Method and system for measuring cardiovascular health
Patent Information
- Application Number
- PCT/CN2025/085369
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-03-27
- Publication Date
- 2026-10-01
Smart Images

Figure CN2025085369_01102026_PF_FP_ABST
Abstract
Description
Method and System for Measuring Cardiovascular HealthFIELD OF THE INVENTION
[0001] The present invention is in the field of biomedical engineering. More specifically, the present invention relates toa non-invasivemethod and a system for generating pulse wave velocityin a subject using a smart device for cardiovascular health monitoring. The present invention employs pulse transit time (PTT) and vascular transit time (VTT) to compute pulse wave velocity and assess the subject’s cardiovascular health efficiently.BACKGROUND OF THE INVENTION
[0002] Continuous cardiovascular health monitoring has evolved significantly with advancements in photoplethysmography (PPG) and pulse transit time (PTT) technologies. For example, in the monitoring of cardiovascular health (BP) , traditional devices utilizing the volume clamp method provide continuous beat-to-beat BP readings, making them valuable tools for research applications. Other monitoring systems integrate finger PPG with electrocardiography (ECG) to estimate BP using pulse wave velocity, enhancing non-invasive tracking. More recent wearable solutions leverage PTT technology for on-the-go BP measurement, requiring periodic calibration with cuff-based devices.
[0003] Innovative wrist-worn and bracelet-style monitors focus on continuous, cuffless tracking via optical sensors, offering greater convenience. However, widespread adoption faces challenges, including hardware limitations, accuracy concerns, and signal quality issues. Many solutions require multiple sensors, leading to increased complexity, integration difficulties, and reduced user comfort. Motion artifacts, skin tone variations, and environmental factors further impact signal reliability. While wearable BP monitoring holds great potential, improvements in calibration methods, sensor integration, and artifact reduction remain crucial for enhancing accuracy and usability.
[0004] For example, United States Patent No. 9485345 B2 disclosed methods that employ a smart device to measure vitals of a person and that the phone is connected to 911 services with capacity to implement 911 dispatch protocols. This invention lacks key advancements that will increase accuracy and precision. More specifically, the algorithms in the said methods primarily focus on timestamp synchronization for calculating time differences but not applicable to reduce noise. This invention also requires internet connectivity for processing and synchronization, thus, limiting usability in remote or offline environments. The reliance on Vascular Transit Time (VTT) also limits accuracy in blood pressure estimation in this invention. Furthermore, this approach requires two mobile phones for differential blood pressure estimation, adding complexity to the measurement process and this invention also relies on the phone’s flash during PPG measurement, which is not essential and may cause additional power consumption or interference.
[0005] US Patent Application No. 20230056557 A1 may have disclosed devices and methods for estimating blood pressure, which implement a pulse-transit-time-based blood pressure model that can be calibrated. While implementing a pulse-transit-time-based blood pressure model with calibration capabilities, however this invention relies on a wearable device such as a smartwatch connected to a mobile phone for signal detection. The necessity of a wearable device in this invention makes its implementation more complex in terms of accessibility for real-time blood pressure measurement without requiring external sensors. Additionally, velocity transit time (VTT) andthis invention only relies onpulse transit time (PTT) and it can potentially lead to fewer accurate inferences and it relies mainly on peaks for blood pressure (BP) estimation, which may not capture the full range of relevant physiological signal. This invention also uses a quadrant-based approach for pixel selection, which may not adapt well across different phone models. There is also a likelihood that this invention may not prioritize synchronization, leading to potential errors due to time misalignment.
[0006] US Patent Application No. 20230056557 A1 disclosed a system and method for pulse transit time (PTT) measurements primarily based on optical data obtained from multiple cameras on a mobile phone, specifically from the user’s face and fingers. However, this invention does not incorporate cardiac motion such as cardiac-induced vibrations and subtle pre-ejection phase movements, which leads to a reliance solely on optical and acoustic signals. This can reduce the precision of time delay calculations between cardiac events and pulse wave propagation, making the system more susceptible to motion artifacts and external noise. Consequently, blood pressure estimation may be less reliable, as it lacks the additional validation and redundancy that cardiac motion signals provide. Additionally, this invention focuses on facial PPG, which is often rejected by hospitals due to privacy concerns. Additional processing step on video feeds is also required in this invention, which can introduce delays. Furthermore, this invention may have mentioned the use of a microphone, however, the noise collected by the microphone may induce noisy signals, causing lower data accuracy.
