on an arterial wall chosen from a carotid wall and a supra-aortic trunk wall, said system comprising: a monitoring device suitable for placement near the vascular plaque, said device comprising at least one temperature sensor configured to measure the temperature of the vascular plaque, at least one vibration sensor configured to measure mechanical waves propagated in said arterial wall, at least one vascular plaque motion sensor configured to measure vascular plaque movements, a memory suitable for storing signals transmitted by said sensors, a communication interface, a power source configured to supply said sensors and the communication interface, a computing unit suitable for communicating with the monitoring device and configured to analyze measurements from the vibration sensor,The temperature sensor and the vascular plaque movement sensor from the monitoring device are used by artificial intelligence trained to detect if there is a risk of at least partial rupture or detachment of the vascular plaque.

[0014] According to one embodiment, the monitoring device further includes at least one electrical sensor configured to measure a parameter related to the electrical impedance of the vascular plaque.

[0015] According to another embodiment, the monitoring device further includes at least one acoustic wave sensor configured to measure acoustic waves from the arterial wall and / or vascular plaque.

[0016] Thanks to the monitoring device attached directly to the patient, it is possible to monitor the evolution of the integrity of the vascular plaques in real time.

[0017] The system makes it possible to predict the risk of a stroke without requiring complex and often invasive examinations of the patient.

[0018] The features described in the following paragraphs may optionally be implemented, independently of each other or in combination with each other:

[0019] The monitoring device comes in the form of a patch suitable for being stuck onto an external surface of a patient's skin.

[0020] The monitoring device comes in the form of a subcutaneous implant suitable for insertion under a patient's skin.

[0021] The temperature sensor is a sensor of thermal waves emitted by the vascular plaque.

[0022] The vibration sensor includes an accelerometer.

[0023] The vibration sensor includes a 3-axis accelerometer and a 3-axis gyroscope.

[0024] The vascular plaque movement sensor is an ultrasound probe.

[0025] The power source is an induction-rechargeable battery.

[0026] The communication interface is chosen from either a short-range radio interface or a near-field communication interface.

[0027] According to another embodiment, the system further comprises at least one mobile communication device suitable for communicating remotely with the monitoring device via said communication interface, which is a short-range interface, and for transmitting signals to the computing unit via a long-range communication interface belonging to the computing unit.

[0028] According to one embodiment, the monitoring device includes at least one temperature sensor and / or at least one motion sensor and / or at least one acoustic wave sensor, the measurement frequency of said sensors being set to emit and / or receive a signal from an internal jugular vein.

[0029] According to another aspect, a method is proposed to predict at least partial rupture or detachment of a vascular plaque that could lead to a stroke, said vascular plaque being present on an arterial wall selected from a carotid wall and a supra-aortic trunk wall of a patient, using a system as described above, the method comprising: continuously acquire over a specified period a plurality of signals representative of the evolution of the risk of rupture or detachment of the vascular plaque by means of the monitoring device placed near the vascular plaque, analyze the plurality of signals in a computing unit by an artificial intelligence trained to detect the risk of at least partial rupture or detachment of the vascular plaque.

[0030] In one embodiment, the artificial intelligence is a neural network and the process further includes a preliminary learning stage comprising: acquire a plurality of signals, called reference signals, from monitoring devices worn by a population whose risks of rupture or detachment of the vascular plaque are known, train the neural network with said reference signals until it converges. Brève description des dessins

[0031] Other features, details, and advantages will become apparent upon reading the detailed description below and analyzing the attached drawings, on which: Fig. 1 [ Fig. 1 ] There figure 1 is an illustration of a system for predicting a rupture or detachment of a vascular plaque that could lead to a stroke according to one embodiment. Fig. 2 [ Fig. 2 ] There figure 2 is an illustration of a prediction system according to another embodiment. Fig. 3 [ Fig. 3 ] There figure 3 shows the monitoring device of the prediction system positioned at the neck of a patient, near the internal carotid artery after the bifurcation of the common carotid artery, opposite the vascular plaque. Fig. 4 [ Fig. 4 ] There figure 4 shows an example of the implementation of the monitoring device fixed to an external surface of the skin, near the carotid artery. Fig. 5 [ Fig. 5 ] There figure 5 shows another example of the implementation of the monitoring device implanted under the skin, near the carotid artery. Fig. 6 [ Fig. 6 ] There figure 6 is a flowchart representing the prediction process implemented by the prediction system according to a mode of embodiment. Fig. 7 [ Fig. 7 ] There figure 7 is a flowchart representing the learning stage to train the neural network used by the computing unit. Fig. 8 [ Fig. 8 ] There figure 8 is a schematic illustration of the monitoring device further comprising an electrical sensor configured to measure a parameter related to the electrical impedance of the vascular plaque according to one embodiment. Fig. 9 [ Fig. 9 ] There figure 9 is a schematic illustration of the monitoring device further comprising an acoustic wave sensor configured to measure acoustic waves from the arterial wall and / or vascular plaque. Description des modes de réalisation

