WO2024041727A1

WO2024041727A1 - Detection of a cardiovascular risk

Info

Publication number: WO2024041727A1
Application number: PCT/EP2022/073425
Authority: WO
Inventors: Virginie VISSAC; Zhao ZHAO; Heikki Vilho NIEMINEN
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2022-08-23
Filing date: 2022-08-23
Publication date: 2024-02-29

Abstract

This disclosure is concerned with cardiovascular diseases and risks. The disclosure provides a system for detecting a cardiovascular risk. The system acquires a cardiac reference signal by measuring an electrocardiogram (ECG) or phonocardiogram (PCG). Further, the system measures a first pulse waveform signal at a first ear of the user, the first pulse waveform signal having a timing clock synchronized with that of the reference signal, and a second pulse waveform signal at a second ear of the user, the second pulse waveform signal having a timing clock synchronized with that of the reference signal. The system measures a third pulse waveform signal at an arm, finger or wrist of the user, the third pulse waveform signal having a timing clock synchronized with that of the reference signal. Then, the system calculates the cardiovascular risk of the user based on the measured pulse waveform signals and the reference signal.

Description

DETECTION OF A CARDIO VASCULAR RISK

TECHNICAL FIELD

The present disclosure relates to the field of medical technology. In particular, the disclosure is concerned with cardiovascular diseases and risks. To this end, the disclosure provides a system and method for detecting a cardiovascular risk.

BACKGROUND

Cardiovascular diseases are still today the major cause of death worldwide, accounting for 31% of the total deaths according to the World Health Organization (WHO) in 2016. The majority of these diseases are caused by a poor diet (food, tobacco, etc.) and, therefore, could be prevented. However, since the majority of the examinations targeting the heart are done in medical facilities, if a cardiovascular disease is detected in a patient, it means that the body already suffered the first damages. If those diseases were detected earlier, the patient could change his/her way of life to improve his/her cardiovascular health, before the disease can develop.

In particular, atherosclerosis can generate a plaque (made of fat, cholesterol, calcium or other substances that can be found in the blood), which builds up in the arteries. With time, the plaque can harden and narrow the arteries, which can cause a limitation of oxygen-rich blood flow to the organs and parts of the body placed after the blockage. Fig. 1 illustrates the effect of plaque 102 which is built up in arteries 101. In particular, Fig. la shows a normal cross-section of an artery 101, i.e., without plaque, while Fig. lb shows the same cross-section of the artery 101 with plaque 102.

Depending on the concerned arteries, different related diseases are associated with atherosclerosis: coronary heart disease, when the plaque builds up in the coronary arteries limiting the blood supply to the heart; carotid artery disease (CAD), when it affects the carotid arteries limiting the blood supply to the brain; peripheral artery disease (PAD), when it affects the arteries supplying the arms, legs or pelvis; and chronic kidney disease, when this affects the renal arteries. Apart from the damages that can be caused to the body by reduction of blood supply, atherosclerosis may increase the risk of thrombosis (part of the plaque ruptures, thereby damaging the lining of the artery, and blood starts to clot at this location and causes a blockage) or embolism (a clot formed on the plaque breaks off and flows somewhere else where it blocks an artery or a vessel). Unfortunately, most of the time, no symptoms are experienced by the patients, until a blockage has occurred.

Since atherosclerosis is mainly asymptomatic, it is hard to determine its incidence. However, it is considered as the major cause of cardiovascular diseases. For example, 75% of acute myocardial infarctions are caused by plaque rupture. Furthermore, according to a study published in 2020 (see ‘Global and regional prevalence, burden, and risk factors for carotid atherosclerosis: a systematic review, meta-analysis, and modelling study Lancet Glob Health, e721-e729 (2020)), the prevalence of carotid artery disease is 1.5 % worldwide (carotid stenosis) with 21.1% of the population presenting carotid plaque. Atherosclerosis also accounts for 90 % of the cases of PAD (see e.g. ‘Update on peripheral artery disease: Epidemiology and evidence-based facts, Atherosclerosis, 275, 379-381 (2018)), which concerns 200 million people worldwide.

Conventional methods used to diagnose atherosclerosis include blood tests (to look for cholesterol levels), electrocardiogram (ECG), chest X-ray, echocardiography, computed tomography scan, stress testing, angiography and ankle-brachial index (blood pressure measurement of both arms and legs to determine the presence of PAD). All of these methods need to be done by professionals in specific facilities, and this explains why this disease is generally not diagnosed before the apparition of the first symptoms. There is currently no mobile heath solution available for normal users’ daily use to early-warn, detect and monitor the progress of atherosclerosis.

Atherosclerosis requires a lot of exams to be diagnosed. The simplest methods are finally the ones targeting a specific associated disease such as the listening of turbulence sounds in the carotid arteries for CAD or the ankle-brachial index measurement for PAD. Even in those simpler cases, the need of a specialist is required, meaning that the patient has to make a specific consultation.

The so-called pulse wave velocity can also be a good indicator of the stiffening of the arteries, which could be a good indicator of the risk of having or developing arteriosclerosis (arteriosclerosis is defined as a stiffening or thickening of the arteries and atherosclerosis is a particular type of arteriosclerosis). However, the medical system uses tonometric probes to measure it, which are bulky and difficult to use.

In order to detect a cardiovascular risk, some easy and not very precise methods are embedded into some wearable devices, such as scales or watches. However these methods only look at one of the arteries to deduce information: e.g. one leg or one arm. However, the development of those diseases is mainly asymmetrical (see e.g. ‘ Asymmetrical limbs arterial pressures: a new marker of atherosclerosis, Hypertension Research, 36, 394-395 (2013)). This means that an atherosclerotic lesion could go unnoticed, if it is present in the non-observed limb, and that the assumption of symmetrical values made to determine other parameters (for example, a watch using photoplethysmography (PPG) only in two different locations on both arms to determine blood pressure directly from the time difference between the two arms) may be wrong for users presenting such diseases. This can lead to an inaccurate determination of certain parameters.

Finally, devices measuring blood pressure non-invasively and without a cuff using a simple pulse wave velocity method generally focus on the wrist. However, because the radial artery (as any peripheral artery basically) has the property to change its radius (vasodilation and vasoconstriction) to adapt the blood flow, the link between pulse wave velocity and blood pressure is not unique (it changes with the radius of the artery). As a consequence, when measuring only the pulse wave velocity on the radial artery without knowing the peripheral pulse wave velocity value or the radius value of the artery, it is very difficult to calculate the corresponding blood pressure.