[0007] These limitations highlight the need for a solution that integrates a comprehensive signal processing approach, leveraging mobile phone sensors without additional hardware while ensuring high-quality, noise-filtered data for accurate and real-time cardiovascular health monitoring including blood pressureand arterial stiffness measurements.SUMMARY OF THE INVENTION
[0008] It is an objective of the present invention toprovidea non-invasive cardiovascular health monitoring systemthat utilizes a smart devicethat employs sensors for accurate health monitoring. More specifically, the present invention provides methods and system with capacities to monitor cardiovascular health through monitoring of blood pressure (BP) and / or arterial stiffness measurements without additional hardware.
[0009] It is also an objective of the present invention to provide a non-invasive cardiovascularhealth monitoring system that enhances and arterial stiffness measurements with accuracy by integrating automatic calculations based on sensors data collected, allowing continuous cardiovascular health monitoring with ease and convenience.
[0010] Accordingly, these objectives can be achieved by following the teachings of the present invention, which relates to a method for continuously monitoring cardiovascularhealth of a subject using a smart device, the method comprising: activating a sensor initialisation protocol on the smart device; collecting raw signals using an integrated sensory module embedded within the smart device simultaneously; pre-processing the raw signals collected via a pre-processing module to obtain pre-processed signals; analysing the pre-processed signals via an analytical module to generate cardiovascular health measurements based on pulse transit time (PTT) and ventricular transit time (VTT) data; displaying the cardiovascular health measurements with diastolic and systolic values via a user interface of the smart device, along with confidence metrics and signal quality indicators; and, storing the cardiovascular health measurements in a database with data validation checks for continuous monitoring.
[0011] Additionally, these objectives also can be achieved by following the teachings of the present invention, which relates to a system for continuously monitoring vascular health using a smart device, the system comprising: a processing unit having instructions stored, therein, the instructions, when executed by one or more processors, cause the one or more processors to: activate a sensor initialization protocol; collect raw signals using an integrated sensory module embedded within the smart device simultaneously; pre-process the raw signals collected via a pre-processing module to obtain pre-processed signals; analyse the pre-processed signals via an analytical module to generate vascular health measurements based on pulse transit time (PTT) and ventricular transit time (VTT) signals; display the cardiovascular health measurements with diastolic and systolic values via a user interface of the smart device, along with confidence metrics and signal quality indicators; and, store the cardiovascular health measurements in a database with data validation checks for continuous monitoring.BRIEF DESCRIPTION OF THE DRAWINGS
[0012] The features of the invention will be more readily understood and appreciated from the following detailed description when read in conjunction with the accompanying drawings of the preferred embodiment of the present invention, in which:
[0013] FIG. 1 illustrates an overview of the present invention in a form of a flow chart;
[0014] FIG. 2 illustrates extraction of photoplethysmography (PPG) features;
[0015] FIG. 3illustratesextraction of seismocardiography (SCG) features;
[0016] FIG. 4 illustrates extraction of phonocardiography (PCG) features; and,
[0017] FIG. 5 illustrates data analysis via an analytical module to generate cardiovascular health measurements based on PTT and VTT data. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
[0018] For the purposes of promoting and understanding the principles of the invention, reference will now be made to the embodiments illustrated in the drawings and described in the following written specification. It is understood that the present invention includes any alterations and modifications to the illustrated embodiments and includes further applications of the principles of the invention as would normally occur to one skilled in the art to which the invention pertains.
[0019] The present invention teaches a method for continuously monitoring cardiovascular health using a smart device, the method comprising: activating a sensor initialisation protocol a cardiovascular health monitoring system 100; collecting raw signals using an integrated sensory module embedded within the smart device120 simultaneously; pre-processing the raw signals collected via a pre-processing module to obtain pre-processed signals 140; analysing the pre-processed signals via an analytical module 500to generate cardiovascular health measurements based on pulse transit time (PTT) and ventricular transit time (VTT) data 160; displaying the cardiovascular health measurements with diastolic and systolic values via a user interface of the smart device, along with confidence metrics and signal quality indicators 180; and, storing the cardiovascular health measurements in a database with data validation checks for continuous monitoring.