[0032] There figure 1 illustrates one embodiment of a system 1 for predicting the risk of rupture or detachment of a vascular plaque that could lead to a plaque rupture stroke. System 1 comprises: A monitoring device 2 (SURV) configured to be placed near a patient's carotid artery or supra-aortic trunks to measure signals representative of the mechanical stability of one or more vascular plaques (of atherosclerotic or other origins) formed on the inner wall of the carotid artery, and a computing unit 10 (CALC) adapted to communicate with the monitoring device 2 and configured to analyze the signals from the monitoring device 2 to detect if there is a risk of rupture of at least a portion of the vascular plaque. Typically, the computing unit 10 can be operated by a practitioner or a group of practitioners and can collect and process data from a large number of monitoring devices 2.

[0033] Monitoring device 2 may include: one or two vibration sensors 4 (SENSE 1), including one or more accelerometer(s), configured to measure mechanical waves that may be generated by the arterial pulse wave passing through the vascular plaque and the underlying walls, a temperature sensor 5 (SENSE 2) configured to measure the temperature of the vascular plaque, and a plaque motion sensor 6 (SENSE 3), including an ultrasonic transducer such as a piezoelectric crystal, configured to measure vascular plaque movements.

[0034] The monitoring device 2 also includes a memory 8 (MEM) capable of storing the signals transmitted by the sensors 4, 5, 6 and a communication interface 9 (INT 1). The communication interface 9 can be, for example: a short-range radio interface, in particular of the "Bluetooth", "Wifi" type or other, or a near-field communication interface, in particular of the NFC or RFID type.

[0035] The signals stored in memory 8 can be transmitted to the computing unit 10 via the communication interface 9.

[0036] The computing unit 10 is configured to analyze signals, notably using trained artificial intelligence, and to detect the occurrence of a stroke due to vascular plaque rupture. The computing unit 10 includes, for example, a central processing unit 12 (CPU) in which the trained neural network is stored. The computing unit 10 also includes a communication interface 11 (CPI 2) for receiving signals from the monitoring device 2. The communication interface 11 and the CPU 12 may optionally be located remotely from each other. The CPU 12 may optionally be a server.

[0037] The function of sensors 4, 5 and 6 is detailed below.

[0038] The rupture or detachment of the vascular plaque attached to the inner wall of the artery can be caused by a mechanical wave generated, for example, by the arterial pulse wave. The mechanical wave comprises a shear component and a compression component, and can be likened to a seismic wave. By measuring these mechanical waves, the vibration sensor 4 is capable of measuring a signal representative of the biomechanical properties of the vascular plaque and the underlying walls. The vibration sensor 4 may include at least one accelerometer. For example, the vibration sensor may include a 3-axis accelerometer and a 3-axis gyroscope. The data representing the signal measured by the vibration sensor 4 are stored in memory 8.

[0039] The temperature sensor 5 is capable of measuring a local temperature variation representative of plaque stability. Indeed, inflammation is one of the main characteristics of an unstable plaque. The intensity of the inflammatory process results in a temperature rise within the plaque. In one embodiment, the temperature sensor 5 can be configured to measure a thermal wave emitted by the plaque in the microwave frequency range. Such a sensor may include, for example, an antenna to detect these thermal waves emitted by the vascular plaque. The data representing the signal measured by the temperature sensor 5 are stored in memory 8.

[0040] The plate motion sensor 6 is a piezoelectric sensor configured to emit ultrasound to detect plate movements in real time and determine whether they are regular or irregular and chaotic, indicating arrhythmia or a disturbance of the arterial pulse wave. Data representing the signal measured by the motion sensor 6 are stored in memory 8.

