Apparatus and method for determining cardiovascular risk score of user

Abstract

A method comprising: providing a device configured to measure a cardiovascular signal of a user; measuring, using the device, a cardiovascular signal during a monitoring period having a duration of at least 24 hours; the monitoring period is subdivided into monitoring periods, each monitoring period having a duration of 24 hours and comprising a plurality of measurement periods; determining, for each measurement cycle, a cardiovascular value; aggregating the cardiovascular values determined for the respective measurement cycles for each monitoring period into a population of cardiovascular parameters; constructing a 24-hour diurnal map of the parameter group relative to the corresponding measurement period; determining a plurality of physiological parameters of the user using the diurnal map; and calculating a cardiovascular risk score for the user using the determined physiological parameters.

Description

Technical Field

[0001] The present invention relates to an apparatus and method for determining a cardiovascular risk score of a user. Background Art

[0002] Hypertension remains a leading risk factor for mortality worldwide. Despite its prevalence, successful management of blood pressure (BP) remains elusive, and part of the difficulty lies in the tool that remains used to diagnose, measure, and treat hypertension: the sphygmomanometer, introduced in 1867 by Samuel Siegfried Karl von Basch. In recent years, there has been a proliferation of devices that attempt to provide BP estimates without a cuff, overcoming many of the limitations of cuff-based BP monitors. Unfortunately, fundamental technical differences between traditional BP cuffs and newer cuffless devices, as well as a hesitation to change well-implemented standards, still generate understandable skepticism and reluctance regarding the adoption of cuffless BP monitors in clinical practice.

[0003] The scale of the hypertension problem is difficult to fathom—estimated to affect more than 1.3 billion people worldwide. Yet, the current paradigm for diagnosis, treatment, and monitoring of high blood pressure (HTN) results in poor control rates. Maintaining the status quo carries significant risks: for more than 40 years, HTN has been the single largest contributor to cardiovascular disease and death, costs the U.S. healthcare system an estimated $131 billion annually, and is the leading preventable risk factor for premature death worldwide. Given the enormous potential benefits, any changes that could improve hypertension care must be fully explored.

[0004] The past two decades have seen a dramatic digital transformation of healthcare, primarily through the widespread adoption of EHRs and the accompanying multitude of digital health applications that enable more robust, real-time, and actionable data and information exchange between patients and providers. In fact, 20 years ago—seven years before the first iPhone was released—the Institute of Medicine (IoM) wrote a book in which the authors highlighted the potential of computer-assisted (now known as digital health or mHealth) tools to help automatically transfer clinical data to clinicians to both improve clinical care and further our understanding of disease.

[0005] Two decades later, technology is beginning to deliver on the IoM’s vision and call to action. Cuffless BP technology holds the promise to improve the diagnosis, treatment, and monitoring of hypertension, with the potential to benefit millions of people with hypertension. Innovation has long been focused on treating the disease once it manifests, but technology now offers the opportunity to help prevent uncontrolled HTN and all its attendant risks. The commercialization of cuffless BP devices will allow us to address many of the behavioral and practical challenges of treating a widespread, chronic disease for which progress has long eluded. Treatment of hypertension is constrained by the limitations of both office-based BP measurements and cuff-based home measurements. It is reasonable to expect that large-scale adoption of cuffless BP devices will realize their benefits and lead to better control of hypertension worldwide.

[0006] In 1948, one of the most important cardiovascular risk studies was initiated in Framingham, Massachusetts. There is little doubt that the insights gained from this study were groundbreaking over the ensuing three decades. As part of the study protocol, office BP measurements were determined by the only technology available at the time—auscultation of Korotkoff sounds to estimate systolic and diastolic BP. Subjects sat with their backs against a chair, and measurements were taken on the left arm only. To this day, all major guidelines recommend that BP measurements be taken as originally described by the Framingham Study methodology, with the notable exception of the shift from manual mercury manometers to automated oscillometric devices.

[0007] This BP collection methodology has been enshrined in collective medical thought as the standard for estimating BP, following decades of clinical trials and the goal of cuff measurement in practice guidelines. Further, it supports the assumption that individuals have physiologically stable and predictable BP. Even expert consensus documents, such as the American Heart Association guidelines, state that "it is generally accepted that routine clinical readings, in the correct setting, are surrogate markers of the patient's true BP, which is considered to be the average over a long period of time and is considered the most important component of BP in determining its adverse effects."

[0008] However, as the understanding of hypertension has evolved, it has become clear that BP continues to change and adapt over a 24-hour period based on lifestyle, daily activities, medications, physical / emotional stressors, and changes in body position. Expert guidelines also recommend that office readings should be confirmed by out-of-office readings in subsequent weeks and months, implying that the true nature of an individual's BP pattern in daily life cannot be fully estimated by relaxing for 5 minutes in a quiet, climate-controlled environment without exercise, talking, caffeine, noise, and with feet flat on the floor.

[0009] Ambulatory BP monitors (ABPM) are beginning to show some of this variability, but usually only over 24 hours.Recently, the development of continuous cuffless BP devices has enabled for the first time a much more representative longitudinal depiction of an individual's BP.

[0010] The importance of out-of-office BP measurement is recognized in all major HTN guidelines for confirming office BP readings. ABPM is considered the gold standard for out-of-office BP measurement. Sadly, however, ABPM remains underutilized for a variety of reasons. Although ABPM may be the currently recommended tool for monitoring out-of-office BP, it is almost never used in the United States. The percentage of Medicare beneficiaries—for whom the estimated prevalence of HTN is 50%—with ABPM claims is only ~0.1% per year. In China, only 1.6% of primary care providers surveyed reported using ABPM to diagnose HTN. A simpler, less expensive, and more widely available solution for BP monitoring would be of significant benefit to both providers and patients.

[0011] All major HTN guidelines also recommend home BP monitoring (HBPM) as a key adjunct to the diagnosis, monitoring, and management of HTN. However, in practice, it is difficult for patients to monitor their BP at home and send meaningful data. In addition, patients require training to follow the same standard procedures for BP measurement as for office BP measurement. Despite the availability of relatively inexpensive home BP cuffs, the percentage of active HBPM among real-world patients is surprisingly low. Half of hypertensive patients report never checking their BP at home, 10% check BP less than once a month, and only 24% of hypertensive patients report checking BP >1 time / week. The data suggest that although HBPM is routinely recommended by expert panels and consensus guidelines, in reality most patients with HTN do not perform it, and certainly not twice a day for at least seven days as recommended.

[0012] There are many possible explanations for the significant gap between recommendations and real-world practice. A study completed in 2017 explored the barriers that primary care providers face in recommending HBPM to their patients. More than two-thirds of respondents cited one or more of the following reasons as barriers to obtaining HBPM data:

[0013] Patients were unable to complete HBPM due to low health literacy and time requirements.