SUMMARY

In view of the above, an objective of this disclosure is to provide a system and method for a better detection of a cardiovascular risk. For instance, the cardiovascular risk should be detected more accurately and in an earlier stage.

This and other objectives are achieved by the solution provided in the independent claims. Advantageous implementations are further defined in the dependent claims.

According to a first aspect, a system for detecting a cardiovascular risk is provided. The system is configured to: acquire a cardiac reference signal by measuring an electrocardiogram (ECG) or phonocardiogram (PCG), measure at least a first pulse waveform signal at a first ear of the user, the first pulse waveform signal having a timing clock synchronized with that of the reference signal, measure at least a second pulse waveform signal at a second ear of the user, the second pulse waveform signal having a timing clock synchronized with that of the reference signal, measure at least a third pulse waveform signal at an arm, finger or wrist of the user, the third pulse waveform signal having a timing clock synchronized with that of the reference signal, and calculate a cardiovascular risk of the user based on the measured pulse waveform signals and the reference signal.

The system of the first aspect has the advantage that a cardiovascular risk of the user could be detected before the onset of symptoms.

In an implementation of the first aspect, the system is further configured to: determine a first set of features of a first side of the body of the user based on the first pulse waveform signal and the reference signal, determine a second set of features of a second side of the body of the user based on the second pulse waveform signal and the reference signal, determine one or more symmetry parameters based on the first set of features and the second set of features, and/or determine one or more timing parameters of the first set of features and the second set of features with respect to the reference signal, and the system being further configured to calculate the cardiovascular risk of the user based on at least one of the one or more symmetry parameters and the one or more timing parameters.

In an implementation of the first aspect, the first set of features and/or the second set of features, respectively, comprises at least one of: a pulse rise time (PRT); an augmentation index (Al); a pulse arrival time (PAT), if the reference signal is the ECG reference signal; and a pulse transit time (PTT), if the reference signal is the PCG reference signal.

In an implementation of the first aspect, the system is further configured to: determine a first pulse arrival time (PAT), or pulse transit time (PTT) related to a first carotid of the user based on the first pulse waveform signal and the reference signal and deduce a first carotid pulse wave velocity (PWV) from it and biometric data of the user, such as gender, height and age, determine a second PAT or PTT related to a second carotid of the user based on the second pulse waveform signal and the reference signal and deduce a second carotid PWV from it and biometric data of the user, such as gender, height and age, compare the first carotid pulse wave velocity with the second carotid pulse wave velocity, and calculate the cardiovascular risk of the user based on the result of the comparison.

In an implementation of the first aspect, the system is further configured to: determine a first PAT or PTT related to a first arm of the user based on the third pulse waveform signal and the reference signal and deduce a first arm PWV from it and biometric data of the user, such as gender, height and age, determine a second PAT or PTT related to a second arm of the user based on the third or a fourth pulse waveform signal and the reference signal and deduce a second arm PWV from it and biometric data of the user, such as gender, height and age, compare the first arm pulse wave velocity with the second arm pulse wave velocity, and calculate a cardiovascular risk of the user based on the comparison.

In an implementation of the first aspect, the system is further configured to: determine a ratio between a carotid pulse wave velocity and an arm pulse wave velocity using at least one of the carotid pulse wave velocities calculated by the system and at least one of the arm pulse wave velocities calculated by the system, calculate a cardiovascular risk of the user based on the pulse wave velocity ratio.

In an implementation of the first aspect, the system further comprises: a first ear-mounted device, a second ear-mounted device, wherein the first ear-mounted device and the second earmounted device are, respectively, configured to measure the first pulse waveform signal and the second pulse waveform signal, either simultaneously or consecutively.

In an implementation of the first aspect, the first ear-mounted device or the second ear-mounted device comprises a processor configured to calculate the cardiovascular disease risk. Additionally or alternatively, the first ear-mounted device or the second ear-mounted device is configured to measure the reference signal and/or the third pulse waveform signal.

In a an implementation of the first aspect, the first ear-mounted device and the second earmounted device are configured to communicate with each other, in order to synchronize the first pulse waveform signal and the second pulse waveform signal with the reference signal. In an implementation of the first aspect, the system comprises a smartphone or smartwatch, the smartphone or smartwatch being configured to measure the third pulse waveform signal, and optionally to measure the reference signal.

In an implementation of the first aspect, the smartwatch or smartphone is configured to communicate with the first ear-mounted device and/or the second ear-mounted device, in order to synchronize the third pulse waveform signal and the first and/or second pulse waveform signal with the reference signal.

In an implementation of the first aspect, at least one of the first ear-mounted device and the second ear-mounted device comprises an ECG sensor or a PCG sensor configured to measure the reference signal.

In an implementation of the first aspect, at least one of the first ear-mounted device and/or the second ear-mounted device comprises a pulse waveform measuring sensor configured to measure the third pulse waveform signal.

In an implementation of the first aspect, the system further comprises a multi- wavelength photoplethysmography (PPG) sensor comprising at least a red, infrared and green LED.

In a preferred embodiment of the first aspect, the system is further configured to: determine an oxygen saturation (SPO2) value using the red and infrared PPG sensors; and determine an arterial pulse waveform on the basis of the at least red, infrared and green PPG signals.

In a preferred embodiment of the first aspect, the system is further configured to measure, in the first ear of the user, a first SPO2 value, measure, in the second ear of the user, a second SPO2 value, and calculate the cardiovascular risk based on a comparison of the first SPO2 value and on the second SPO2 value.

According to a second aspect, a method for detecting a cardiovascular risk is provided. The method comprises measuring an ECG or PCG reference signal, measuring at least a first pulse waveform signal at a first ear of the user, the first pulse waveform signal having a timing clock synchronized with that of the reference signal, measuring at least a second pulse waveform signal at a second ear of the user, the second pulse waveform signal having a timing clock synchronized with that of the reference signal, measuring at least a third pulse waveform signal at an arm, finger or wrist of the user, the third pulse waveform signal having a timing clock synchronized with that of the reference signal, and calculating a cardiovascular risk of the user based on the measured pulse waveform signals and the reference signal.