[0041] In relation to the figure 2 A prediction system according to another embodiment is described. It further includes a mobile communication device 13 communicating with the monitoring device 2 via a short-range communication interface. The mobile communication device 13 includes an application through which the patient, wearing the monitoring device, can retrieve the signals stored in memory 8. The monitoring device transmits the signals to the mobile communication device 13 via its communication interface 9, which is a short-range interface.

[0042] The mobile communication device 13 can also communicate with the computing unit 10. The mobile communication device can, for example, transmit the received signals via a 2G, 3G, 4G, or 5G communication interface. The computing unit 10 can receive the signals via its communication interface 11, which is a long-range interface.

[0043] The mobile communication device 13 can be, for example, a smart mobile phone or a smart watch.

[0044] The monitoring device 2 is a self-contained unit. The monitoring device 2 includes a power source 7 (BATT) which can, for example, be recharged by an external device, particularly by induction. The battery and electronic components of the data acquisition device can be selected so that the device's autonomy is at least one month.

[0045] There figure 8 illustrates a schematic view of a monitoring device 20 according to another embodiment in which it includes, in addition to the three other sensors 4, 5, 6 described above, a fourth sensor which is an electrical sensor 24 (SENS 4).

[0046] The electrical sensor 24 is configured to measure the electrical impedance of the vascular plaque, which reflects the plaque stiffness level identified as influencing plaque stability. The vascular plaque and the surrounding medium behave as a dielectric medium. By measuring the impedance, it is possible to extract the dielectric parameters, namely the conductivity and permittivity of this medium, which are representative of the plaque stiffness level. In one embodiment, the electrical impedance sensor is configured to apply a current and measure a potential difference. Knowing the voltage and current, the vascular plaque impedance value is determined using Ohm's law. Continuous impedance measurement allows monitoring of changes in plaque stiffness. The data representing the signal measured by the impedance sensor 24 are stored in memory 8.

[0047] There figure 9 illustrates a schematic view of a monitoring device 30 according to another embodiment in which it comprises, in addition to the four sensors 4, 5, 6, 24 of the figure 8 , a fifth passive acoustic wave sensor 35 configured to measure acoustic waves from the arterial wall and plaque.

[0048] It should be noted that the prediction system can also predict the risk of rupture or detachment of a vascular plaque based on one or more signals measured by the five sensors 4, 5, 6, 24, and 35 of the monitoring device. The monitoring device may therefore include any one or more, or any combination of, the five sensors 4, 5, 6, 24, and 35 described above to measure a signal representative of the mechanical stability of the vascular plaque.

[0049] According to one embodiment, the monitoring device includes only one of the four sensors out of the five sensors 4, 5, 6, 24, 35 described above.

[0050] According to another embodiment, the monitoring device may include a combination of two sensors from among the five sensors 4, 5, 6, 24, 35 described above.

[0051] According to yet another embodiment example, the monitoring device may include a combination of three sensors from among the five sensors 4, 5, 6, 24, 35 described above.

[0052] According to yet another embodiment example, the monitoring device may include a combination of four sensors from among the five sensors 4, 5, 6, 24, 35 described above.

[0053] Advantageously, the measurement frequency of the sensors of the monitoring device can be set to emit and / or receive signals from a region of interest located at a certain depth under the skin.

[0054] The combination of sensors varies according to clinical needs.

[0055] When using a monitoring device to track changes over time in the biomechanical properties of a vascular plaque and / or the underlying wall of the carotid artery, the monitoring device may include, for example, a vibration sensor 4 and / or a temperature sensor 5 and / or a motion sensor 6 and / or an electrical sensor 24 and / or an acoustic wave sensor 35 configured to transmit and / or receive signals from a vascular plaque and / or the underlying wall of the carotid artery located at a variable depth below the skin surface, typically between 1 and 3 cm. The measurement frequency of the vibration sensor 4, the temperature sensor 5, the motion sensor 6, the electrical sensor 24, and the acoustic wave sensor 35 are set to transmit and / or receive the signal from the vascular plaque and / or the underlying wall of the carotid artery.