[0014] The invasiveness of the test, the routine requirement, and the requirement to bring the HBPM to the office.

[0015] Inaccurate test results due to patient nonadherence to HBPM protocols (e.g., incorrect cuff size, poor timing of BP readings, failure to record readings, “cherry-picked” normal BP readings displayed to physicians.

[0016] Inaccurate results due to patient factors such as physical constitution.

[0017] Test results or cuff inflation may increase patient anxiety and therefore affect accuracy. Summary of the invention

[0018] It is known that blood pressure (BP) changes over time and follows a circadian rhythm. It typically exhibits higher systolic and diastolic values ​​during the day and decreases at night and during sleep. This nocturnal drop in blood pressure is often referred to as "night-dipping". However, the exact trend of this night-dipping may vary greatly between individuals and may even change slowly over time. Some parameters of this trend, such as the amplitude of the night-dipping or the slope of the morning rise (also known as the "morning surge"), are clinically relevant because they are associated with different cardiovascular risk factors.

[0019] The present disclosure relates to a method for determining a cardiovascular risk score of a user, comprising:

[0020] providing an apparatus configured to measure at least a cardiovascular signal of a user;

[0021] measuring said at least cardiovascular signal during a monitoring period having a duration of at least 24 hours;

[0022] The monitoring cycle is subdivided into at least one monitoring period, wherein the at least one monitoring period has a duration of 24 hours and comprises a plurality of measurement cycles;

[0023] determining a cardiovascular value for each measurement cycle of a plurality of measurement cycles within the at least one monitoring period;

[0024] aggregating the cardiovascular values ​​determined for a given measurement cycle of each of the at least one monitoring period into a cardiovascular parameter group;

[0025] For each of the at least one monitoring period, constructing a 24-hour circadian graph of the cardiovascular parameter group relative to the corresponding measurement period;

[0026] Calculating multiple physiological parameters of the user based on the diurnal graph; and

[0027] A cardiovascular risk score of the user is calculated based on the determined physiological parameters.

[0028] The present disclosure also relates to an apparatus for determining a cardiovascular risk score of a user.

[0029] The present disclosure also relates to a computer program comprising instructions which, when a computer executes the program, cause the computer to implement the steps of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Exemplary embodiments are disclosed in the specification and illustrated in the accompanying drawings, in which:

[0031] Figure 1 a- Figure 1 d illustrates an apparatus and method for calculating a cardiovascular risk score of a user;

[0032] Figures 2 to 6 An example of a possible fitting model is shown, namely a trapezoidal model ( Figure 2 ), rectangular model( Figure 3 ), Gaussian model( Figure 4 ), skewed Gaussian model ( Figure 5 ) and the skew-flat Gaussian model ( Figure 6 );

[0033] Figure 7 Group comparisons of the mean daytime BP values ​​obtained by device versus HBPM measurements obtained during device initialization on the same day were reported;

[0034] Figure 8 a and Figure 8 b illustrates the poor within-subject reproducibility of ABPM examinations: because individual diurnal BP excursions vary over time, by choosing an arbitrary measurement day(s), ABPM examinations may generate phenotypes that misrepresent the patient's underlying BP phenotype;

[0035] Fig. 9 Comparison of estimates of nocturnal descent measured by ABPM and by device:

[0036] Fig.10 a to Fig.10 j shows a new generation of dynamic BP control metrics that can be generated based on data captured from the device;

[0037] Fig.11 An apparatus for determining a cardiovascular risk score of a user according to an embodiment is shown; and

[0038] Fig.12 is a cross-sectional view of a device according to an embodiment. DETAILED DESCRIPTION

[0039] The present disclosure describes a method and apparatus for constructing a BP circadian rhythm model for a specific user given his / her BP measurement history over a period of time, and for deriving a set of relevant parameters from the model, which can be used to assess the user's cardiovascular risk score and further communicate the cardiovascular risk score to the user.

[0040] Specifically (see Figure 1 a to Figure 1 b), in the monitoring period Tm( Figure 1b) during which the cardiovascular signal of the user is measured by using the device 10 configured to measure such a cardiovascular signal. The monitoring period Tm has a duration of at least 24 hours and is subdivided into a plurality of measurement periods (not shown). For each measurement period, a cardiovascular parameter of the user is determined.

[0041] Each cardiovascular parameter determined for a respective measurement period is aggregated into a cardiovascular parameter group.

[0042] A circadian graph of the cardiovascular parameter cluster relative to the corresponding measurement period can then be constructed.

[0043] For a given user, 24 / 7 (24 hours a day, 7 days a week) BP systolic and diastolic values ​​for a continuous time period (e.g., two weeks) are aggregated together. In terms of the temporality of each measurement, the exact day of each measurement is ignored, and only the time information (hours and minutes) is retained. This process combines all BP measurements (systolic or diastolic, respectively) for the entire survey period centered at midnight within a single 24-hour cycle. For each one-hour measurement period (e.g., from 7:00am to 7:59am), any BP measurement outside the limits defined by the median ±2x IQR (interquartile range) for the period is discarded before further analysis (outlier removal).

[0044] On the remaining data, a constrained piecewise linear model was fitted via least squares optimization. The model was constrained in such a way that the BP trend was constrained to be constant during the nighttime decline and to be a different constant during the daytime, and two linear transitions (ramps) between these two regimes were allowed.

[0045] Once established, the model provides a set of six physiological parameters that uniquely describe an individual's BP profile (see Figure 1 b), namely:

[0046] Daytime BP value (Y0),

[0047] Absolute night drop amplitude (ampl),

[0048] The timing of the start of the decline (X0),

[0049] Duration of the previous night ramp (dl),

[0050] The duration of the decreasing plateau (nl), and

[0051] Duration of the post-nocturnal ramp (al).

[0052] Physiological parameters can be further combined into additional clinically relevant parameters such as (non-exhaustive):

[0053] Nighttime BP value,

[0054] Relative night drop,

[0055] Full night fall duration,

[0056] Time in Target Range (TTR), or

[0057] Morning peak slope.

[0058] The calculated physiological parameters may further allow determination of the user's blood pressure phenotype, including classifications such as: true normotensive, white coat hypertension, masked hypertension, sustained hypertension, hypotensive, night-falling, night-rising.

[0059] Physiological parameters and / or clinically relevant parameters and / or blood pressure phenotypes may further be combined into additional cardiovascular risk scores.