It has to be noted that all devices, elements, units and means described in the present application could be implemented in the software or hardware elements or any kind of combination thereof. All steps which are performed by the various entities described in the present application as well as the functionalities described to be performed by the various entities are intended to mean that the respective entity is adapted to or configured to perform the respective steps and functionalities. Even if, in the following description of specific embodiments, a specific functionality or step to be performed by external entities is not reflected in the description of a specific detailed element of that entity which performs that specific step or functionality, it should be clear for a skilled person that these methods and functionalities can be implemented in respective software or hardware elements, or any kind of combination thereof.

BRIEF DESCRIPTION OF DRAWINGS

The above described aspects and implementation forms of the present invention will be explained in the following description of specific embodiments in relation to the enclosed drawings, in which:

FIG. 1 illustrates the effect of plaque built up in arteries;

FIG. 2 shows a schematic representation of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 3 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment;

FIG. 4 shows a schematic representation of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 5 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment;

FIG. 6 shows a schematic representation of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 7 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment; FIG. 8 shows a schematic representation of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 9 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment;

FIG. 10 shows a schematic representation of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 11 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment;

FIG. 12 shows a schematic representation of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 13 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment;

FIG. 14 shows a schematic representation of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 15 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment;

FIG. 16 shows a schematic representation of part of a system for calculating a cardiovascular risk according to an embodiment;

FIG. 17 shows a schematic representation of some steps of a method for calculating a cardiovascular risk according to an embodiment;

FIG. 18 shows a schematic representation of a flowchart for extracting all the parameters that can be measured with the system and will be used to determine a cardiovascular risk

FIG. 19 shows a schematic representation of a flowchart for calculating a cardiovascular risk according to an embodiment;

FIG. 20 shows a schematic representation of a flowchart for calculating a cardiovascular risk according to an embodiment; and

FIG. 21 shows a schematic representation of a method for calculating a cardiovascular risk according to an embodiment.

DETAILED DESCRIPTION OF EMBODIMENTS

FIG. 2 shows a schematic representation of a system 202 for calculating a cardiovascular risk of a user 201 according to an embodiment. The system may comprise at least one ear-mounted device like an earphone. The system may comprise at least one smart device, like a smartphone or smartwatch.

The system 202 is configured to acquire a cardiac reference signal a by measuring an electrocardiogram (ECG) or phonocardiogram (PCG). For instance, the system 202 may comprise an ECG sensor or PCG sensor for this measurement, wherein the sensor may be included in an ear-mounted device of the system 202.

The system 202 is further configured to measure at least a first pulse waveform signal b at a first ear of the user 201, the first pulse waveform signal b having a timing clock synchronized with that of the reference signal a. To this end, the system 202 may include a first pulse waveform measuring sensor, e.g., included in a first ear-mounted device of the system 202.

The system 202 is also configured to measure at least a second pulse waveform signal c at a second ear of the user 201, the second pulse waveform signal c having a timing clock synchronized with that of the reference signal a. To this end, the system 202 may include a second pulse waveform measuring sensor, e.g., included in a second ear-mounted device of the system 202.

The system 202 is further configured to measure at least a third pulse waveform signal d at an arm, finger or wrist of the user 201, the third pulse waveform signal d having a timing clock synchronized with that of the reference signal a. To this end, the system 202 may comprise a smartphone or smartwatch equipped a pulse waveform measuring sensor.

The system 202 is configured to calculate the cardiovascular risk e of the user 201 based on the measured pulse waveform signals b, c, d, and the reference signal a.

The system 202 has the advantage that the cardiovascular risk e of the user 201 could be detected before the onset of the symptoms with daily used objects. In particular, the system 202 can take into account a symmetry by measuring the pulse waveform signals at the two ears and at least one arm, finger, or wrist of the user. FIG. 3 shows a schematic representation of some steps of a method 2100 for calculating the cardiovascular risk of the user 201 according to an embodiment. The method 2100 can be performed by the system 202 of FIG. 2.

In the step 301, the acquisition of the signals a, b, c, and d is performed by the system 202.

Then, in step 302, the calculation of one or more parameters extracted from the signals is performed by the system 202. In particular, the system 202 may be configured to determine a first set of features of a first side of the body of the user 201 based on the first pulse waveform signal b and the reference signal a. Further, the system 202 may be configured to determine a second set of features of a second side of the body of the user 201 based on the second pulse waveform signal c and the reference signal a. Further, the system 202 may be configured to determine one or more symmetry parameters based on the first set of features and the second set of features, and/or to determine one or more timing parameters of the first set of features and the second set of features with respect to the reference signal a. In other words, the one or more calculated parameters may include the one or more symmetry parameters and/or the one or more timing parameters. Then, the system 202 is configured to calculate the cardiovascular risk of the user 201 based on at least one of the one or more symmetry parameters and the one or more timing parameters.

For example, the first set of features and/or the second set of features, respectively, may comprise at least one of a PRT, an Al, a PAT, if the reference signal a is the ECG reference signal, and a PTT, if the reference signal a is the PCG reference signal.

The system 202 may comprise a smartphone 801 and the method 2100 may further comprise the step 303 of transferring information, in particular, the first set of features and the second set of features, to the smartphone 801.

Then, in the step 304, the system 202 may be configured to perform the step of calculating further information using biometric data of the user 201. Finally, in step 305, the system 202 is configured to perform the step of calculating, as mentioned above, the cardiovascular risk by using the above mentioned calculated parameters. In one example, the system 202 may be configured to determine a first PAT or PTT related to a first carotid of the user 201 based on the first pulse waveform signal b and the reference signal a, and to deduce a first carotid PWV from it and biometric data of the user 201, such as gender, height and age. The system 202 may also be configured to determine a second PAT or PTT related to a second carotid of the user 201 based on the second pulse waveform signal c and the reference signal a, and to deduce a second carotid PWV from it and biometric data of the user 201, such as gender, height and age. Then, the system 202 can be configured to compare the first carotid pulse wave velocity with the second carotid pulse wave velocity, and to calculate the cardiovascular risk e of the user 201 based on the result of the comparison.

In another example, the system 202 may be further configured to determine a first PAT or PTT related to a first arm of the user 201 based on the third pulse waveform signal d and the reference signal a, and to deduce a first arm PWV from it and biometric data of the user 201, such as gender, height and age. The system 202 may be also configured to determine a second PAT or PTT related to a second arm of the user 201 based on the third d or a fourth pulse waveform signal and the reference signal a and deduce a second arm PWV from it and biometric data of the user 201, such as gender, height and age. Then, the system 202 may compare the first arm pulse wave velocity with the second arm pulse wave velocity, and calculate the cardiovascular risk e of the user 201 based on the comparison.