[0020] All embodiments in the present invention, wherein the smart device employed must have capacities to record sounds and videos.
[0021] By performing the above, the present invention enables local processing without requiring an internet connection, making it usable in any location. It eliminates the need for WiFi or VOIP, ensuring independence from network-based communication. The design allows for simultaneous three-modality recording by optimizing phone positioning, reducing the need for multiple devices. Unlike existing solutions that rely on master-slave synchronization between phones, this approach removes the need for signal syncing over transmission. By minimizing error delays, it enhances the accuracy of pulse transit time (PTT) calculations. While other existing methods require multiple phones to capture different modalities, the present invention also enables all measurements to be conducted on a single device. Additionally, the PPG system optimizes flash usage by keeping it on only during the process, therefore saving the energy to power up the system.
[0022] In a non-limiting embodiment of the present invention, the cardiovascular health includes measurements of blood pressure or arterial stiffness or a combination thereof.
[0023] In a non-limiting embodiment of the present invention, the step of collecting the raw signals using the integrated sensory module embedded within the smart device120 further comprising: obtaining raw photoplethysmography (PPG) signals by securing a finger on a camera wherein a PPG module200 determines if the PPG signal is accepted to move forward, or rejected for a corrective measure or repetition signals measurement; obtaining raw phonocardiography (PCG) and seismocardiography (SCG) signals by securing the smart device near the subject’s heart, wherein an SCG module300 determines if the SCG signal is accepted to move forward, or rejected for a corrective measure or repetition of signals measurementand, utilizing the SCG signals to validate the PPG signals by assessing motion artifacts, wherein excessive SCG-detected movement indicates the instability of the smart device for the corrective measure or repetition of signals measurement.
[0024] In a non-limiting embodiment of the present invention, the step of pre-processing the raw signals collected via the pre-processing module to obtain the pre-processed signals 140 further comprising: synchronizing the PPG and SCG signals by aligning and stamping each raw signals with a common time base timestamps for temporal alignments via a signal synchronisation module; evaluating the synchronized signal quality based on signal quality metrics with pre-determined thresholds of the synchronized signals by filtering poor-quality signals from the high-quality signals, wherein, when the signal quality metrics fall below the pre-determined thresholds, the signals indicate the poor-quality signals with the motion artifacts, signal dropouts or other anomalies, thus said signals automatically undergo a corrective measure or repetition of signals measurement; or, when the signal quality metrics exceed the pre-determined thresholds, the signals indicate the high-quality signals and proceed for processing by the analytical module500; and, extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module.
[0025] In a non-limiting embodiment of the present invention, the step of extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module, further comprising: extracting at least one PPG feature via the PPG module 200, wherein, the extraction is performed by: extracting pixels and RGB data from the PPG signals; averaging pixels and RGB data based on weights to output PPG values; performing exponential smoothing and filtering of band pass of the PPG values; normalizing and visualizing the smoothed and filtered PPG values; identifying peaks in the PPG values to extract heart rate (HR) and heart rate viability (HRV) values; verifying signal quality based on the HRV value using the quality control module for continuous acceptable PPG data streaming; and, extracting the at least one PPG feature for analysis by the analytical module500.
[0026] In a non-limiting embodiment of the present invention, the step of extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module, further comprising: extracting at least one SCG feature via the SCG module300, wherein, the extraction is performed by: requesting for highest sampling rate to capture fine cardiac motion; recording the fine cardiac motion for continuous data streaming by an accelerometer; performing filtration of band pass to remove noise; extracting SCG data via data enveloping; normalizing and visualizing the filtered SCG data; identifying peaks in the SCG data to SCG features; verifying signal quality of the SCG data using the quality control module for continuous acceptable SCG data streaming; and, extracting the at least one SCG feature for analysis by the analytical module500.
[0027] In a non-limiting embodiment of the present invention, the step of extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module, further comprising: extracting at least one PCG feature via the PCG module400, wherein, the extraction is performed by: capturing the heart sound via a video recording; obtaining PPG data based on the video recording by communicating with the PPG model; extracting video sound from the video recording; amplifying, normalizing and removing noise from the sound of the video recording; computing a sound envelope to identify relevant signal regions for windowing; identifying and validating a first heart sound (S1) and a second heart sound (S2) components based on time and amplitude criteria; verifying signal quality of the PCG data; synchronizing the PCG data with its corresponding PPG signals; and, extracting the at least one PCG feature for analysis by the analytical module500.