[0060] exist Figure 1 a to Figure 1 In the embodiment illustrated in FIG. d, a method for calculating a cardiovascular risk score of a user includes:

[0061] An apparatus 10 ( Figure 1 a);

[0062] Using the device 10 for measuring said at least cardiovascular signal ( Figure 1 b);

[0063] The monitoring period Tm is subdivided into at least one monitoring period Ts, wherein the at least one monitoring period Ts has a duration of 24 (twenty four) hours and comprises a plurality of measurement periods;

[0064] determining a cardiovascular value for each measurement cycle of a plurality of measurement cycles within the at least one monitoring period Ts;

[0065] aggregating the cardiovascular values ​​determined for a given measurement cycle of each of the at least one monitoring period Ts into a cardiovascular parameter group;

[0066] constructing, for each of the at least one monitoring period Ts, a 24 (twenty four) hour diurnal diagram of the parameter group relative to the corresponding measurement period;

[0067] Use circadian graphs to determine multiple physiological parameters of the user ( Figure 1 c); and

[0068] The determined physiological parameters are used to calculate the user's cardiovascular risk score ( Figure 1 d).

[0069] The measurement period corresponds to the time period during which the cardiovascular signal is measured by the device 10 .

[0070] exist Figure 1 a to Figure 1 D, Figure 1 a shows the device 10 represented as a cuffless blood pressure optical sensor at the wrist, Figure 1 b shows a cardiovascular value (in this case, a BP value) calculated from a cardiovascular signal (in this case, a PPG signal, not shown) provided to the device 10, and Figure 1 c shows the total monitoring period Tm (see Figure 2 ) is aggregated into a cardiovascular parameter group, and a 24-hour circadian graph is constructed, and the model is fitted to the circadian graph to determine multiple physiological parameters. Based on multiple physiological parameters, the user's cardiovascular risk score ( Figure 1 d).

[0071] In one aspect, the monitoring period Tm has a duration of 48 hours, seven days, one month or one year. Note that in the case where the monitoring period Tm has a duration of 24 hours, its duration is equal to the duration of the monitoring period Ts.

[0072] In another aspect, the measurement period can have a duration of at least 10 seconds or 30 seconds, ie, one minute, 5 minutes, one hour, two hours, four hours, or six hours.

[0073] exist Figure 1 In the example of b, the monitoring period Tm is greater than 24 hours. For example, the monitoring period Tm corresponds to two weeks (14 days). In this case, the monitoring period Tm includes more than one monitoring period Ts (for a 2-week monitoring period Tm, 14 monitoring periods of 24 hours duration).

[0074] Each 24-hour monitoring period Ts includes a plurality of measurement cycles. For example, a monitoring period Ts includes twenty-four measurement cycles of one-hour duration. Thus, for each measurement cycle corresponding to a time of day, a plurality of cardiovascular values ​​are calculated (for each time of day, twenty-four cardiovascular values ​​are calculated).

[0075] Therefore, for each time of day corresponding to each measurement period, multiple cardiovascular values ​​are aggregated into a cardiovascular parameter group (see Figure 1 c).

[0076] In one aspect, cardiovascular signals may be measured during workdays and non-workdays, respectively.

[0077] Cardiovascular signals may include physical activity measurements, sleep amount and sleep quality measurements, and any of values ​​derived from electrocardiogram signals, values ​​derived from photoplethysmography signals, values ​​derived from bioimpedance signals, values ​​derived from ultrasound signals, or values ​​derived from any arterial pulsation signal (such as pulse pressure).

[0078] The cardiovascular value may be calculated based on the cardiovascular signal measured during the measurement cycle. Calculating the cardiovascular value may include identifying a plurality of pulses of the cardiovascular signal. Calculating the cardiovascular value may further include determining at least one feature of the identified pulses and calculating the cardiovascular value based on the at least one feature. For example, a method for calculating the cardiovascular value is proposed in European Patent Publication EP3226758. In other embodiments, the cardiovascular value may also be calculated without identifying a plurality of consecutive pulses of the cardiovascular signal and without applying a non-parametric algorithm or a machine learning algorithm.

[0079] The step of determining a cardiovascular value may include calculating a cardiovascular value based on each cardiovascular signal. The cardiovascular value may include any one of a systolic BP value, a diastolic BP value, an average BP value, a heart rate value, or a blood glucose value.

[0080] Cardiovascular values ​​may further include pulse pressure, central pulse wave velocity, peripheral pulse wave velocity, arterial stiffness, aortic pulse transit time, augmentation index, stroke volume, stroke volume variability, pulse pressure variability, cardiac output, systemic vascular resistance, venous pressure, systemic hemodynamic parameters, pulmonary hemodynamic parameters, cerebral hemodynamic parameters, heart rate, heart rate variability, beat-to-beat interval, arrhythmia detection, ejection duration, SpO2, SpHb, SpMet, SpCO, respiratory rate, tidal volume, and any of general cardiovascular and health indicators.

[0081] The step of aggregating cardiovascular values ​​may comprise grouping the cardiovascular values ​​determined by the apparatus 10 during a given measurement period into cardiovascular parameter groups (see Figure 1 b and Figure 1 c).

[0082] The step of constructing a 24-hour diurnal graph may comprise plotting the cardiovascular parameter cluster calculated for each measurement period as a function of the day (24-hour period) (e.g., from midnight to midnight (see Figure 1 c)) as a function of time.

[0083] In one aspect, the cardiovascular parameter cluster can have a duration of 10 seconds, 30 seconds, 1 minute, 5 minutes, 10 minutes, or 30 minutes.

[0084] The step of using the circadian graph may include calculating a plurality of physiological parameters of the user based on the circadian graph.

[0085] In one aspect, the step of using the diurnal graph may include calculating, for each cardiovascular parameter cluster, a cardiovascular representative value among the cardiovascular values ​​aggregated into the cardiovascular parameter cluster.

[0086] In another aspect, the step of using the diurnal graph may include calculating a cardiovascular representative value, including classifying the cardiovascular parameters as inner or outer values, and calculating the cardiovascular representative value using only inner cardiovascular parameters.

[0087] In one aspect, calculating the cardiovascular representative value may include classifying the cardiovascular parameters as inliers or outliers. The cardiovascular parameters may be weighted according to the probability that they are inliers or outliers.

[0088] In one aspect, determining a plurality of physiological parameters of the user may include the step of fitting a model to the circadian graph, the physiological parameters corresponding to parameters of the model.

[0089] The model can be one of a linear model, a nonlinear model, a constrained model, and an unconstrained model fitted via least squares optimization.

[0090] In one aspect, the constraint model may include time constraints on the duration of the physiological parameters (e.g., the night-time plateau period of blood pressure decline cannot be longer than 12 hours, or the duration of the previous night slope cannot be longer than the duration of the decline plateau period), or may include amplitude constraints on the physiological parameters (e.g., the daytime blood pressure value cannot be higher than 200 mmHg, or the SBP morning peak cannot be greater than 50 mmHg per hour).

[0091] like Figure 2 As shown in , the physiological parameters may include any one of the daytime BP value (Y0), the absolute night drop amplitude (ampl), the timing of the start of the drop (X0), the duration of the pre-night slope (dl), the duration of the drop plateau (nl), or the duration of the post-night slope (al).