In another example, the system 202 may further be configured to determine a ratio between a carotid pulse wave velocity and an arm pulse wave velocity using at least one of the carotid pulse wave velocities calculated by the system 202 and at least one of the arm pulse wave velocities calculated by the system 202, and to calculate the cardiovascular risk e of the user 201 based on the pulse wave velocity ratio.

In summary, Fig. 3 shows the principle of the measurement of the cardiovascular risk wherein in step 301 the signals are acquired by the sensors, in step 302 the parameters are extracted from the signals, in step 303 the information is transferred to a smartphone or the like, in step 304 more information is calculated using biometric data like age or gender of the user 201, and in step 305 all those calculated parameters are used into an algorithm calculating the cardiovascular risk of the user 201. In this method 2100, the step 302 and 303 can be exchanged, so that the signals can directly be transferred to the smartphone or the like, to directly calculate the parameters using a bigger processing power. FIG. 4 shows a schematic representation of an exemplary system 202 for calculating the cardiovascular risk of the user 201 according to an embodiment.

In this embodiment, the system 202 comprises a first ear-mounted device 401 and a second earmounted device 402. The first ear-mounted device 401 and the second ear-mounted device 402 can, respectively, be configured to measure the first pulse waveform signal b and the second pulse waveform signal c, either simultaneously or consecutively.

Moreover, the first ear-mounted device 401 or the second ear-mounted device 402 can comprise a processor, which is configured to calculate the cardiovascular risk e. Furthermore, the first ear- mounted device 401 or the second ear- mounted device 402 can be configured to measure the reference signal a and/or to measure the third pulse waveform signal d. The first earmounted device 401 and/or the second ear-mounted device 402 can be earbuds.

For example, the first ear-mounted device 401 and the second ear-mounted device 402 are configured to communicate with each other, in order to synchronize the first pulse waveform signal b and the second pulse waveform signal c with the reference signal a.

In an implementation, at least one of the first ear-mounted device 401 and the second earmounted device 402 comprises a PCG sensor l.a and l.b configured to measure the reference signal a.

For instance in the embodiment shown in Fig. 4, the system 202 may comprise a pair of earbuds 401, 402 (left earbud and right earbud), each containing a multi-wavelength PPG sensor 2. a and 2.b to be placed in contact with the inner part of the ears of the user 201, the PCG sensors (basically the microphones inside the earbuds 401, 402) l.a and l.b, and the pulse-waveform measuring sensors 3. a and 3.b placed on the external sides of the earbuds to contact a finger of each hand of the user 201.

The embodiments of the present disclosure allow the determination of the risk of presenting atherosclerosis, and mainly some of its associated diseases: PAD and CAD. This can be achieved by means of a wearable device that can be used in everyday life and which can track the changes of part of the cardiovascular system as a function of time. The system 202 can comprise the earbuds 401, 402 shown in FIG. 4, which may be used alone or in combination with another device, for instance, a smartphone 801 (see Fig. 8) or a smartwatch 1401 (see Fig. 14). The system 202 may be able to look at both the carotid arteries and the arm arteries and deduce the cardiovascular risk e from it.

The cardiovascular risk e can be determined by comparison of some parameters (e.g., pulse wave velocity, oxygen saturation levels, pulse rise time, augmentation index) between the carotid arteries and the arms, as well as by evaluation of the found values in themselves. The pulse wave velocity ratio (ratio of PWV between the carotid artery and the arm artery) can also be calculated and analyzed to evaluate the mismatch level of the compliance of the arteries between central and peripheral arteries. The method 2100 presented here can also enable to calculate the carotid pulse wave velocity, which is an important cardiovascular parameter in itself. It indicates the degree of stiffening of the user’s 201 arteries and can be used, in turn, to determine the blood pressure of the user 201 via calibration with a certified device.

Embodiments of the present disclosure enable cardiovascular disease detection as early and easy as possible, in order to enable the users to make the right changes in their life before it is too late or even before they have to take any medicine. This disclosure enables to measure several vital sign parameters such as the pulse wave velocity ratio, carotid pulse wave velocity, blood pressure, arterial stiffness, etc.

The system 202 may be further configured to calculate the risk of having or developing PAD or CAD using earbuds 401, 402 alone or in combination with another everyday life object such as a smartwatch 1401 or a smartphone 801.

The system 200 may be further configured to calculate the risk of having or developing PAD or CAD by evaluating the symmetry of different parameters measured in both carotid and arm arteries (oxygen saturation, augmentation index, pulse rise time, pulse wave velocity) of the user 201.

The system 202 may be configured to evaluate the cardiovascular health of the user 201 by calculating the carotid PWV obtained either using the PAT or PTT (depending on the reference measurement used) measured using either a PCG or ECG as a reference signal and the PPG signals inside the ears of the user 201. The system 202 may be configured to determine the blood pressure of the user using the carotid PWV obtained previously and a validated blood pressure monitor to calibrate the values. This method is more precise than other wearables measuring only at the wrist as the carotid artery is an elastic artery which implies that its radius does not change due to vasodilation and vasoconstriction. This means that there is only one relationship between PWV and blood pressure that can be determined by calibration on the contrary to measurements made at the wrist that present several relationships depending on the radius of the artery.

The system 202 may be configured to determine the arterial stiffness mismatch by calculating the PWV ratio obtained using the carotid PWV and the arm PWV.

The system 202 can further be configured to determine the cardiovascular risk e of the user 201 by evaluating the symmetry of the pulse waveform features and timings measured at the ears and the arms of the user 201.

The measurements may be done simultaneously or consecutively (one side after the other). The signals can also either be synchronized using the PCG signals measured in each earbud 401, 402 or by a Bluetooth connection between the two earbuds 401, 402.

In an embodiment, at least one of the first ear-mounted device 401 and/or the second earmounted device 402 comprises a pulse waveform measuring sensor configured to measure the third pulse waveform signal d.

FIG. 5 shows a schematic representation of some steps of the method 2100 for calculating the cardiovascular risk e according to an embodiment.

In particular, Fig. 5 shows the two possible data acquisition processes using the embodiment corresponding to step 301 of Fig. 3 (in Fig. 5, mwPPG stands for multi- wavelength photoplethy smography) .

