System for continuous estimation of hemodynamic parameters by integrating photoplethysmographic and ballistocardiographic signals

Abstract

A system for the continuous estimation of hemodynamic parameters using photoplethysmographic and ballistocardiographic signal integration, wherein the system comprises the following: a photoplethysmographic detection unit configured to emit optical radiation into biological tissue and detect variations in reflected or transmitted light corresponding to pulsating changes in blood volume; a ballistocardiographic detection unit with at least one motion-sensitive sensor for detecting micro-accelerations or displacements of the subject's body caused by the ejection activity of the heart; a signal processing unit operationally connected to the photoplethysmographic acquisition unit and the ballistocardiographic acquisition unit, wherein the signal processing unit comprises an amplification circuit, a filter circuit and an analog-to-digital conversion circuit configured to generate processed digital signals; a synchronization processor configured to align the processed photoplethysmographic and ballistocardiographic signals in time based on identifiable waveform features; a feature extraction processor configured to derive temporal, morphological, and frequency-domain features from the aligned signals; and a hemodynamic estimation processor configured to calculate one or more hemodynamic parameters such as blood pressure, stroke volume, cardiac output and vascular compliance based on the extracted features.

Description

Technical field of the invention

[0001] The present invention relates generally to biomedical signal processing systems and physiological monitoring devices, in particular a system and associated device architecture for the continuous, non-invasive estimation of hemodynamic parameters by the integrated acquisition and processing of photoplethysmographic (PPG) and ballistocardiographic (BCG) signals. The invention further relates to portable and bed-integrated monitoring structures with synchronized sensors, signal processing, adaptive filtering, and computer-aided estimation of cardiovascular parameters such as blood pressure, stroke volume, cardiac output, and vascular compliance. Background of the invention

[0002] Continuous monitoring of hemodynamic parameters is essential for the early detection of cardiovascular abnormalities, the management of chronic diseases, and real-time clinical decision-making. Conventional methods such as invasive arterial catheterization, while offering high accuracy, are associated with risks, discomfort, and limited applicability in outpatient settings. Non-invasive methods such as cuff-based oscillometry are intermittent and unsuitable for continuous monitoring. Photoplethysmography has become a widely used optical method for assessing blood volume changes, while ballistocardiography measures the mechanical recoil forces associated with cardiac output.Existing systems typically use these modalities independently, resulting in limited accuracy and susceptibility to motion artifacts and physiological variability. Therefore, there is a need for an integrated system capable of fusing PPG and BCG signals to enable robust, continuous, and accurate hemodynamic measurements.

[0003] Continuous hemodynamic monitoring has long been an indispensable requirement in clinical medicine, particularly in intensive care, perioperative management, and the treatment of chronic cardiovascular diseases. The timely detection of changes in blood pressure, cardiac output, and vascular resistance can significantly influence treatment success. Traditionally, invasive arterial catheterization has been considered the gold standard for continuous blood pressure measurement. This involves inserting a catheter directly into an artery and connecting it to a pressure transducer to enable real-time measurements. Although this method offers high accuracy and waveform fidelity, it carries significant risks such as infection, thrombosis, bleeding, and vascular injury. Therefore, its use is limited to intensive care settings under strict medical supervision.Furthermore, the need for qualified personnel and sterile conditions makes invasive monitoring unsuitable for long-term or outpatient applications.

[0004] In contrast to invasive methods, non-invasive techniques such as oscillometric cuff devices have become widely used in both clinical and home settings due to their simplicity and safety. These devices work by inflating a cuff around an extremity and recording fluctuations in arterial pressure during deflation to estimate systolic and diastolic blood pressure. However, oscillometric devices provide only intermittent measurements and cannot detect rapid hemodynamic changes, which are often crucial for the early detection of cardiovascular instability. Furthermore, repeated cuff inflation can cause discomfort, restrict blood flow, and is impractical for continuous monitoring, especially during sleep or physical activity.The accuracy of such devices is also affected by factors such as cuff position, extremity movements and arterial stiffness, which leads to measurement deviations.