[0056] When using a monitoring device to measure changes in the biomechanical properties of the internal jugular vein, the device may include, for example, a temperature sensor 5 and / or a motion sensor 6 and / or an acoustic wave sensor 35 configured to transmit and / or receive signals from an internal jugular vein located at a variable depth below the skin surface, depending on the patient's anatomy, generally between 1 and 3 cm below the skin. The measurement frequencies of the temperature sensor 5, the motion sensor 6, and the acoustic wave sensor 35 are different and set to transmit and / or receive the signal from the jugular vein.

[0057] There figure 3 This illustrates a schematic view of a common carotid artery 100, which divides into two branches in the neck of a patient: the internal carotid artery 101, which supplies the brain, and the external carotid artery 102, which supplies the neck and face. The monitoring device 2, as described above, can be positioned adjacent to the internal carotid artery just after the bifurcation of the common carotid artery in order to continuously measure the various signals emitted by one or more plaques present on the inner wall of the internal carotid artery.

[0058] On the figure 3 Also schematically illustrated is the internal jugular vein 107, a deep vein in the neck, in contact with the internal carotid artery. There is an internal jugular vein on each side of the neck. The internal jugular vein runs vertically downwards, lateral to the internal carotid artery, and then to the common carotid artery.

[0059] In general, the device's positioning can be adjusted according to clinical needs. In another embodiment, it is also possible to position a second monitoring device or even several devices to monitor not only the internal carotid artery but also its branches or any other artery of the supra-aortic trunks. For example, a second monitoring device can be positioned just before the bifurcation of the common carotid artery.

[0060] With reference to the figure 4 , the monitoring device 2 is in the form of a patch 3 having an adhesive surface which allows the device to be stuck to an area of ​​the outer surface of the skin 106 of the neck, near the carotid artery 100.

[0061] With reference to the Figure 5 The monitoring device 2 can also be a subcutaneous implantable device. The monitoring device 2 includes, for example, a casing made of a biocompatible material. The monitoring device is placed within an implantable probe. It can be placed under the skin 106, near the carotid artery 100. It is placed by the practitioner who performs a subcutaneous dissection. The device is positioned in contact with the muscle 104 under the skin.

[0062] One aspect of this disclosure is to enable the use of system 1 to continuously monitor changes in the biomechanical properties of an internal plate and / or an underlying wall of the carotid artery. For this use, the monitoring device may include a vibration sensor 4 and / or a temperature sensor.

[0063] The monitoring procedure can be implemented in a non-invasive manner by placing the monitoring device on a patient's skin, near the carotid artery.

[0064] The monitoring procedure can also be implemented by placing the monitoring device under a patient's skin, near the carotid artery.

[0065] With reference to the figure 6 , a process implementing a system for predicting the risk of rupture or detachment of a plaque that could lead to a stroke is described below.

[0066] At step E1, the vibration sensor 4 and / or the temperature sensor 5 and / or the motion sensor 6 and / or the electrical sensor 24 and / or the acoustic wave sensor 35 are set to operate within a suitable frequency range to emit and / or receive signals from a vascular plaque and / or an underlying wall of the carotid artery.

[0067] In step E1, the vibration sensor 4 and / or the temperature sensor 5 and / or the motion sensor 6 and / or the electrical sensor 24 and / or the acoustic wave sensor 35 of the monitoring device measure the signals from the plate for a predetermined period. This period varies from a few hours to a few weeks. The sensors are, for example, pre-programmed to operate for a specific duration. This duration is determined by the practitioner based on clinical needs.

[0068] According to one embodiment, the three sensors 4, 5 and 6 of the monitoring device measure the signals from the plaque and / the wall of the carotid artery for a determined time.

[0069] According to another embodiment, the four sensors 4, 5, 6, 24 of the monitoring device measure the signals from the plaque and / or the wall of the carotid artery for a determined period of time.

[0070] According to yet another embodiment, the five sensors 4, 5, 6, 24, 35 of the monitoring device measure the signals coming from the plaque and / or the wall of the carotid artery for a determined period of time.

[0071] At stage E2, the signals are stored in memory 8.

[0072] At step E3, the signals stored in the memory 8 of the monitoring device are sent to the computing unit 10 via the communication interface 9.

[0073] In step E4, the computing unit 10 analyzes the signals using artificial intelligence trained to detect a risk of rupture or plaque detachment that could lead to a stroke. The artificial intelligence includes a neural network trained to determine from the collected signals whether the plaque(s) present in the carotid artery are at risk of rupture.