In the first process (steps 501 and 502), the user 201, at first, places both its hands against the earbuds 401, 402. Afterwards, in step 502, all the measurements are done simultaneously: left ear mwPPG, right ear mwPPG, left ear PCG, left hand PPG, right hand PPG, and right ear PCG. In the second process (steps 503-506), at first, the user 201 places its right hand against the right earbud 402 in step 503. Then, the system 202, can be configured to perform the following measurements on the right in step 504: right ear mwPPG, right hand PPG, right ear PCG. Afterwards, in step 505, the user 201 places his left hand against the left earbud 401. Then, in step 506, the system 202 can be configured to perform the following measurements on the left: left ear mwPPG, left hand PPG, left ear PCG.

FIG. 6 shows a schematic representation of the system 202 for calculating the cardiovascular risk e according to an embodiment.

In the embodiment presented in Fig. 6, the PCG sensors l .a and l.b are replaced by one ECG sensor (electrodes 4. a, 4.b and 4.c) that is placed on the left earbud 401, so that at least one electrode is made to contact the left ear of the user 201 and one electrode can be contacted by the left hand of the user 201. In that case, the measurements may be done simultaneously on both ears to enable the synchronization of all the pulse waveform signals (in ear and finger PPG) with the ECG signal. The two earbuds 401, 402 can be connected by Bluetooth, enabling a synchronization of the measured signal with an accuracy better than 0.5 ms.

FIG. 7 shows a schematic representation of some steps of the method 2100 for calculating the cardiovascular risk e according to an embodiment;

In particular, in Fig. 7, the process of the data acquisition corresponding to Fig. 6 is shown.

In step 701, the user 201 places both its hands against the earbuds 401, 402. Then, in step 702, the two earbuds 401, 402 can connect together via Bluetooth for timing synchronization. Finally, in step 703, all measurements are done simultaneously, namely: left ear mwPPG, right ear mwPPG, ECG, left hand PPG, and right hand PPG.

FIG. 8 shows a schematic representation of the system 202 for calculating the cardiovascular risk according to an embodiment. In the embodiment shown in Fig. 8, the system 202 further comprises the smartphone 801 or smartwatch 1401 (see Fig. 14). The smartwatch 1401 or smartphone 801 can be configured to communicate with the first earmounted device 401 and/or the second ear-mounted device 402, in order to synchronize the third pulse waveform signal d and the first b and/or second pulse waveform signal c with the reference signal a.

In this embodiment, the external pulse waveform measuring sensors made to contact each arm of the user 3. a and 3.b are replaced by sensors that are placed on another device. Here, for example, the camera of the smartphone 801. In this case, the smartphone 801 can be connected to the earbuds 401, 402 using Bluetooth, the reference signals can be PCG signals embedded in each earbud 401, 402.

FIG. 9 shows a schematic representation of some steps of the method 2100 for calculating the cardiovascular risk according to an embodiment. In particular, in Fig. 9, the process of data acquisition of the embodiment shown in Fig. 8 is shown.

The measurements are made consecutively so that, when the user places one of its finger of its left hand on the camera of the smartphone 801 to obtain a PPG measurement, the smartphone 801 connects in Bluetooth to the left earbud 401 to simultaneously measure the PCG inside the left ear and the multi-wavelength PPG signal inside the left ear. Then, a connection is made to the right earbud 402 to measure the right side of the user 201.

In particular, the method 2100 comprises the steps of step 901: the user places a finger of the left hand against the camera of the smartphone 801; step 902: the smartphone 801 connects to the left earbud 401 via Bluetooth to enable timing synchronization; step 903 : the acquired measurements are: left ear mwPPG, left ear PCG, left hand PPG; step 904: the user places a finger of the right hand against the camera of the smartphone 801; step 905: the smartphone 801 connects to the right earbud 402 via Bluetooth to enable timing synchronization; and step 906: the acquired measurements are right ear mwPPG, right ear PCG, right hand PPG. FIG. 10 shows a schematic representation of the system 202 for calculating the cardiovascular risk e according to an embodiment.

In the embodiment presented in Fig. 10, the PCG sensors l.a and l.b of the embodiment presented in Fig. 8 are replaced by an ECG sensor using electrodes 4. a to 4.c. In this case, the measurement is also done consecutively, as shown in Fig. 11, so that the user 201 contacts its left hand with the electrode 4.b. The left earbud 401 can first be connected in Bluetooth with the right earbud 402 to synchronize the in-ear multi-wavelength PPG signals taken by the right earbud 402 with the reference ECG signal measured by the left earbud 401, while keeping the left hand in contact with electrode 4.b. The user 201 contacts the camera of the smartphone 801 with a finger of the right hand, and the left earbud 401 connects to the smartphone 801 to synchronize the pulse waveform signal measured on the right hand to the reference ECG signal.

Finally, the user switches hands to contact the right hand with the electrode 4.b and the left hand with the camera of the smartphone 3. a, while the left earbud 401 is still connected with the smartphone 801 using Bluetooth to synchronize the pulse waveform signal of the left hand with the reference ECG signal.

FIG. 11 shows a schematic representation of some steps of the method 2100 for calculating the cardiovascular risk e according to an embodiment. In particular, in Fig. 11, the process of data acquisition of the embodiment shown in Fig. 10 is depicted.

In particular, the method 2100 comprises the steps of: step 1101 : the user 201 places a finger of the left hand against the left earbud 401; step 1102: the left earbud 401 connects to the right earbud 402 via Bluetooth for timing synchronization; step 1103: acquired measurements are the left ear mwPPG, the right ear mwPPG, and the ECG signal; step 1104: the user 201 places a finger of the right hand against the camera of the smartphone 801; step 1105: the left earbud 401 connects to the smartphone 801 via Bluetooth for timing synchronization; step 1106: acquired measurements are the left ear mwPPG, the right hand PPG, and the ECG signal; step 1107: the user 201 exchanges placement of the hands. One finger of the left hand is in contact with the camera of the smartphone 801, whereas one finger of the right hand is in contact with the ECG electrodes of the left earbud 401; and step 1108: acquired measurements are the left ear mwPPG, the left hand PPG, and the ECG signal.

FIG. 12 shows a schematic representation of the system 202 for calculating the cardiovascular risk e according to an embodiment.