[0005] To meet the need for continuous and non-invasive monitoring, photoplethysmography (PPL) has emerged as a promising technique, particularly with the increasing prevalence of wearable devices such as smartwatches and fitness trackers. PPL utilizes optical sensors to detect changes in light absorption or reflection caused by pulsatile blood volume changes in the microcirculatory bed. The waveform derived from PPL signals provides valuable information about cardiovascular dynamics, including heart rate, pulse wave velocity, and arterial compliance. Several approaches to blood pressure measurement from PPL signals have been proposed, often based on pulse transit time or pulse arrival time, which is derived from the temporal relationship between PPL and ECG signals.Although these methods offer the advantage of non-invasive and continuous measurement, they have significant limitations. PPL signals are highly susceptible to motion artifacts, interference from ambient light, and variations in skin tone and tissue composition. Furthermore, the relationship between photoplethysmographic features and blood pressure is complex and non-linear, often requiring frequent calibration and individual modeling, thus limiting the generalizability and long-term reliability of such systems.

[0006] Ballistocardiography is another non-invasive method for measuring the mechanical recoil force of the body, generated by the heart's pumping of blood into the vessels. Ballistocardiographic signals can be acquired using accelerometers, piezoelectric sensors, or force-sensitive resistors integrated into portable devices, chairs, or beds. The characteristic waveform of ballistocardiography, particularly the identification of the J-wave, which corresponds to rapid blood acceleration, provides information about cardiac contractility and stroke volume. Unlike photoplethysmography, ballistocardiography is less susceptible to optical interference and, in some applications, can be measured discreetly and without direct skin contact.However, ballistocardiographic signals are inherently weak and highly sensitive to external disturbances such as body movement, respiration, and environmental vibrations. Signal morphology can vary considerably between individuals and measurement conditions, complicating reliable feature extraction and interpretation. Furthermore, ballistocardiography alone does not provide sufficient information to directly and accurately determine blood pressure, limiting its use as a standalone monitoring method.

[0007] Electrocardiography-based methods have also been investigated for hemodynamic estimation, particularly in combination with photoplethysmography to calculate pulse wave transit time, which is inversely proportional to blood pressure. While this multimodal approach improves the accuracy of the estimation compared to single-modality systems, it introduces additional complexity regarding sensor placement, signal synchronization, and energy consumption. Electrocardiographic electrodes require good skin contact and are prone to problems such as electrode drying, motion artifacts, and discomfort during prolonged wear. Furthermore, the reliance on multiple sensor modalities increases overall system costs and reduces everyday acceptance.

[0008] Recent advances in wearable and telemedicine monitoring technologies have led to the development of cuffless blood pressure monitoring systems that utilize machine learning and data-driven models to estimate hemodynamic parameters from physiological signals. These systems typically use features extracted from photoplethysmographic signals, sometimes in combination with other signals such as electrocardiography or impedance cardiography, and employ regression or deep learning techniques to predict blood pressure values. While such approaches yield promising results in controlled environments, their performance in everyday use often deteriorates due to variations in sensor placement, ambient noise, and interindividual differences.The lack of standardized datasets and validation protocols further complicates the evaluation of these systems, and regulatory approval remains a significant challenge due to concerns regarding accuracy and reliability.

[0009] Another limitation of existing solutions lies in their inability to provide a comprehensive hemodynamic assessment beyond basic parameters such as heart rate and blood pressure. Parameters like stroke volume, cardiac output, and systemic vascular resistance are crucial for understanding overall cardiovascular function but are typically measured using specialized equipment such as echocardiography or thermodilution techniques, which are not suitable for continuous or non-invasive monitoring. Consequently, a gap exists in current technologies regarding the holistic and continuous assessment of cardiovascular health in a non-invasive and user-friendly manner.

[0010] Many existing systems operate under the assumption of static physiological conditions and do not account for dynamic changes in vascular tone, hydration status, and autonomic regulation, which can significantly influence hemodynamic parameters. This leads to reduced accuracy during activities such as exercise, postural changes, or stress. The calibration effort is also a significant drawback, as many cuffless systems require regular recalibration using conventional cuff-based measurements, limiting their practicality and user independence.

[0011] Besides technical challenges, ergonomic and user-friendly aspects play a crucial role in the acceptance of continuous monitoring systems. Devices requiring close skin contact, multiple sensors, or complex setup procedures often result in low user acceptance. Battery life, data privacy, and seamless integration into existing healthcare infrastructure are further factors that influence the effectiveness of such systems in practical use.