[0092] Figures 2 to 6 An example of a possible fitting model is shown, namely a trapezoidal model including six parameters Y0, X0, ampl, dl, nl, al ( Figure 2 ), including a rectangular model with four parameters Y0, X0, ampl, and nl ( Figure 3 ), including the Gaussian model with four parameters Y0, X0, sd, and ampl ( Figure 4 ), including the skewed Gaussian model with five parameters Y0, X0, sd0, sd1, and ampl ( Figure 5 ), and a skew-flat Gaussian model consisting of six parameters Y0, X0, sd0, sd1, ampl, and nl ( Figure 6 ).

[0093] The physiological parameters may include differences in physiological parameters calculated during the work week and weekends, or more generally, differences in physiological parameters calculated during work days and non-work days.

[0094] Physiological parameters may further include the temporal dynamics of any physiological parameter. The temporal dynamics of a physiological parameter may include day-to-day variability of a parameter, spread of a parameter over a number of days, trend of a parameter over a number of days, or days when a parameter is above or below a threshold.

[0095] In one aspect, the model further uses non-physiological parameters. Non-physiological parameters include parameters such as: user's geographic location including altitude, weather forecasts and observations, heat waves, cold waves, travel information, pollution information, public health information including the status of infectious diseases in the user's location, time of year, allergy information, daylight patterns, social information, education level, family status (marital status, number and age of children), financial information, political interests and views, professional status including level of responsibility, contract type, regularity of working hours, information from the calendar including working days / non-working days and holidays, workload, work schedule, eating patterns, exercise / activity patterns, sedentary level, leisure information including consumption of caffeine, alcohol, drugs, medications (not limited to antihypertensive drugs), health status of the user and close relatives, type and size of residence, owner or tenant pets, use of social media, religious practices, questionnaires on mood and general state.

[0096] The method may further include combining at least two physiological parameters to obtain one or more correlated physiological parameters. The combination may include adding, multiplying or dividing at least two physiological parameters. Such combination may further include calculating a correlation coefficient or synchronization coefficient of two or more physiological parameters.

[0097] Relevant physiological parameters may include any of the following: daytime BP value, nighttime BP value, relative nighttime drop, duration of complete nighttime drop, time in target range (TTR), BP variability pattern, morning surge slope, nighttime SBP, nighttime DBP, nighttime HR, SBP drop, DBP drop, HR drop, SBP morning surge, DBP morning surge, HR morning surge, duration of SBP drop, duration of DBP drop, duration of HR drop, SBP / DBP / HR synchronization, response to medication type, BP medication compliance, patient engagement measures, or response to lifestyle interventions.

[0098] In one aspect, the relevant physiological parameters may include a blood pressure phenotype of the user.

[0099] In one aspect, relevant physiological parameters include any of true normotension, white coat hypertension, masked hypertension, sustained hypertension, hypotension, night drop, night rise, or a phenotype predictive of response to a particular drug or treatment such as renal denervation.

[0100] The step of calculating a cardiovascular risk score for the user using the determined physiological parameters may include calculating a cardiovascular risk score for the user based on the determined physiological parameters.

[0101] In one aspect, calculating a cardiovascular risk score for a user includes using user data.

[0102] User data may include any of the following: age, weight, height, sex, race, lipid levels, diabetes status, smoking, CT calcium (Agatston score), family history, risk genetic markers, actigraphy information, exercise information, diet information, stress level, general feeling, hormone data, menstrual cycle information, medication intake, weekday / weekend information, season information, sleep quality information, bedtime pattern, or any parameter used to calculate cardiovascular risk score in clinical guidelines such as ACC / AHA guidelines, ESC guidelines, or MESA database. User data may also be any non-physiological parameter.

[0103] User data can be provided manually by the user or automatically integrated from external systems.

[0104] The cardiovascular risk score may be any of a 10-year (ten-year) cardiovascular disease risk, a 10-year heart attack risk, a 10-year stroke risk, or any other clinically relevant cardiovascular risk score.

[0105] In one aspect, the cardiovascular parameter may be at least any one of a blood pressure value, a heart rate value, a cardiac output value, a blood glucose value, a physical activity measure, or a sleep quantity and sleep quality measure.

[0106] Also disclosed is an apparatus for determining a cardiovascular risk score of a user.

[0107] Device 10 (see Fig.11 ) comprises a measuring module 20, which is configured to measure the cardiovascular signals of the user during a monitoring cycle Tm having a duration of at least 24 (twenty four) hours, wherein the monitoring cycle Tm is subdivided into monitoring periods Ts, each monitoring period Ts having a duration of 24 (twenty four) hours and comprising a plurality of measurement cycles.

[0108] The device 10 further includes a processor 30, which is configured to determine a cardiovascular value based on the cardiovascular signal measured for each measurement cycle, aggregate the cardiovascular values ​​determined for the corresponding measurement cycle of each monitoring period Ts into a cardiovascular parameter group, construct a 24 (twenty-four) hour circadian graph of the parameter group relative to the corresponding measurement cycle, use the circadian graph to determine multiple physiological parameters of the user, and use the determined physiological parameters to calculate the user's cardiovascular risk score.

[0109] The device further comprises an interface 40 for displaying and / or transmitting the calculated cardiovascular risk score. The interface 40 is operably connected to the processor 30.

[0110] The processor 30 may be configured to perform the following steps: determining cardiovascular values, aggregating cardiovascular values, constructing a 24-hour circadian graph, calculating a plurality of physiological parameters of the user, and calculating a cardiovascular risk score for the user.

[0111] The device 10 may be operably connected to wired or wireless communication circuitry, which may include WiFi or Bluetooth or cellular support. The device 10 may further be operably connected to a memory.

[0112] The interface 40 may include an application on a smartphone, tablet, computer, smart watch, or any portable device.

[0113] Interface 40 may be configured to generate external signals directed to the user, for example, to provide guidance on how to optimize the user's calculated risk by suggesting lifestyle, medication, or treatment modifications.

[0114] The interface 40 may also be configured to manually or automatically input non-physiological data or user data.

[0115] The interface can be positioned close to the user or far away from the user.

[0116] In one aspect, the measurement module 20 may be configured to automatically measure cardiovascular signals without user interaction.

[0117] The measurement module 20 may include any of the following: an arterial pulsation sensor, such as a galvanic skin response (GSR) sensor array, a bioimpedance (BioZ) sensor array, an electrocardiogram sensor (ECG), a sensor based on radio frequency (RF) detection, a radar sensor, a mechanical sensor, a pressure sensor, an invasive sensor, an intra-arterial sensor, a minimally invasive sensor, a subcutaneous sensor, a tensiometer, a strain sensor, a plethysmographic sensor, a microphone, an ultrasound sensor, a capacitive sensor, an electromagnetic sensor, a Raman sensor, or any sensor capable of measuring a pulsation signal from the capillary bed of the skin or from any other part of the arterial tree. The cardiovascular signal measured by the device 10 via the measurement module 20 may therefore correspond to the signal measured by.