In the embodiment shown in Fig. 12, the smartphone 801 also embeds the ECG electrodes 4. a and 4.b. In this case, the measurements can be done consecutively one side after the other, connecting via Bluetooth the smartphone 801 with the earbud 401, 402 corresponding to the hand being measured as described in Fig. 13. For example, if the user 201 places a finger of the left hand first in front of the camera 3. a of the smartphone 801 for a PPG measurement, the smartphone 801 should connect via Bluetooth to the left earbud 401 to synchronize with the in- ear PPG signals. Then, the user 201 can place a finger of the right hand on the camera 3. a and the smartphone 801 can connect to the right earbud 402.

FIG. 13 shows a schematic representation of some steps of the method 2100 for calculating the cardiovascular risk e according to an embodiment. In particular, in Fig. 13, the process of data acquisition of the embodiment shown in Fig. 12 is shown.

In particular, the method 2100 comprises: step 1301 : the user 201 places its left hand in contact with the left ECG electrode placed on the smartphone 801; step 1302: the user 201 places its right hand in contact with the right ECG electrode placed on the smartphone 801; step 1303: the user 201 places a finger of the left hand in contact with the camera 3. a of the smartphone 801; step 1304: the smartphone 801 connects via Bluetooth with the left earbud 401 for timing synchronization; step 1305: acquired measurements are the left ear mwPPG, the left hand PPG, and the ECG signal; step 1306: the user 201 places a finger of its right hand in contact with the camera 3. a of the smartphone 801; and step 1307: the smartphone 801 connects via Bluetooth with the right earbud 402 for timing synchronization step 1308: acquired measurements are the right ear mwPPG, the right hand PPG, and the ECG signal;

FIG. 14 shows a schematic representation of the system 202 for calculating the cardiovascular risk e according to an embodiment.

In this embodiment, the earbuds 401, 402 are combined with a smartwatch 1401 embedding ECG and PPG sensors on the back to measure at the wrist of the user 201. In this case, measurements can be done on both arms by switching the side of the smartwatch 1401. In which case, the smartwatch 1401 can connect via Bluetooth to the earbud 401, 402 on the side of the wrist wearing the watch, as it is shown in Fig. 15.

FIG. 15 shows a schematic representation of some steps of the method 2100 for calculating the cardiovascular risk e according to an embodiment. In particular, in Fig. 15, the process of data acquisition of the embodiment shown in Fig. 14 is shown.

In particular, the method 2100 comprises: step 1501 : the user 201 places the smartwatch 1401 on its left wrist; step 1502: the smartwatch 1401 connects to the left earbud 1401 via Bluetooth for timing synchronization; step 1503: the user 201 places a finger of its right hand in contact with the top ECG electrode of the smartwatch 1401; step 1504: acquired measurements are the left ear mwPPG, the left hand PPG, and the ECG signal; step 1505: the user 201 places the smartwatch 1401 on its right wrist; step 1506: the smartwatch 1401 connects to the right earbud 402; step 1507: the user 201 places a finger of the left hand in contact with the top ECG electrode of the smartwatch 1401; step 1508: acquired measurements are the right ear mwPPG, the right hand PPG and the ECG signal. FIG. 16 shows a schematic representation of a part of the system 202 for calculating the cardiovascular risk e according to an embodiment.

In another embodiment shown in Fig. 16, the smartwatch 1401 can embed an additional PPG sensor close to the ECG electrode 4.b so that the user 201 can measure both the ECG signals and both arms PPG signals simultaneously. In this case, the smartwatch 1401 can connect to any earbud 401, 402 consecutively to synchronize the in-ear PPG signals to the reference ECG signal as shown in Fig. 17.

FIG. 17 shows a schematic representation of some steps of the method 2100 for calculating the cardiovascular risk e according to an embodiment. In particular, in Fig. 17, the process of data acquisition of the embodiment shown in Fig. 16 is shown.

In particular, the method 2100 comprises: step 1701 : the user 201 places the smartwatch 1401 on the left wrist; step 1702: the smartwatch 1401 connects to the left earbud 401 via Bluetooth for timing synchronization; step 1703: the user 201 places a finger of the right hand in contact with the top ECG electrode of the smartwatch 1401 and top PPG sensor on the smartwatch 1401; step 1704: acquired measurements are the left ear mwPPG, the left hand PPG and the ECG signal; step 1705: the smartwatch 1401 connects to the right earbud 402; and step 1706: acquired measurements are the right ear mwPPG, the right hand PPG and the ECG signal.

The possible embodiments compatible with this method 2100 are not limited to the devices cited here, for example a ring embedding electrodes and PPG sensors can also be used instead of the smartwatch 1401.

FIG. 18 shows a schematic representation of a flow chart 1800 of the measured parameters by the system 202 on each side of the user and used for calculating the cardiovascular risk e according to an embodiment. The system 202 can further comprise a multi- wavelength PPG sensor comprising at least a red, infrared and green LED. The system 202 can further be configured to determine an oxygen saturation (SPO2) value using the red and infrared PPG sensors, and to determine an arterial pulse waveform on the basis of the at least red, infrared and green PPG signals.

Moreover, the system 202 can further be configured to measure, in the first ear of the user 201, a first SPO2 value, and to measure, in the second ear of the user 201, a second SPO2 value. Further the system 202 can calculate the cardiovascular risk e based on a comparison of the first SPO2 value and on the second SPO2 value.

The system 202 may be able to measure the reference signal (ECG or PCG), in-ear PPG signals and arms pulse waveform signals. Therefore, the information that can be obtained with such system is described in Fig. 18.

In each ear, the multi-wavelength PPG sensor can enable first to determine an SPO2 value, then the multi-wavelength PPG signals can be combined to reconstruct an arterial pulse waveform by taking into account the fact that each wavelength can penetrate at a different depth. Shorter wavelengths that do not penetrate to vascular bed, but probe the surface layers of skin tissue, can be used to remove artefacts from the long wavelength signal that passes through surface layers before reaching the vascular bed. This allows the reconstruction of arterial pulse waveform from PPG signals. This arterial pulse waveform improves the accuracy of the physiological parameters that can be determined with PPG signals.

Examples of physiological parameters that can be derived from pulse waveform are: the PRT, the Al and other time based parameters that can be calculated by combining arterial pulse waveform with the reference signal. Comparison of this signal to the reference signal can, for instance, give a PAT in case the reference signal is ECG, or a PTT in case it is PCG.