[0012] Despite significant advances in non-invasive monitoring technologies, existing solutions exhibit trade-offs between accuracy, continuity, ease of use, and robustness to external disturbances. Single-modal systems such as photoplethysmography or ballistocardiography are insufficient for reliable and comprehensive hemodynamic assessment, while multimodal systems are often associated with increased complexity and reduced practicality. These limitations underscore the need for an integrated approach that leverages complementary sensor modalities, robust signal processing techniques, and adaptive modeling to enable precise, continuous, and non-invasive hemodynamic monitoring suitable for both clinical and everyday use. Summary of the invention

[0013] The present invention describes a system and associated device for the continuous estimation of hemodynamic parameters by synchronous acquisition of photoplethysmographic and ballistocardiographic signals, as well as multidomain-based signal fusion and computer-aided modeling. The system comprises a sensor structure with an optical PPG acquisition unit and a mechanical BCG acquisition unit integrated into a portable device or carrier surface, a signal conditioning and synchronization unit for temporal alignment of the acquired signals, and a processing unit for extracting morphological, temporal, and frequency-domain-based features for estimating cardiovascular parameters. The invention also includes adaptive calibration and machine learning to improve estimation accuracy under varying physiological and environmental conditions.The embodiment of the device integrates sensors, processing circuits and communication interfaces in a compact, ergonomically designed housing suitable for continuous monitoring.

[0014] The main objective of the present invention is to provide a system for the continuous, non-invasive estimation of hemodynamic parameters through the integrated acquisition and processing of photoplethysmographic and ballistocardiographic signals. This overcomes the limitations of intermittent and invasive monitoring methods. A further objective of the invention is the precise and timely estimation of critical cardiovascular parameters such as systolic and diastolic blood pressure, stroke volume, cardiac output, and vascular compliance by means of synchronized multimodal physiological signal analysis. Another objective of the invention is to improve signal reliability and robustness by combining optical and mechanical sensor modalities. This compensates for the respective weaknesses of the individual modalities, particularly with regard to motion artifacts, environmental influences, and physiological variability.

[0015] A further objective of the invention is to provide a structurally integrated monitoring device configured as a portable or surface-mounted system. Photoplethysmographic and ballistocardiographic sensors are arranged in a compact and ergonomically optimized manner to enable continuous and unobtrusive monitoring in both clinical and non-clinical settings. Another objective is the integration of a signal processing and synchronization mechanism that aligns the temporal characteristics of multiple physiological signals to enable the precise calculation of derived parameters such as pulse transit time and pulse arrival time. A further objective is the implementation of advanced feature extraction methods that capture morphological, temporal, and frequency-domain-specific features of the acquired signals for improved hemodynamic assessment.

[0016] A further objective of the invention is to provide a computing architecture that utilizes hybrid modeling approaches and combines physiological models with data-driven techniques to improve the accuracy of the estimation and its adaptability to different individuals and varying physiological states. Furthermore, the invention aims to integrate adaptive calibration and real-time parameter adjustment mechanisms that reduce the reliance on frequent external calibration and ensure consistent performance over extended monitoring periods. Additionally, the invention provides an artifact detection and correction function that improves signal quality by identifying and mitigating the effects of noise, motion, and external disturbances.

[0017] A further objective of the invention is the seamless transmission of processed hemodynamic data to external devices or remote monitoring systems via integrated wireless interfaces, thereby facilitating telemedicine applications and long-term patient monitoring. Another objective is to ensure low power consumption and efficient resource utilization within the device to support extended operation in portable configurations. The invention also aims to provide a scalable and modular system architecture that can be adapted to various application scenarios, including home care, hospital monitoring, and remote patient treatment.

[0018] A further aim of the invention is to improve wearing comfort and compliance by minimizing sensor inconspicuousness and eliminating cumbersome components such as inflatable cuffs or invasive probes. Finally, the invention aims to provide a comprehensive and reliable solution for continuous cardiovascular monitoring that supports the early detection of abnormalities, improves clinical decision-making, and contributes to better patient outcomes through timely and precise hemodynamic assessment. BRIEF DESCRIPTION OF THE IMAGE

[0019] These and other features, aspects and advantages of the present invention will be better understood if the following detailed description is read with reference to the accompanying drawing, in which the same symbols represent the same parts: Fig. Figure 1 shows a block diagram of a system for the continuous estimation of hemodynamic parameters using photoplethysmographic and ballistocardiographic signal integration.