[0074] Steps E3 and E4, for example, are performed by the practitioner. This allows the practitioner to obtain a reliable prediction of the patient's risk of developing a stroke due to plaque rupture, even if the patient presents no symptoms of the condition.

[0075] Alternatively, in step E3, the patient can also use a mobile communication device 13 to communicate with the monitoring device periodically via a short-range communication interface. The device is, for example, a smartphone. It includes, for example, an application through which the patient can query the monitoring device to receive the signals stored in the device's memory 8. The communication device 13 then transmits the signals via a 4G or 5G network to the processing unit 10. In this way, a diagnosis of the evolution of the plaque stability status can be established remotely by the practitioner periodically, for example, once a week.

[0076] According to one embodiment and with reference to the figure 7 , when the artificial intelligence is a neural network, the process may further include a preliminary learning step to train the artificial intelligence to determine from the collected signals whether the plaque(s) present in the carotid artery present a risk of rupture or detachment.

[0077] More specifically, the learning stage may include the following sub-stages.

[0078] In substep E01, a plurality of monitoring devices is used to collect signals from a population with a known risk of vascular plaque rupture. These signals, referred to as reference signals, are stored on a server.

[0079] In substep E02, the neural network is trained with the reference signals until it converges. The trained neural network is then stored in the computing unit 10, specifically in the central processing unit 12.

[0080] Through the continuous acquisition of a set of signals representative of changes in vascular plaque stability and the use of artificial intelligence to analyze these signals, the system described in this disclosure can predict the risk of plaque rupture that could lead to stroke. This system can be used by clinicians, in conjunction with morphological and histological studies, to monitor and assess local changes in plaque rupture risk and thus prevent serious clinical consequences such as stroke. The system described in this disclosure therefore allows for the rapid identification of individuals who require intervention, even in the absence of pathological symptoms.

[0081] Another aspect of this disclosure is to enable the system to be used for continuous monitoring of changes in the biomechanical properties of an internal jugular vein, which is located deeper under the skin than the carotid artery. For this use, the monitoring device may include a temperature sensor 5, a motion sensor 6, and an acoustic wave sensor 35.

[0082] The monitoring procedure can be implemented in a non-invasive manner by placing the monitoring device on a patient's skin, near the internal jugular vein.

[0083] The monitoring procedure can also be implemented by placing the monitoring device under a patient's skin, near the internal jugular vein.

[0084] With reference to the figure 6 , in step E1, the temperature sensor 5, the motion sensor 6 and the acoustic wave sensor 35 of the monitoring device 2 are set to operate within a suitable frequency range which allows to transmit and / or receive signals from the internal jugular vein for a determined duration.

[0085] The measurement duration varies from a few hours to a few weeks. For example, the sensors are pre-programmed to operate for a specific duration. This duration is determined by the practitioner based on clinical needs. The collected signals represent changes in the biomechanical properties of the internal jugular vein over time.

[0086] At stage E2, the signals are stored in memory 8.

[0087] At step E3, the signals stored in the memory 8 of the monitoring device are sent to the computing unit 10 via the communication interface 9.

[0088] In step E4, computing unit 10 analyzes the signals using artificial intelligence trained to detect a risk of vascular thrombosis. The artificial intelligence includes a neural network trained to determine, from the collected signals, whether changes in the biomechanical properties of the internal jugular vein present a risk of vascular thrombosis.

[0089] Steps E3 and E4, for example, are performed at the practitioner's office. This allows the practitioner to obtain a reliable prediction of the patient's risk of developing vascular thrombosis, even if the patient presents no symptoms of the condition.

[0090] Alternatively, in step E3, the patient can also use a mobile communication device 13 to communicate with the monitoring device periodically via a short-range communication interface. The device is, for example, a smartphone. It includes, for example, an application through which the patient can query the monitoring device to receive the signals stored in the device's memory 8. The communication device 13 then transmits the signals via a 4G or 5G network to the computing unit 10. In this way, the practitioner can remotely monitor changes in the biomechanical properties of the internal jugular vein periodically, for example, once a week.

[0091] According to one embodiment and with reference to the figure 7, when the artificial intelligence is a neural network, the process may further include a preliminary learning step to train the artificial intelligence in such a way as to determine from the signals collected over time to characterize a risk of occurrence of a vascular thrombosis.

[0092] More specifically, the learning stage may include the following sub-stages.