[0028] In a non-limiting embodiment of the present invention, the corrective measure or repetition of signals measurement includes readjusting finger position or realigning gyroscope-guided smart device position.
[0029] In a non-limiting embodiment of the present invention, the step of analysing the pre-processed signals via the analytical module 500to generate the cardiovascular health measurements 160 further comprising: applying a time-delay calculation module to align and compare a plurality of peak points from the extracted key features for optimal time delays measurement; computing a PTT calculation using a weighted averaging algorithm embedded in the PTT calculation module based on the synchronized SCG and PPG data to output PTT values; computing a VTT calculation using the weighted averaging algorithm embedded in the VTT calculation module based on PPG and PCG data to output VTT values; and, applying the analytical module to embed and analyse PTT values, VTT values, subject’s input data and subject’s clinical input data to output the cardiovascular health measurements with diastolic and systolic values.
[0030] The above approach utilizes a pixel selection mesh, allowing adaptation to different phone models. By combining VTT and PTT, the present invention optimizes results and enhances inference through the computation of their differences. Unlike conventional models that rely on peak detection for BP estimation, this solution incorporates multiple modalities for improved accuracy. Time synchronization is prioritized, performed before all processes, and continuously refined by recording application process delay times.
[0031] In a non-limiting embodiment of the present invention, the method further comprising: utilizing additional physiological signals including but not limited to pulse wave velocity variations, arterial stiffness factors, temperature compensation and motion artifact correction to compute the cardiovascular health measurement.
[0032] In a non-limiting embodiment of the present invention, the method further comprising inputting real cuff-based cardiovascular health data and model for subsequent calibration.
[0033] In a non-limiting embodiment of the present invention, the method further comprising: pre-training the analytical module 500 with the subject’s demographic data including age, gender, height, and weight as mandatory inputs, and optionally incorporating additional parameters derived from a selfie image of left torso, including body arm, for detecting arm length and skin colour.
[0034] In a non-limiting embodiment of the present invention, the step of pre-training the analytical module 500 further comprising: calibrating signals with the subject’s existing measurements including but not limited to the subject’s in-clinic or at-homecardiovascular health measurements.
[0035] The present invention also teaches a system for continuously monitoring cardiovascular health using a smart device on a subject, the system comprising: a processing modulehaving instructions stored, therein, the instructions, when executed by one or more processors, cause the one or more processors to: activate a sensor initialization protocol of the smart device; collect raw signals using an integrated sensory module simultaneously; pre-process the raw signals collected via a pre-processing module to obtain pre-processed signals; analyse the pre-processed signals via an analytical module500 to generate cardiovascular health measurements based on pulse transit time (PTT) and ventricular transit time (VTT) signals; display the cardiovascular health measurements with diastolic and systolic values via a user interface of the smart device or a display unit, along with confidence metrics and signal quality indicators; and, store the cardiovascular health measurements in a database with data validation checks for continuous monitoring.
[0036] In a non-limiting embodiment of the present invention, the processing module further comprising of a central processing unit (CPU) , a math processing unit, a graphic processing unit (GPU) and a tensor processing unit (TPU) .
[0037] In a non-limiting embodiment of the present invention, the display unit includes but is not limited to a liquid crystal display, a light emitting display, or any other suitable display.
[0038] In a non-limiting embodiment of the present invention, the smart device includes but is not limited to a mobile phone, VR glasses and the like.
[0039] In a non-limiting embodiment of the present invention, the subject includes but is not limited to a human or an animal.
[0040] In a non-limiting embodiment of the present invention, the raw signals are physiological signals obtained from finger pulse, heart sound and cardiac motion.
[0041] The present invention relies on a finger PPG and heart sound, resulting in a fundamentally different data collection process that includes heart sound and motion modalitieswith only one device. Additionally, no extra processing is needed on the video feed, as data quality is ensured by the present invention which in turn reduces delays in video-extracted data.
[0042] In a non-limiting embodiment of the present invention, the sensor initialization protocol is configured to establish initial sensor parameters and baseline conditions for signals acquisition.