[0020] Furthermore, those skilled in the art will recognize that the elements in the drawing are simplified and not necessarily drawn to scale. For example, the flowcharts illustrate the process by highlighting the main steps to facilitate understanding of the present disclosure. With regard to the construction of the device, one or more components may be represented in the drawing by conventional symbols. The drawing may show only the specific details relevant to understanding the embodiments of the present disclosure, so as not to clutter the drawing with details that are already apparent to those skilled in the art from the description contained herein. Detailed description of the invention

[0021] To facilitate understanding of the principles of the invention, reference is made below to the embodiment shown in the drawing, which is described using specific terms. It is understood, however, that this does not limit the scope of protection of the invention. Rather, modifications and further developments of the depicted system, as well as further applications of the inventive principles shown therein, are conceivable, insofar as they would normally occur to a person skilled in the art in the field of the invention.

[0022] It will be clear to those skilled in the art that the foregoing general description and the following detailed description are exemplary and explanatory of the invention and are not to be understood as a limitation of it.

[0023] References to “an aspect”, “another aspect”, or similar phrases in this description mean that a particular feature, structure, or property described in connection with the embodiment is included in at least one embodiment of the present disclosure. Therefore, phrases such as “in one embodiment”, “in another embodiment”, and similar expressions in this description may, but do not necessarily, all refer to the same embodiment.

[0024] The terms "includes," "comprehensive," or similar expressions denote non-exclusive inclusion. Thus, a procedure or method containing a list of steps does not only include those steps but may also include further steps not explicitly listed or inherent in the procedure or method. Likewise, the statement "includes..." for one or more devices, subsystems, elements, structures, or components, without further limitations, does not preclude the existence of other devices, subsystems, elements, structures, or components.

[0025] Unless otherwise defined, all technical and scientific terms used herein have the same meanings generally known to those skilled in the art in the field to which this invention belongs. The systems, methods, and examples described herein serve only for illustration and are not to be understood as limiting.

[0026] Embodiments of the present disclosure are described in detail below with reference to the attached drawing.

[0027] Fig.1 eigt a system for the continuous estimation of hemodynamic parameters by means of photoplethysmographic and ballistocardiographic signal integration. The system 100 comprises: a photoplethysmographic acquisition unit (102) that emits optical radiation into biological tissue and detects variations in the reflected or transmitted light corresponding to pulsatile changes in blood volume; a ballistocardiographic acquisition unit (104) with at least one motion-sensitive sensor for detecting micro-accelerations or displacements of the body caused by cardiac ejection activity; a signal processing unit (106) that is operationally linked to the photoplethysmographic and ballistocardiographic acquisition units and includes amplification, filtering, and analog-to-digital converter circuits for generating processed digital signals;a synchronization processor (108) that temporally synchronizes the processed photoplethysmographic and ballistocardiographic signals based on identifiable waveform features; a feature extraction processor (110) configured to derive temporal, morphological, and frequency-domain-specific features from the aligned signals; and a hemodynamic estimation processor (112) configured to calculate one or more hemodynamic parameters, such as blood pressure, stroke volume, cardiac output, and vascular compliance, based on the extracted features.

[0028] In one embodiment, the photoplethysmographic detection unit (102) comprises a plurality of light-emitting diodes operating at different wavelengths and at least one photodetector arranged to detect fluctuations in light absorption corresponding to arterial pulsations. The unit further comprises a control circuit configured to modulate the emission intensity and the sampling frequency to optimize signal quality under various physiological conditions.

[0029] In one embodiment, the ballistocardiographic detection unit (104) comprises at least one microelectromechanical accelerometer embedded in a structural mount configured to maintain mechanical coupling with the subject's body, the accelerometer being configured to detect multi-axis motion signals corresponding to body movements caused by the heart.

[0030] In one embodiment, the signal processing unit (106) comprises a cascaded filter arrangement with a bandpass filter for isolating physiological frequency components and a notch filter for suppressing mains interference. The amplification circuit includes a low-noise amplifier for amplifying weak signal components before digitization.

[0031] In one embodiment, the synchronization processor (108) is configured to identify reference points, including systolic peaks in the photoplethysmographic signal and corresponding mechanical peaks in the ballistocardiographic signal, and to calculate time intervals that represent physiological delays between the cardiac ejection phase and the arrival of the peripheral pulse.

[0032] In one embodiment, the feature extraction processor (110) is configured to calculate the slope of the waveform, the amplitude variation, the pulse width, the area under the waveform and higher-order derivatives from the photoplethysmographic signal and is further configured to calculate the peak-to-peak intervals, the energy distribution and the waveform morphology from the ballistocardiographic signal.

[0033] In one embodiment, the hemodynamic estimation processor (112) comprises a computing unit configured to implement a hybrid estimation approach that combines physiological modeling and data-driven regression techniques, wherein the processor is further configured to generate continuous estimates of systolic and diastolic blood pressure without the use of an inflatable cuff.