[0093] In substep E01, a plurality of monitoring devices is used to collect signals from a population with a known risk of vascular thrombosis. These signals, referred to as reference signals, are stored on a server.

[0094] In substep E02, the neural network is trained with the reference signals until it converges. The trained neural network is then stored in the computing unit 10, specifically in the central processing unit 12.

[0095] Through the continuous acquisition of a set of signals representative of the evolution of changes in the biomechanical properties of the internal jugular vein, and the use of artificial intelligence to analyze these signals, the system described in this disclosure can also predict the risk of blood clot formation in the internal jugular vein of the neck, which can lead to vascular thrombosis. The method described below helps clinicians, in addition to existing examinations such as Doppler ultrasound to visualize the clot, to assess the risk of jugular thrombosis, enabling them to provide prompt patient care.

[0096] Furthermore, the system's use is not limited solely to predicting stroke. It can be implemented in a patient after a stroke to predict the risk of recurrence following treatment of vascular plaques.

[0097] Although the system of the present invention offers particular advantages for predicting the occurrence of a stroke, the system can be applied more generally for the early detection of patients at risk of developing cardiovascular diseases. The system can also be used in clinical trials to evaluate the efficacy of a new treatment for vascular plaques.

Claims

1. A system (1) for predicting an at least partial rupture or the detachment of vascular plaque that could lead to a stroke, with said vascular plaque being present on an arterial wall selected from among a carotid wall and a supra-aortic trunk wall, said system comprising: - a monitoring device (2) able to be placed in the vicinity of the vascular plaque, said device comprising at least one temperature sensor (5) configured to measure the temperature of the vascular plaque, at least one vibration sensor (4) configured to measure mechanical waves propagated in said arterial wall, at least one motion sensor (6) for the vascular plaque configured to measure movements of the vascular plaque, a memory (8) able to store signals transmitted by said sensors, a communication interface (9), a power source (7) configured to power said sensors and the communication interface (9); - a computation unit (10) adapted to communicate with the monitoring device (2) and configured to analyze measurements of the vibration sensor (4), of the temperature sensor (5) and of the vascular plaque motion sensor (6) originating from the monitoring device (2) by artificial intelligence trained to detect whether there is a risk of at least partial rupturing or detachment of the vascular plaque.

2. The system as claimed in claim 1, wherein the monitoring device (20) further comprises at least one electrical sensor (24) configured to measure a parameter related to the electrical impedance of the vascular plaque.

3. The system as claimed in claim 1 or 2, wherein the monitoring device (30) further comprises at least one acoustic wave sensor (35) for measuring the acoustic waves originating from the arterial wall and / or from the vascular plaque.

4. The system as claimed in any of claims 1 to 3, wherein the monitoring device (2, 20, 30) is in the form of a patch able to be adhered to an external surface of the skin (106) of a patient.

5. The system as claimed in any of claims 1 to 3, wherein the monitoring device (2, 20, 30) is in the form of a subcutaneous implant able to be inserted under the skin of a patient.

6. The system as claimed in any of claims 1 to 5, wherein the temperature sensor (5) is a sensor for detecting thermal waves emitted by the vascular plaque.

7. The system as claimed in any of claims 1 to 6, wherein the vibration sensor (4) comprises an accelerometer.

8. The system as claimed in claim 7, wherein the vibration sensor (4) comprises a 3-axis accelerometer and a 3-axis gyroscope.

9. The system as claimed in any of claims 1 to 8, wherein the vascular plaque motion sensor (6) is an ultrasound probe.

10. The system as claimed in any of claims 1 to 9, wherein the power source (7) is an induction-rechargeable battery.

11. The system as claimed in any of claims 1 to 10, wherein the communication interface (9) is selected from among a short-range radio interface or a near-field communication interface.

12. The system as claimed in any of claims 1 to 11, further comprising at least one mobile communication appliance (13) adapted to remotely communicate with the monitoring device (2) via said communication interface (9), which is a short-range interface, and to transmit signals to the computation unit (10) via a long-range communication interface (11) forming part of the computation unit (10).

13. The system as claimed in claim 3, wherein the monitoring device comprises at least one temperature sensor (5) and / or at least one motion sensor (6) and / or at least one acoustic wave sensor (35), with the measurement frequency of said sensors being set so as to transmit and / or receive a signal originating from an internal jugular vein.