[0043] In a non-limiting embodiment of the present invention, the integrated sensory module further comprises: a phonocardiography (PCG) module 400 configured to obtain raw PCG signals and extract PCG data from the heart sound via a video recording with a microphone enabled; a seismocardiography (SCG) module300, configured to obtain raw SCG signals and extract SCG data from the cardiac motion via an accelerometer; and, a photoplethysmography (PPG) module200, configured to obtain raw PPG signals and extract PPG data from the finger pulse via a camera.
[0044] In a non-limiting embodiment of the present invention, raw PCG signals are obtained from at least 30 fps video recordings.
[0045] In a non-limiting embodiment of the present invention, raw SCG signals are obtained from 100 Hz accelerometer data streams.
[0046] In a non-limiting embodiment of the present invention, raw PPG signals are obtained from 30 to 120 fps camera feeds.
[0047] In a non-limiting embodiment of the present invention, the pre-processing module comprises: a signal synchronization module configured to obtain temporal alignments for PPG and SCG signals collected by the integrated sensory module, generating synchronized signals; and, a quality control module that employs a reference point algorithm to evaluate signal quality based on signal quality metrics with pre-determined thresholds of the synchronized signals.
[0048] In a non-limiting embodiment of the present invention, the signal quality metrics are signal-to-noise ratios, baseline stability and artifacts presence.
[0049] In a non-limiting embodiment of the present invention, the quality control module is further configured to filter poor-quality signals if the signals fall below pre-determined thresholds and to extract features with peaks in the signal quality metrics, generating pre-processed signals for analysis by the analytical module500.
[0050] In a non-limiting embodiment of the present invention, the features are cardiovascular parameters, the cardiovascular parameters further comprising: time-series features including but not limited to PPG peaks, PPG 50%peaks, Heart SCG envelopes, first and second (S1, S2) heartbeat sounds from Heart SCG; and, non time-series features including but not limited to PPG notch position, third and fourth (S3, S4) heartbeat sounds, R-R interval, pulse duration, area of diastolic and systolic and contraction amplitude.
[0051] In a non-limiting embodiment of the present invention, the analytical module 500 comprises: a time-delay calculation module configured to align and compare detected reference points to measure time delays accurately in the PPG, SCG and the PCG features; PTT and VTT calculation modules configured to generate PTT and VTT values and reject the PTT and VTT values below their thresholds based on the signal quality metrics; and, a predetermined parametric and non-parametric algorithm or functions configured to compute cardiovascular health measurements based on an adaptive correlation between the PTT, VTT and the signals.
[0052] The following details describe the present invention with details and working examples. Advantages of the present invention are also described below.
[0053] More specifically, the present invention combines three distinct physiological measurements in a single smartphone, namely: heart sound detection via microphone; cardiac motion vibration via accelerometer and PPG measurement via camera. In view of this, no additional hardware is required, unlike existing solutions.
[0054] The present invention is a multi-signal processing invention. More specifically, the present invention can simultaneously capture and synchronize multiple cardiovascular signals with advanced signal processing algorithms for noise reduction, real-time data integration and analysis, PTT calculation method using multiple reference points.
[0055] The present invention initiates measurement through a coordinated sequence of sensor activations and signal processing steps. Upon measurement initiation, the present invention performs a comprehensive sensor initialization protocol that prepares three primary measurement channels: acoustic, motion, and optical. The initialization phase ensures proper configuration of sensor parameters and establishes baseline conditions for accurate data acquisition.
[0056] Following initialization, the present invention simultaneously activates three distinct measurement channels. The microphone subsystem is configured for high-fidelity acoustic capture of heart sounds with specific emphasis on S1 and S2 components, utilizing a sampling rate of 2 kHz with 16-bit resolution. The accelerometer subsystem is activated to capture subtle cardiac-induced vibrations, operating at a sampling rate of 100 Hz with tri-axial measurement capability. The camera and LED subsystem engages with precise timing control for PPG signal acquisition, maintaining a consistent frame rate of 60 fps with controlled flash-LED illumination intensity.
[0057] The present invention implements a sophisticated signal synchronization mechanism that aligns the three independent data streams with microsecond precision. This synchronization utilizes a common time base and hardware-level timestamps to ensure temporal alignment of all physiological signals. The synchronization module compensates for varying sensor latencies and ensures that all signals maintain precise temporal relationships necessary for accurate PTT calculations.