[0118] The apparatus 10 may include a wearable device. Fig.12 A possible configuration of the apparatus 10 as a wearable device is illustrated in a cross-sectional view of FIG. The apparatus 10 may include a wristband 15, which includes a measurement module 20. The measurement module 20 may include at least one pulsation sensing unit 21. For example, the measurement module 20 includes four pulsation sensing units 21, which are distributed along the inner side of the outer circumference of the wristband 15 so as to contact the user's wrist skin when the apparatus 10 is worn. Other arrangements of the pulsation sensing units 21 on the wristband 15 are possible.

[0119] In one embodiment, the pulsation sensing unit 21 may include a photoplethysmography (PPG) sensor array that can measure arterial pulsation, arterial diameter, blood flow, and / or blood content. In this case, the cardiovascular signal is a photoplethysmographic signal. In this embodiment, the pulsation sensing unit 21 may be arranged on the wristband 15 so that the optical sensor array 21 straddles or otherwise addresses an artery, such as the ulnar artery 111 near the ulna 113, or the radial artery 112 near the radius 114 (e.g., Fig.12 ) or any arterial vascular bed 117 in the skin of the wrist.

[0120] In an embodiment, the apparatus 10 further comprises a trigger module 50 (see Fig.11 ), which is configured to start or stop a measurement cycle through the measurement module 20.

[0121] The trigger module 50 can control the measurement module 20 according to the trigger parameter. The trigger parameter can be user-specific. Examples of trigger parameters can include trigger signals, such as motion signals representing user movement. Such motion signals can be measured by using a motion sensor 60 placed on the user (e.g., on the device 10). The motion sensor 60 can include any one of an inertial measurement unit (IMU), an accelerometer, a gyroscope, a magnetometer, or a combination of these devices.

[0122] In one aspect, the cardiovascular parameter is a blood pressure (BP) value.

[0123] Blood pressure values ​​may be measured by using cuff-based BP measurement techniques and / or by using optical sensors. Blood pressure values ​​are measured at the wrist.

[0124] The blood pressure value includes at least one of systolic BP, diastolic BP or average BP.

[0125] The device 10 disclosed herein allows for the transmission of blood pressure values ​​in a continuous manner, for example, during the day and night.

[0126] The apparatus 10 disclosed herein corresponds to a cuffless BP monitoring device. Compared to traditional methods, a cuffless BP device has the potential to address many practical and behavioral issues, to overcome the barriers to recommended routine BP monitoring, and to obtain significantly more BP data. The ability to collect continuous BP readings at home, outside the clinic, during daily activities, and while sleeping over a period of weeks, months, and years provides patients and providers with a much more complete assessment of BP than intermittent examinations performed in controlled locations and environments (which only give doctors and patients a limited understanding of the full performance of BP).

[0127] The ability to take continuous BP readings outside of the clinic and during the day and night has traditionally been limited because available monitoring technologies require the cuff to be inflated for each measurement. Devices that do not require the cuff to be inflated can overcome most of the inherent limitations in traditional BP cuff monitors.

[0128] The device 10 allows for the non-invasive determination of an individual's BP without causing any arterial occlusion.

[0129] Device 10 may be placed at a body location such as the wrist, fingertip, chest, ear, forehead, or a combination.

[0130] The device 10 provides an indirect BP estimation, which relies on the analysis of the arterial pulse at one or more body locations using sensors that do not apply pressure to the location. The device 10 does not provide a direct pressure measurement, but rather a quantity calculated by a computer program based on the waveform analysis of the pressure pulse, which is typically mapped to a BP value after an initialization phase. Several sensor technologies (sensing unit 21) are currently used to capture the waveform of the pressure pulse, ranging from optical sensors (evaluating the pulsation of skin arterioles via reflection or transmission photoplethysmography sensors), camera sensors (evaluating the pulsation of skin arterioles via reflection video-based photoplethysmography), biopotential sensors (evaluating different electromagnetic features of cardiac activity, or evaluating arterial pulsation based on impedance plethysmography signals at different body locations), radar sensors (evaluating arterial pulsation at different body locations based on radar reflections) and tension sensors (evaluating the pulsation of surface arteries by sensing the displacement of the skin). Depending on the number of body locations on the patient's body where the pressure pulse is captured, the waveform analysis is performed based on a pulse wave velocity algorithm (typically when at least two body locations are involved) or based on a pulse wave analysis algorithm (typically when a single body location is involved). Since no pressure measurement is involved in the evaluation of this pulse waveform, most cuffless BP monitors still require an initialization procedure that involves the use of an oscillometric device to provide information in "mmHg".

[0131] The main attractive feature of the device 10 is the potential to provide significantly more BP data points. The ability to collect continuous BP readings at home, outside the clinic, in daily life, and at night and during sleep over a period of days to years gives patients and providers a much more representative BP assessment than occasional cuff estimates. A snapshot of BP measured at a certain point in time in the clinic or by HBPM represents only a small part of the complete dynamic data set of BP. Without this data, doctors and patients are essentially in the dark about the true nature of BP.

[0132] To illustrate these limitations, real-world systolic BP (SBP) data from male subjects were recorded over a two-month period using device 10. The data demonstrated significant differences between simulated office readings, occasional home BP monitoring, and ambulatory BP monitoring. Office estimates implied absolute SBP values ​​that were significantly higher than mean and did not capture longitudinal BP data, which is most evident and prevalent in white coat and latent HTN syndrome (up to 40% of individuals). Home BP estimates of SBP—when performed routinely—may correlate with overall mean values ​​but do not capture daily and diurnal variability. Finally, ambulatory BP monitoring only reveals information during very short (24 or 48 hour) periods. All three traditional methods of BP estimation represent only a modest understanding of dynamic BP, which is well demonstrated by device 10.

[0133] Compared to other existing technologies, the device 10 can provide significantly more BP readings, better demonstrate BP variability, and provide nighttime BP measurements, all of which have important clinical implications. However, clinicians may wonder how the daytime average BP provided by the device 10 compares to HBPM readings performed on the same day? To answer this question, anonymized data from 2928 users of the device 10 were analyzed offline ( Figure 7 a and 7b). The analysis compared the daytime BP data (between 8am and 8pm) measured by the device 10 on the day of the initialization procedure against the brachial cuff BP measurements (HBPM) obtained during the same initialization procedure. The analysis was repeated for both systolic BP and diastolic BP. In both cases, paired T tests showed that the difference in the measurements under the two modalities was statistically significant (p<0.001 for both), although the difference (2.25 mmHg for SBP and 0.44 mmHg for DBP) was below the resolution and margin of error of any automated oscillometric BP monitor.