From those time based parameters, a PWV can be calculated (from PTT) or estimated (from PAT) using information from the user like height, gender or age. This PWV may correspond to a carotid value since in ear PPG measurements are made very close to the carotid artery. Similarly, the pulse waveform measurements made on the arm can enable to extract features like the PRT and comparison to reference signal can also enable to obtain a PWV for the arms. Finally, the combination of the carotid over arm PWVs can enable to calculate a pulse wave velocity ratio. All these information can be obtained for each side of the body of the user 201.

The information presented here may be basic information that can be obtained with the system 202. However, additional information like arm SPO2 values could be obtained in case a SPO2 sensor is used to measure the pulse waveform of the arms. Similarly, the augmentation index and pulse rise time are cited here as extracted from the signals but additional features can be calculated and used for the algorithm.

FIG. 19 shows a schematic representation of a flowchart 1900 for calculating a cardiovascular risk e according to an embodiment.

First, the SPO2 values and PRT obtained on both ears of the user 201 can be compared to find if there is a significant difference. This could mean that one of the carotid arteries is narrowed or blocked (both SPO2 values and PRT have been found to be correlated with the stage of PAD). The maximum acceptable difference between those parameters should be previously determined during the development of the device enabling to determine the values a and 0 of Fig. 19. A similar comparison can be made for the arms in case the SPO2 sensor is used to obtain the pulse waveform at each arm.

Then, the augmentation indexes determined using the reconstructed arterial pulse waveform in the ears can be compared. The Al can give information about the arterial stiffness, therefore both the asymmetry and the absolute values are of interest for the carotid arteries. Indeed, an asymmetry, determined as a difference higher than a predetermined value obtained during calibration y, can be indicative of the beginning of carotid artery disease with one side getting stiffer than the other one. A value (calculated as mean, max or other combination of both side values, depending on symmetrical results) superior to a certain value dependent on age and gender and obtained by calibration, is indicative of a higher risk of carotid artery disease.

Then, the PWV determined for both carotid arteries can be compared. If it is found during development that the timing comparison is more accurate, then the determined PATs or PTTs can be compared instead of the PWV. PWV is an indication of arterial stiffness and can also enable to compare the arterial stiffness in both carotid arteries, thus, enabling to double check the information obtained with the augmentation index for more accuracy. Furthermore PWV is also expected to increase in case of narrowing of the artery or presence of a blockage. Therefore, an asymmetry in carotid PWVs can be indicative of a risk of having carotid artery disease. The asymmetry can be determined again by a side difference higher than a predetermined value 5 obtained during calibration. The carotid PWV in itself is also an interesting parameter. Therefore, its absolute value (determined depending on symmetrical result as average, maximum or other combination of both side values) can also be compared to a value 8 dependent of age and gender to determine if it presents a risk for the user 201. Finally, the value of the carotid PWV can be used to determine the blood pressure of the user 201 using a validated blood pressure monitor for calibration. Indeed, the carotid artery is an elastic artery, and, therefore, the relationship between blood pressure and pulse wave velocity is more stable than what could be obtained with peripheral arteries that suffer vasodilatation and vasoconstriction.

The PWV (or if found more accurate PATs or PTTs) obtained for both arms will also be compared. If an asymmetry is detected, then it means that there is a chance that the blood pressure values in both arms could be different, increasing the risk of having or developing PAD or CAD.

Finally, the PWV ratio (defined as the ratio between the carotid PWV over the arm PWV) can be calculated using a possible combination of both sides values depending on previous symmetrical results obtained. The PWV ratio is indicative of arterial stiffness mismatch between central and peripheral arteries. Therefore, a value indicative of a too important mismatch (defined as a value higher than a predetermined value r| dependent on age and gender or smaller than a predefined value or out of a predetermined range) would be indicative of a higher cardiovascular risk for the user 201.

The method 2100 presented here is only a possible way of analyzing the different measured parameters to obtain information of the risk of a user of having or developing a cardiovascular disease. A possibility is also to use a subset of the data depending on how the user 201 is using the system. For example, if the embodiment of Fig. 14 is used but without doing the steps 5 to 8 of Fig. 15, then an algorithm similar to the one shown in Fig. 20 could be used. In fact, Fig. 20 shows principle of the method 2100 using a subpart of the measured parameters, such as: right carotid parameters (SPO2, Al, PRT, PWV); left carotid parameters (SPO2, Al, PRT, PWV); asymmetry of carotid parameters (ASPO2, AAI, APRT, APWV) expressed as values or percentage; left arm parameters (PRT, PWV); the PWV ratio for one side.

All these parameters can be calculated and can be used in the algorithm but can also be shown to the user 201.

FIG. 21 shows a schematic representation of the method 2100 for calculating the cardiovascular risk e according to an embodiment.

The method 2100 comprises a step 2101 of measuring 2101 an ECG or PCG reference signal a; a step 2102 of measuring 2102 at least a first pulse waveform signal b at a first ear of a user, the first pulse waveform signal b having a timing clock synchronized with that of the reference signal a; a step 2103 of measuring 2103 at least a second pulse waveform signal c at a second ear of the user, the second pulse waveform signal c having a timing clock synchronized with that of the reference signal a; a step 2104 of measuring 2104 at least a third pulse waveform signal d at an arm, finger or wrist of the user 201, the third pulse waveform signal d having a timing clock synchronized with that of the reference signal a; and a step 2105 of calculating 2105 the cardiovascular risk e of the user 201 based on the measured pulse waveform signals b, c, d and the reference signal a.

The present invention has been described in conjunction with various embodiments as examples as well as implementations. However, other variations can be understood and effected by those persons skilled in the art and practicing the claimed invention, from the studies of the drawings, this disclosure and the independent claims. In the claims as well as in the description the word “comprising” does not exclude other elements or steps and the indefinite article “a” or “an” does not exclude a plurality. A single element or other unit may fulfill the functions of several entities or items recited in the claims. The mere fact that certain measures are recited in the mutual different dependent claims does not indicate that a combination of these measures cannot be used in an advantageous implementation.

Claims

1. A system (202) for detecting a cardiovascular risk (e) of a user (201), the system (202) being configured to: acquire a cardiac reference signal (a) by measuring an electrocardiogram, ECG or phonocardiogram, PCG; measure at least a first pulse waveform signal (b) at a first ear of the user (201), the first pulse waveform signal (b) having a timing clock synchronized with that of the reference signal (a); measure at least a second pulse waveform signal (c) at a second ear of the user (201), the second pulse waveform signal (c) having a timing clock synchronized with that of the reference signal (a); measure at least a third pulse waveform signal (d) at an arm, finger or wrist of the user (201), the third pulse waveform signal (d) having a timing clock synchronized with that of the reference signal (a); and calculate the cardiovascular risk (e) of the user (201) based on the measured pulse waveform signals (b, c, d) and the reference signal (a).