[0058] Quality assurance is performed through real-time signal analysis that evaluates signal-to-noise ratios, baseline stability, and artifact presence. The quality check module implements adaptive thresholding to identify motion artifacts, signal dropouts, and other anomalies that could compromise measurement accuracy. If quality metrics fall below predetermined thresholds, the present invention automatically initiates a corrective measure or repetition for signals measurement.
[0059] The final stage encompasses comprehensive data processing, where synchronized signals undergo specific filtering, feature extraction, and PTT calculation algorithms. This processing pipeline includes adaptive noise reduction, peak detection, and sophisticated signal analysis techniques to extract relevant cardiovascular parameters. The present invention maintains continuous data buffering to ensure no information loss during processing transitions.
[0060] This integrated approach ensures robust and accurate physiological measurements while maintaining system efficiency and reliability through coordinated sensor operation and precise timing control.
[0061] The PTT Calculation system implements a signal processing algorithm that processes and analyses multiple physiological signals to determine cardiovascular health values. The present invention begins with three synchronized processed signals: heart sound data from the microphone, cardiac motion vibration data from the accelerometer, and PPG waveform data from the smartphone camera.
[0062] In the reference point detection stage, the present invention employs advanced peak detection algorithms to identify specific physiological landmarks within each signal stream. For heart sounds, the algorithm precisely detects S1 (first heart sound) peaks using wavelet transformation techniques. Simultaneously, the present invention identifies the characteristic points in the PPG waveform, specifically focusing on the foot point of the pulse wave, while also detecting peak acceleration points in the cardiac motion signal.
[0063] The time delay calculation module implements a cross-correlation analysis between these detected reference points. The present invention calculates multiple time intervals: the delay between S1 and the PPG foot point (PTT-S1) , the delay between the acceleration peak and PPG foot point (PTT-ACC) , and additional temporal relationships between various signal features. These multiple time delay measurements enable robust PTT computation by providing redundant measurements that improve accuracy.
[0064] The PTT Computation stage synthesizes these multiple time delay measurements using a weighted averaging algorithm. The weighting factors are dynamically adjusted based on signal quality metrics, ensuring optimal utilization of the available measurements. This stage also implements artifact rejection algorithms to eliminate unreliable PTT values caused by motion or poor signal quality.
[0065] The BP Estimation Algorithm employs a machine learning model that converts the computed PTT values into cardiovascular health measurements. The algorithm utilizes a calibrated mathematical relationship between PTT and cardiovascular health, incorporating additional parameters such as pulse wave velocity variations and arterial stiffness factors. The present invention applies temperature compensation and motion artifact correction to enhance measurement accuracy.
[0066] Finally, the result display module presents the calculated systolic and diastolic cardiovascular health values through a user-friendly interface, along with confidence metrics and signal quality indicators. The present invention stores these results in a structured database for trend analysis and historical comparison, while also implementing data validation checks to ensure physiological plausibility of the measurements.
[0067] The present invention explained above is not limited to the aforementioned embodiment and drawings, and it will be obvious to those having an ordinary skill in the art of the present invention that various replacements, deformations, and changes may be made without departing from the scope of the invention.