[0134] Figure 7 a and Figure 7 b shows a population comparison of daytime BP averages by apparatus 10 versus HBPM measurements taken during device initialization on the same day. Figure 7 a presents one data point for systolic BP and diastolic BP from each of the 2928 users of the device 10. On the X-axis is a single measurement from the HBPM reading during device initialization, compared to the mean BP during the day (8am to 8pm) measured by the device 10 on the Y-axis. Left, SBP. Right, DBP. The dashed line is calculated using Huber linear regression. Figure 7 The letter values ​​on b depict the distribution of the paired differences between the values ​​measured by the device 10 and the corresponding HBPM values ​​for systolic BP and diastolic BP. The mean and standard deviation of each distribution are presented at the bottom. Asterisks indicate statistically significant differences between HBPM and measurements obtained by the device 10 (p<0.001 for both).

[0135] When compared to conventional monitoring, using device 10, using real-world data from over 2000 patients, the distinct advantages of a significantly richer BP data set, the ability to measure nighttime BP longitudinally, and the automatic, passive nature of the device were significant (Table 1). Further, the systematic differences in the daytime BP averages for device 10 compared to daytime HBPM readings across this population were small and within an acceptable margin of error. In summary, device 10 has great potential to greatly improve the ability to monitor BP in an ambulatory setting.

[0136]

[0137] Table 1 Guidance for interpretation of daily averages obtained from device 10 when compared to daily averages obtained from HBPM.

[0138] Despite the low utilization of ABPM in clinical practice, it is still the recommended modality when a more complete analysis of the patient's BP pattern is required for the diagnosis / monitoring of HTN (see previous section). Although ABPM is currently the only recommended method that can obtain both daytime and nighttime BP readings, its infrequent use raises issues of reproducibility. The diurnal deviation of BP is known to be dynamic, and by arbitrarily choosing a short monitoring period (24 or 48 hours), clinicians may obtain data that represent only a small fraction of the overall BP.

[0139] To further illustrate the reproducibility issue of ABPM, Figure 8 a and Figure 8 b shows two examples of repeated ABPM records from ongoing clinical trials (below). A meta-analysis of 35 observational studies showed that for 1 / 3 of patients, the classification of dipper / non-dipper status for two consecutive ABPM nights was not reproducible (e.g., on the first night, the patient was classified as a dipper, but on the following night was classified as a non-dipper), and the observed differences in the nightly averages of SBP and DBP between two consecutive nights could vary between -19.6 and 21.3 mmHg and between -11.3 and 12.3 mmHg, respectively. These data confirm that the reproducibility of ABPM in assessing inter-individual decline states and daytime and nighttime BP values ​​is limited. Figure 8 a and Figure 8 b ABPM records of selected patients from the ABPM study are provided. Figure 8 a showed very poor inter-subject reproducibility, while Figure 8 b presents a better reproducibility in terms of the measured nocturnal drop in BP and the daytime / nocturnal averages of BP.

[0140] Given the poor reproducibility of ABPM as documented, device 10 can overcome the arbitrariness of cuff-based ABPM by leveraging the ability of device 10 to generate large amounts of data over long periods of time. However, data comparison between the two modalities requires consideration of two factors. First, device 10 measures BP fundamentally differently than conventional oscillometric ABPM measures BP. Second, the frequency and periodicity of ABPM measurements (e.g., once every 20 minutes) differ from the frequency and periodicity of device 10 (e.g., when the user is stationary for a sufficient period of time). This difference captures BP during different daily activities, with ABPM capturing far more readings during periods of physical activity than device 10. Therefore, these two differences (technique and trigger timing) are expected to generate daytime and nighttime averages of BP that may vary between modalities.

[0141] Fig. 9 a to Fig. 9 d provides an initial understanding of the systematic differences observed between ABPM and Device 10 in estimating a patient's nighttime state.

[0142] exist Fig. 9 a and Fig. 9 In b, the difference in estimated SBP reduction among patients in the NCT04548986 trial is illustrated. Fig. 9 c shows that the system factor across the initial cohort of patients in the same trial was 3.4. Fig. 9 d shows that the system factor is 3.1 across N=4644 users of device 10 when the phenotypic distribution is matched to that of a large independent trial (N=6359). As expected, different monitoring modalities tend to provide similar phenotypic information, but require the application of technology-dependent conversion factors.

[0143] Fig. 9 a and Fig. 9 b Graph showing the ABPM monitor (Diasys 3Plus, Novacor, France, Fig. 9 a) and device 10 (Aktiia BP monitor, Aktiia, Switzerland, Fig. 9 b) An example of simultaneous BP data collected from a patient enrolled in the NCT04548986 trial. Note that although the ABPM data were recorded over a 24-hour period, the data were recorded by the device 10 during and within a week after the ABPM recording. The monitoring period using the device 10 was extended to one week to account for the day-to-day variability of the diurnal pattern and to increase the number of data points registered during the day and night (because the sampling frequency of the device 10 is lower). On the same figure, two estimates of the BP drop are extracted. To calculate the drop on the ABPM recording, a general method of the difference between daytime and nighttime BP is used. The daytime and nighttime sub-periods are defined based on fixed clock time intervals: 9am to 9pm during the day and 12am to 6am at night. Data recorded during the transition period are excluded to avoid excessive dispersion between individual users. Data points located at a distance above the interquartile range from the median of each sub-period are considered to be outliers, and finally, the drop is calculated as the difference between the sub-period medians. In order to calculate the drop registered by the device 10, a statistical method was implemented. Taking advantage of the fact that the cuffless circadian plot presents a higher density of data points, a parametric model was used to fit the circadian rhythm for SBP and DBP (see the continuous "fit" line in the figure): The night drop was then extracted from one of the model parameters. The estimated night drop for this patient already differs between the two modalities, with the ABPM drop (A) appearing to be greater than the cuffless drop (B).

[0144] Fig. 9 c shows a statistical analysis of the A>B phenomenon for the preliminary patient cohort of the NCT04548986 trial. Data from the initial N=20 enrolled patients were processed to estimate the systematic gain difference in the descent measured by ABPM compared to the descent measured by device 10. In this cohort, and after bootstrapping the available samples, we calculated that the ABPM descent was characteristically 3.4 times greater than the cuffless descent, with a 95% confidence interval ranging from 2.3 to 4.4. It is important to note that NCT04548986 has not yet been completed, and a comprehensive analysis of the collected data will be further presented in a dedicated publication.