2. The system (202) of claim 1, wherein the system (202) is further configured to: determine a first set of features of a first side of the body of the user (201) based on the first pulse waveform signal (b) and the reference signal (a); determine a second set of features of a second side of the body of the user (201) based on the second pulse waveform signal (c) and the reference signal (a); determine one or more symmetry parameters based on the first set of features and the second set of features; and/or determine one or more timing parameters of the first set of features and the second set of features with respect to the reference signal (a); and the system being further configured to calculate the cardiovascular risk of the user (201) based on at least one of the one or more symmetry parameters and the one or more timing parameters.

3. The system (202) of claim 2, wherein the first set of features and/or the second set of features, respectively, comprises at least one of: a pulse rise time, PRT; an augmentation index, Al; a pulse arrival time, PAT, if the reference signal (a) is the ECG reference signal; and a pulse transit time, PTT, if the reference signal (a) is the PCG reference signal.

4. The system (202) of claim 3, wherein the system (202) is further configured to: determine a first pulse arrival time, PAT, or pulse transit time, PTT, related to a first carotid of the user (201) based on the first pulse waveform signal (b) and the reference signal (a) and deduce a first carotid pulse wave velocity, PWV, from it and biometric data of the user (201), such as gender, height and age; determine a second PAT or PTT related to a second carotid of the user (201) based on the second pulse waveform signal (c) and the reference signal (a) and deduce a second carotid PWV from it and biometric data of the user (201), such as gender, height and age; compare the first carotid pulse wave velocity with the second carotid pulse wave velocity; and calculate the cardiovascular risk (e) of the user (201) based on the result of the comparison.

5. The system (202) of claim 4, wherein the system (202) is further configured to: determine a first pulse arrival time, PAT, or pulse transit time, PTT, related to a first arm of the user (201) based on the third pulse waveform signal (d) and the reference signal (a) and deduce a first arm pulse wave velocity, PWV, from it and biometric data of the user (201), such as gender, height and age; determine a second PAT or PTT related to a second arm of the user (201) based on the third (d) or a fourth pulse waveform signal and the reference signal (a) and deduce a second arm PWV from it and biometric data of the user (201), such as gender, height and age; compare the first arm pulse wave velocity with the second arm pulse wave velocity; and calculate the cardiovascular risk (e) of the user (201) based on the comparison.

6. The system (202) of claim 5, wherein the system (202) is further configured to: determine a ratio between a carotid pulse wave velocity and an arm pulse wave velocity using at least one of the carotid pulse wave velocities calculated by the system (202) and at least one of the arm pulse wave velocities calculated by the system (202); calculate the cardiovascular risk (e) of the user (201) based on the pulse wave velocity ratio.

7. The system (202) of anyone of the preceding claims, wherein the system (202) further comprises: a first ear-mounted device (401); a second ear-mounted device (402); wherein the first ear-mounted device (401) and the second ear-mounted device are (402), respectively, configured to measure the first pulse waveform signal (b) and the second pulse waveform signal (c), either simultaneously or consecutively.

8. The system (202) of claim 7, wherein the first ear-mounted device (401) or the second ear-mounted device (402) comprises a processor configured to calculate the cardiovascular risk (e); and/or the first ear-mounted device (401) or the second ear-mounted device (402) is configured to measure the reference signal (a) and/or the third pulse waveform signal (d).

9. The system (202) of claim 7 or 8, wherein the first ear-mounted device (401) and the second ear-mounted device (402) are configured to communicate with each other, in order to synchronize the first pulse waveform signal (b) and the second pulse waveform signal (c) with the reference signal (a).

10. The system (202) of any one of the preceding claims, wherein the system (202) comprises a smartphone (801) or smartwatch (1401), the smartphone (801) or smartwatch (1401) being configured to measure the third pulse waveform signal (d), and optionally to measure the reference signal (a).

11. The system (202) of claim 10 and of one of the claims 7 to 9, wherein the smartwatch (1401) or smartphone (801) is configured to communicate with the first ear-mounted device (401) and/or the second ear-mounted device (402), in order to synchronize the third pulse waveform signal (d) and the first (b) and/or second pulse waveform signal (c) with the reference signal (a).

12. The system (202) of any one of the claims 8 to 11, wherein at least one of the first earmounted device (401) and the second ear-mounted device (402) comprises an ECG sensor or a PCG sensor configured to measure the reference signal (a).

13. The system (202) of any one of the preceding claims 7 to 12, wherein at least one of the first ear-mounted device (401) and/or the second ear-mounted device (402) comprises a pulse waveform measuring sensor configured to measure the third pulse waveform signal (d).

14. The system (202) of anyone of the preceding claims, wherein the system (202) further comprises : a multi-wavelength photoplethysmography, PPG, sensor comprising at least a red, infrared and green LED.

15. The system (202) of claim 14, wherein the system (202) is further configured to: determine an oxygen saturation, SPO2, value using the red and infrared PPG sensors; and determine an arterial pulse waveform on the basis of the at least red, infrared and green PPG signals.

16. The system (202) of claim 15, wherein the system (202) is further configured to: measure, in the first ear of the user, a first oxygen saturation, SPO2, value; measure, in the second ear of the user, a second oxygen saturation, SPO2, value; and calculate the cardiovascular risk (e) based on a comparison of the first SPO2 value and on the second SPO2 value.

17. A method (2100) for detecting a cardiovascular risk (e) of a user (201), the method (2100) comprising: measuring (2101) an electrocardiogram, ECG, or phonocardiogram, PCG, reference signal (a); measuring (2102) at least a first pulse waveform signal (b) at a first ear of a user, the first pulse waveform signal (b) having a timing clock synchronized with that of the reference signal (a); measuring (2103) at least a second pulse waveform signal (c) at a second ear of the user, the second pulse waveform signal (c) having a timing clock synchronized with that of the reference signal (a); measuring (2104) at least a third pulse waveform signal (d) at an arm, finger or wrist of the user (201), the third pulse waveform signal (d) having a timing clock synchronized with that of the reference signal (a); and calculating (2105) a cardiovascular risk (e) of the user (201) based on the measured pulse waveform signals (b, c, d) and the reference signal (a).