Claims
1.A method for continuously monitoring cardiovascular health of a subject using a smart device, the method comprising:activating a sensor initialisation protocolon the smart device (100) ;collecting raw signals using an integrated sensory module embedded within the smart device (120) simultaneously;pre-processing the raw signals collected via a pre-processing module to obtain pre-processed signals (140) ;analysing the pre-processed signals via an analytical module (500) to generate cardiovascular health measurements based on pulse transit time (PTT) and ventricular transit time (VTT) data (160) ;displaying the cardiovascular health measurements with diastolic and systolic values via a user interface of the smart device, along with confidence metrics and signal quality indicators (180) ; and,storing the cardiovascular health measurements in a database with data validation checks for continuous monitoring.2.The method according to claim 1, wherein the cardiovascular health includes measurements of blood pressure or arterial stiffness or a combination thereof.3.The method according to claim 1, wherein the step of collecting the raw signals using the integrated sensory module embedded within the smart device (120) further comprising:obtaining raw photoplethysmography (PPG) signals by securing a finger on a camera wherein a PPG module (200) determines if the PPG signal is accepted or rejected for a corrective measure or repetition of signals measurement;obtaining raw phonocardiography (PCG) and seismocardiography (SCG) signals by securing the smart device near the subject’s heart, wherein an SCG module (300) determines if the SCG signal is accepted or rejected for a corrective measure or repetition of signals measurement; and,utilizing the SCG signals to validate the PPG signals by assessing motion artifacts, wherein excessive SCG-detected movement indicates the instability of the smart device for the corrective measure or repetition of signals measurement.4.The method according to claim 3, wherein the step of pre-processing the raw signals collected via the pre-processing module to obtain the pre-processed signals (140) further comprising:synchronizing the PPG and SCG signals by aligning and stamping each raw signals with a common time base timestamps for temporal alignments via a signal synchronisation module;evaluating the synchronized signal quality based on signal quality metrics with pre-determined thresholds of the synchronized signals by filtering poor-quality signals from the high-quality signals, wherein,when the signal quality metrics fall below the pre-determined thresholds, the signals indicate the poor-quality signals with the motion artifacts, signal dropouts or other anomalies, thus said signals automatically undergo a corrective measure or repetition of signals measurement; or,when the signal quality metrics exceed the pre-determined thresholds, the signals indicate the high-quality signals and proceed for processing by the analytical module (500) ; and,extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module.5.The method according to claim 4, wherein the step of extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module, further comprising:extracting at least one PPG feature via the PPG module (200) , wherein, the extraction is performed by:extracting pixels and RGB data from the PPG signals;averaging pixels and RGB data based on weights to output PPG values;performing exponential smoothing and filtering of band pass of the PPG values;normalizing and visualizing the smoothed and filtered PPG values;identifying peaks in the PPG values to extract heart rate (HR) and heart rate viability (HRV) values;verifying signal quality based on the HRV value using the quality control module for continuous acceptable PPG data streaming; and,extracting the at least one PPG feature for analysis by the analytical module (500) .6.The method according to claim 4, wherein the step of extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module, further comprising:extracting at least one SCG feature via the SCG module (300) , wherein, the extraction is performed by:requesting for highest sampling rate to capture fine cardiac motion;recording the fine cardiac motion for continuous data streaming by an accelerometer;performing filtration of band pass to remove noise;extracting SCG data via data envelope;normalizing and visualizing the filtered SCG data;identifying peaks in the SCG data to SCG features;verifying signal quality of the SCG data using the quality control module for continuous acceptable SCG data streaming; and,extracting the at least one SCG feature for analysis by the analytical module (500) .7.The method according to claim 4, wherein the step of extracting at least one key feature in each signal by each module using a reference point detection algorithm employed in the quality control module, further comprising:extracting at least one PCG feature via the PCG module (400) , wherein, the extraction is performed by:capturing the heart sound via a video recording;obtaining PPG data based on the video recording by communicating with the PPG model;extracting video sound from the video recording;amplifying, normalizing and removing noise from the sound of the video recording;computing a sound envelope to identify relevant signal regions for windowing;identifying and validating a first heart sound (S1) and a second heart sound (S2) components based on time and amplitude criteria;verifying signal quality of the PCG data;synchronizing the PCG data with its corresponding PPG signals; and,extracting the at least one PCG feature for analysis by the analytical module (500) .8.The method according to claim 3, wherein the corrective measure or repetition of signals measurement includes readjusting finger position or realigning gyroscope-guided smart device position.9.The method according to claim 1, wherein the step of analysing the pre-processed signals via the analytical module (500) to generate the cardiovascular health measurements (160) further comprising:applying a time-delay calculation module to align and compare a plurality of peak points from the extracted key features for optimal time delays measurement;computing a PTT calculation using a weighted averaging algorithm embedded in the PTT calculation module based on the synchronized SCG