[0145] Fig. 9 d shows further statistical investigation of the same A>B phenomenon, now merging data from an independent ABPM study of N=6359 patients and real-world data from N=4644 users of device 10. According to Kario et al., in the general population, one would expect to observe the following distribution of night-fall phenotypes measured by ABPM measurements (phenotype being defined as an observable characteristic in the diurnal pattern of BP): 16% of individuals are extreme dippers (fall greater than 20 mmHg), 40% of individuals are normal dippers (fall between 10 and 20 mmHg), 32% of individuals are non-dippers (fall between 0 and 10 mmHg), and 12% of individuals are risers (positive fall). However, when observing the fall pattern recorded by device 10, the same distribution is not met, with the fall distribution being significantly compressed. The current analysis involves estimating the optimal factor required to expand the fall distribution of device 10 to an ABPM-equivalent phenotypic representation. In this cohort of N = 4644 users, and after bootstrapping, we calculated that the ABPM decline was characteristically 3.1 times greater than the continuous cuffless decline, with a 95% confidence interval ranging from 2.8 to 3.4.

[0146] By merging real-world data from over 4000 users of the device 10 with clinical data from a controlled clinical trial, it was demonstrated that the device 10 can estimate a patient's BP phenotype. However, due to technology differences between the cuffless technology of the device 10 and oscillometric ABPM monitors, technology-dependent conversion factors may be required to compare estimates from the two modalities. Table 2 provides general guidance on how BP phenotype data from the device 10 should be interpreted when compared to BP characteristics obtained from ABPM.

[0147] Thus, the apparatus of the present invention may further comprise the step of transforming the calculated cardiovascular values, the calculated cardiovascular parameters, the calculated circadian graphs or the calculated physiological parameters into ABPM equivalent values ​​and ABPM equivalent graphs according to a mapping function. The mapping function may be a pre-calculated affine function as presented above (the ABPM-style decline may be calculated as 3.1 times the decline calculated by the apparatus 10), or any other type of mapping function pre-calculated from recorded data from a large population, or calculated from any user data as described in claim 15, or a combination of both. The same mapping method may also be applied to transform the calculated cardiovascular values, the calculated cardiovascular parameters, the calculated circadian graphs or the calculated physiological parameters into HBPM equivalent values ​​and HBPM equivalent graphs according to a mapping function. Then, when calculating the cardiovascular risk score of the user, the ABPM equivalents or HBPM equivalents may be used instead of the relevant physiological parameters, or in addition to the relevant physiological parameters.

[0148]

[0149] Table 2 Guidance for interpreting BP phenotype data obtained from device 10 when compared to BP characteristics obtained from ABPM.

[0150] To demonstrate the potential of the device 10, Fig.10 a to Fig.10 J illustrates a new set of ambulatory BP control metrics estimated by means of the device 10 for a male patient (51 years old) during five months of continuous monitoring.

[0151] Specifically, Fig.10 10a to 10j show a new generation of dynamic BP control metrics that can be generated from data captured from the device 10. The time series presented were captured over a five month period for the subject and in addition to standard BP metrics (e.g., 24h, daytime, and nighttime averages of SBP, DBP, and HR) ( Fig.10 a. Fig.10 c. Fig.10 e. Fig.10 f and Fig.10 g), it also displays novel dynamic metrics such as time in therapeutic range (TTR, Fig.10 b) SBP variability ( Fig.10 d) Dynamic day and night model ( Fig.10 h) Dynamic SBP Night Drop ( Fig.10 i) and dynamic night fall duration and morning peak acceleration ( Fig.10 j).

[0152] Fig.10a Reports all 4729 SBP readings performed during the monitoring period, as well as the evolution of the 24hSBP mean. SBP readings within the target range (<120 mmHg) are shown as green dots, and those outside the target range (>120 mmHg) are shown as red dots. Based on these data, Fig.10 b shows a novel SBP-related metric for BP control: SBP time in target range (quartiles on Y-axis: 75%-100% of time = green, 50%-75% of time = yellow, 25%-50% of time = orange, and 0-25% of time = red), with the total percentage of time spent in each quartile of the monitoring period on the right. Fig.10 c further shows the nighttime average of SBP, where SBP>120 mmHg is colored in red and SBP<120 mmHg is colored in green. Fig.10 d shows the mid-term BP variability (SD in mmHg) during the day (orange), night (black) and mean (green). Fig.10 e and Fig.10 f shows the mean diastolic BP and 24-h mean heart rate (green lines) as well as all individual data points (green dots). Fig.10 g shows the daytime (orange), nighttime (black), and 24-hour mean (green) SBP. In the same figure, the shaded time periods a, b, and c correspond to Fig.10 h, and is also highlighted in panels I and J. Therefore, Fig.10 h shows the diurnal pattern of SBP at time periods a, b, and c, demonstrating the different patterns of night drop, night drop duration, and morning peak in this patient. Regarding sleep-related BP parameters, Fig.10 i shows the tracking of the evolution of the night drop in SBP in mmHg (dark green) and in % (light green). and Fig.10 j shows the tracking of the evolution of the duration of the nocturnal drop in SBP (red) and the morning peak (orange) is quantified.

[0153] The emergence of 24h-ABPM shows that the phenotype of BP is more complex than just a binary variable (hypertension or not), and it can demonstrate the diurnal variation of BP, including daytime and nighttime BP components and the nocturnal physiological decline of BP. The predictive values ​​of these different components are compared, and each variable shows the predictive value without considering the absolute BP. Average nighttime BP is a stronger predictor of cardiovascular events, whether systolic or diastolic. In both the hypertension group and the general population, the superior predictive value of the average nighttime BP, daytime average BP and HBPM of 24-hour ABPM is shown. This strong predictive value is noteworthy because the low reproducibility of the decline pattern has been shown in small studies and recent large studies. In fact, only a small number of hypertensive patients maintain their initial decline phenotype within 4 years. In addition, the tolerance of ABPM at night is lower than that during the day, and therefore may affect the quality of sleep. Sleep disorders caused by cuff inflation also show that they can affect the association between nighttime ABPM and prognosis.

[0154] Device 10 has the potential to overcome these negative features of nocturnal ABPM. First, nocturnal blood measurements can be repeated over days to months, allowing for the derivation of a more consistent nocturnal phenotype. Second, in the absence of any cuff inflation, the impact on sleep quality is expected to be negligible. However, these unique characteristics may affect the reference level of normal nocturnal cuffless BP, which may need to be redefined.

[0155] To facilitate the exploration of novel phenotype-driven cardiovascular risk assessment, Table 3 provides a list of existing and promising BP phenotypes that could be supported by large-scale deployment of devices 10 .

[0156]

[0157] Table 3 is a suggested list of BP phenotypes that can already be identified via HBPM / ABPM examinations, and an expanded list of BP phenotypes that will be supported by deploying the device 10 .

[0158] Before we can determine the use of the classic and novel phenotypes in daily practice, their utility and predictive value will have to be demonstrated in longitudinal epidemiological studies. Their independent predictive value for mean 24-h or nighttime BP will be necessary if evidence-based personalized therapy is to become a reality. Studying these phenotypes will take time but will certainly provide physicians with a new set of physiological or induced BP responses that may help tailor antihypertensive therapy in the future.

[0159] The present disclosure further relates to a non-transitory computer-readable storage medium comprising a computer program product including instructions for causing at least one processor to perform a method for determining a cardiovascular risk score for a user.

Claims

1. A method comprising: Providing an apparatus (10) configured to measure at least a cardiovascular signal of a user; measuring said at least cardiovascular signal during a monitoring period (Tm) having a duration of at least 24 hours; The monitoring period (Tm) is subdivided into at least one monitoring period (Ts), wherein the at least one monitoring period (Ts) has a duration of 24 hours and comprises a plurality of measurement cycles; determining a cardiovascular value for each measurement cycle of the plurality of measurement cycles within the at least one monitoring period (Ts); aggregating said cardiovascular values ​​determined for a given measurement cycle of each said at least one monitoring period (Ts) into a cardiovascular parameter group; constructing, for each of the at least one monitoring period (Ts), a 24-hour circadian graph of the cardiovascular parameter group relative to the corresponding measurement period; calculating a plurality of physiological parameters of the user according to the diurnal graph; as well as A cardiovascular risk score of the user is calculated based on the determined physiological parameters.

2. The method according to claim 1, wherein the monitoring period (Tm) has a duration of 48 hours, seven days, one month or one year.

3. A method according to claim 1 or claim 2, wherein the measurement period has a duration of at least 10 seconds or 30 seconds, i.e. a duration of 10 seconds, 30 seconds, 1 minute, 5 minutes, one hour, two hours, four hours or six hours. 4 . The method according to claim 1 , comprising calculating, for each cardiovascular parameter group, a cardiovascular representative value of the cardiovascular values ​​aggregated into the cardiovascular parameter group. 5 . The method according to claim 4 , wherein calculating the cardiovascular representative value comprises classifying cardiovascular parameters as inner values ​​or outer values, and calculating the cardiovascular representative value using only inner cardiovascular parameters.

6. The method of claim 5, wherein calculating the cardiovascular representative value comprises classifying the cardiovascular parameters as insider values ​​or outsider values; and wherein the cardiovascular parameters are weighted according to the probability that they are insider values ​​or outsider values.

7. The method according to any one of claims 1 to 6, wherein: Determining the plurality of physiological parameters of the user comprises the step of fitting a model to a circadian pattern, the physiological parameters corresponding to parameters of the model.

8. The method of claim 7, wherein the model is one of a linear model, a nonlinear model, a constrained model, and an unconstrained model fitted via least squares optimization.

9. The method according to any one of claims 1 to 8, wherein the physiological parameter comprises any one of the following parameters: daytime BP value (Y0), absolute nocturnal drop amplitude (ampl), timing of the onset of the drop (X0), duration of the pre-nocturnal slope (dl), duration of the plateau of the drop (nl) or duration of the post-nocturnal slope (al).

10. The method according to any one of claims 7 to 9, wherein: The model further uses non-physiological parameters.

11. The method according to any one of claims 1 to 10, further comprising combining at least two physiological parameters to obtain one or more correlated physiological parameters.

12. The method of claim 11, wherein the one or more relevant physiological parameters include any of the following: daytime BP value, nighttime BP value, relative nighttime drop amplitude, complete nighttime drop duration, time in target range (TTR), BP variability pattern, morning surge slope, nighttime SBP, nighttime DBP, nighttime HR, SBP drop, DBP drop, HR drop, SBP morning surge, DBP morning surge, HR morning surge, SBP drop duration, DBP drop duration, HR drop duration, SBP / DBP / HR synchronization, response to medication type, BP medication adherence, patient engagement measure, or response to lifestyle intervention.

13. A method according to claim 11 or claim 12, wherein the relevant physiological parameters include a blood pressure phenotype of the user.

14. The method of claim 13, wherein the relevant physiological parameters include any of the following: true normotension, white coat hypertension, masked hypertension, sustained hypertension, hypotension, night drop, night rise, or a phenotype predicting response to a particular drug or treatment.

15. The method of any one of claims 1 to 14, wherein calculating the cardiovascular risk score of the user comprises using user data.

16. A method according to claim 15, wherein the user data includes any of the following: age, weight, height, gender, race, blood lipid level, diabetes status, smoking, CT calcium (Agatston score), family history, risk genetic markers, actigraphy information, exercise information, diet information, stress level, overall feeling, hormone data, menstrual cycle information, medication intake, weekday / weekend information, season information, sleep quality information, bedtime pattern, or any parameter used to calculate cardiovascular risk score in clinical guidelines such as ACC / AHA guidelines, ESC guidelines or MESA database, or any non-physiological parameter.

17. The method according to any one of claims 1 to claim 16, wherein the cardiovascular risk score is any one of a 10-year cardiovascular disease risk, a 10-year heart attack risk, a 10-year stroke risk, or any other clinically relevant cardiovascular risk score.

18. The method according to any one of claims 1 to claim 17, wherein the cardiovascular parameter is at least any one of a blood pressure value, a heart rate value, a cardiac output value, a blood glucose value, a physical activity measurement, a sleep amount and sleep quality measurement, an electrocardiogram signal, a photoplethysmography signal, a bioimpedance signal or an ultrasound signal.

19. An apparatus for determining a cardiovascular risk score of a user, comprising: a measuring module (20) configured to measure a cardiovascular signal of the user during a monitoring period (Tm) having a duration of at least twenty-four hours, wherein the monitoring period (Tm) is subdivided into monitoring periods (Ts), each monitoring period (Ts) having a duration of twenty-four hours and comprising a plurality of measuring periods (Tm); a processor (30) configured to determine a cardiovascular value for each measurement cycle (Tm), aggregate the cardiovascular values ​​determined for the corresponding measurement cycle (Tm) of each monitoring period (Ts) into a cardiovascular parameter group, construct a twenty-four-hour circadian graph of the parameter group relative to the corresponding measurement cycle (Tm), determine a plurality of physiological parameters of the user using the circadian graph, and calculate a cardiovascular risk score of the user using the determined physiological parameters; as well as An interface (40) is used to display and / or transmit the calculated cardiovascular risk score.

20. The apparatus of claim 19, wherein the interface (40) comprises an application on a smartphone, tablet, computer, smart watch, or any portable device.

21. A device as claimed in claim 19 or claim 20, connectable via any wired or wireless connection including WiFi or Bluetooth or cellular support.

22. A non-transitory computer-readable storage medium comprising a computer program product, the computer program product comprising instructions causing at least one processor to perform the method according to any one of claims 1 to 18.

Images