and PPG data to output PTT values;computing a VTT calculation using the weighted averaging algorithm embedded in the VTT calculation module based on PPG and PCG data to output VTT values; and,applying the analytical module (500) to embed and analyse PTT values, VTT values, subject’s input data and subject’s clinical input data to output the cardiovascular health measurements with diastolic and systolic values.10.The method according to claim 1, wherein the method further comprising:utilizing additional physiological signals including but not limited to pulse wave velocity variations, arterial stiffness factors, temperature compensation and motion artifact correction to compute the cardiovascular health measurement.11.The method according to claim 1, wherein the method further comprising inputting real cuff-based cardiovascular health data and model for subsequent calibration.12.The method according to claim 1, wherein the method further comprising:pre-training the analytical module (500) with the subject’s demographic data including age, gender, height, and weight as mandatory inputs, and optionally incorporating additional parameters derived from a selfie image of left torso, including body arm, for detecting arm length and skin colour.13.The method according to claim 12, wherein the step of pre-training the analytical module (500) model further comprising:calibrating signals with subject’s existing measurements including but not limited to the subject’s in-clinic or at-home cardiovascular health measurements.14.A system for continuously monitoringcardiovascular health using a smart device on a subject, the system comprising:a processing module having instructions stored, therein, the instructions, when executed by one or more processors, cause the one or more processors to:activate a sensor initialization protocol of the smart device;collect raw signals using an integrated sensory module simultaneously;pre-process the raw signals collected via a pre-processing module to obtain pre-processed signals;analyse the pre-processed signals via an analytical module (500) to generate cardiovascular health assessments based on pulse transit time (PTT) and ventricular transit time (VTT) signals;display the cardiovascular health measurements via a user interface of the smart device or a display unit, along with confidence metrics and signal quality indicators; and,store the cardiovascular health measurements in a database with data validation checks for continuous monitoring.15.The system according to claim 14, wherein the processing module further comprising of a central processing unit (CPU) , a math processing unit, a graphic processing unit (GPU) and a tensor processing unit (TPU) .16.The system according to claim 14, wherein the display unit includes but is not limited to a liquid crystal display, a light emitting display, or any other suitable display.17.The system according to claim 14, wherein the smart device includes but is not limited to a mobile phone, VR glasses and the like.18.The system according to claim 14, wherein the subject includes but is not limited to a human or an animal.19.The system according to claim 14, wherein the raw signals are physiological signals obtained from finger pulse, heart sound and cardiac motion.20.The system according to claim 14, wherein the sensor initialization protocol is configured to establish initial sensor parameters and baseline conditions for signals acquisition.21.The system according to claim 14, wherein the integrated sensory module further comprises:a phonocardiography (PCG) module (400) configured to obtain raw PCG signals and extract PCG data from the heart sound via a video recording with a microphone enabled;a seismocardiography (SCG) module (300) , configured to obtain raw SCG signals and extract SCG data from the cardiac motion via an accelerometer; and,a photoplethysmography (PPG) module (200) , configured to obtain raw PPG signals and extract PPG data from the finger pulse via a camera.22.The system according to claim 14, wherein raw PCG signals are obtained from at least 30 fps video recordings.23.The system according to claim 14, wherein raw SCG signals are obtained from 100 Hz accelerometer data streams.24.The system according to claim 14, wherein raw PPG signals are obtained from 30 to 120 fps camera feeds.25.The system according to claim 14, wherein the pre-processing module (140) comprises:a signal synchronization module configured to obtain temporal alignments for PPG and SCG signals collected by the integrated sensory module, generating synchronized signals; and,a quality control module that employs a reference point algorithm to evaluate signal quality based on signal quality metrics with pre-determined thresholds of the synchronized signals.26.The system according to claim 14, wherein the signal quality metrics are signal-to-noise ratios, baseline stability and artifacts presence.27.The system according to claim 14, wherein the quality control module is further configured to filter poor-quality signals if the signals fall below pre-determined thresholds and to extract features with peaks in the signal quality metrics, generating pre-processed signals for analysis by the analytical module (500) .28.The system according to claim 14, wherein the features are cardiovascular parameters, the cardiovascular parameters further comprising:time-series features including but not limited to PPG peaks, PPG 50%peaks, Heart SCG envelopes, first and second (S1, S2) heartbeat sounds from Heart SCG; and,non time-series features including but not limited to PPG notch position, third and fourth (S3, S4) heartbeat sounds, R-R interval, pulse duration, area of diastolic and systolic and contraction amplitude.29.The system according to claim 14, wherein the analytical module (500) (500) comprises:a time-delay calculation module configured to align and compare detected reference points to measure time delays accurately in the PPG, SCG and the PCG features;PTT and VTT calculation modules configured to generate PTT and VTT values and reject the PTT and VTT values below their thresholds based on the signal quality metrics; and,a predetermined parametric and non-parametric algorithm or functionsconfigured to compute cardiovascular health measurements based on an adaptive correlation between the PTT, VTT and the signals.