[0034] In an embodiment further comprising an adaptive calibration unit configured to update estimation parameters based on patient-specific baseline data and continuously improve estimation accuracy by compensating for variations in vascular properties and physiological state.

[0035] In an embodiment further comprising an artifact detection unit configured to calculate signal quality indices based on noise levels, motion disturbances and waveform consistency, and selectively excludes or corrects damaged signal sections using interpolation or filtering techniques.

[0036] In one embodiment, the photoplethysmographic acquisition unit (102) and the ballistocardiographic acquisition unit (104) are integrated into a portable structure comprising a flexible housing for attachment to an extremity, the structural arrangement ensuring stable sensor positioning and consistent signal acquisition.

[0037] The described system is realized through tangible, physical components that operate via measurable electrical and mechanical interactions, thus ensuring practical applicability. The photoplethysmographic acquisition unit consists of discrete optical emitters and photodetectors arranged in a physical array to transmit and receive light through biological tissue, generating analog electrical signals that correspond to changes in blood volume. The ballistocardiographic acquisition unit comprises motion-sensitive elements such as micro-accelerometers or displacement sensors mounted in a structural holder. This holder ensures direct mechanical coupling to the subject's body to detect cardiac motion.The signals are routed through a signal conditioning unit, which includes physically implemented amplifier circuits, analog filters, and analog-to-digital converters, converting the raw analog input signals into digitized electrical representations. Temporal alignment is achieved by dedicated processing circuits configured to detect waveform features through clocked hardware operations. Feature extraction is performed by electronic processing units that calculate signal characteristics through deterministic circuit behavior. Hemodynamic estimation is carried out by a computing unit that processes the extracted parameters to generate physiological outputs.Additional units such as calibration and artifact handling are also implemented as electronic circuits configured to adjust and refine signal interpretation based on measurable input variations.

[0038] The present invention, relating to the claimed system, implements a method for the continuous estimation of hemodynamic parameters through tightly coupled acquisition, synchronization, feature derivation, and adaptive computer-aided processing of photoplethysmographic and ballistocardiographic signals. The process begins with the simultaneous acquisition of optical and mechanical physiological signals. The photoplethysmographic acquisition unit emits controlled optical radiation into the tissue and detects time-varying intensity modulations corresponding to the pulsating blood volume. The ballistocardiographic acquisition unit, on the other hand, detects micro-accelerations associated with body movement caused by the heart.The captured analog signals are first subjected to a signal processing unit, where low-noise amplification is performed to emphasize weak physiological components, followed by cascade filtering. The filter stage includes a bandpass filter, which preserves the frequency components associated with the heart cycles while simultaneously suppressing baseline deviations and high-frequency noise, as well as a notch filter to eliminate interference from external electrical sources. The processed signals are then digitized using high-resolution analog-to-digital conversion to maintain waveform fidelity for subsequent processing.

[0039] After digitization, the signals are fed to a synchronization processor, which temporally aligns the photoplethysmographic and ballistocardiographic waveforms. The synchronization process involves identifying reference points in each signal. The photoplethysmographic waveform is analyzed to identify characteristic features such as systolic peaks, diastolic inflection points, and the waveform foot. The ballistocardiographic waveform is analyzed to identify corresponding mechanical events, including the primary ejection peak. The synchronization processor calculates the temporal offsets between these reference points and aligns the signals in a common temporal reference frame. This alignment enables the precise determination of physiologically relevant intervals corresponding to the propagation of the pulse wave from the heart to the peripheral measurement sites.

[0040] Following synchronization, the feature extraction processor performs a cross-domain analysis of the aligned signals to obtain a comprehensive feature set. In the time domain, the processor calculates parameters such as peak-to-peak intervals, rise time, decay time, pulse width, and interbeat variability. Morphological analysis determines waveform amplitude, slope characteristics, and curvature variations, including higher derivations that capture the inflection dynamics of the photoplethysmographic waveform. Simultaneously, the ballistocardiographic signal is analyzed to extract kinetic features such as peak amplitude, peak interval, and signal energy distribution across the cardiac cycle. Frequency domain analysis is performed using spectral decomposition to quantify the power distribution within physiological frequency bands, thereby capturing autonomic modulation and cardiovascular variability.The extracted features are normalized and combined into a feature vector that represents the current physiological state of the subject.

[0041] The hemodynamic estimation processor receives the feature vector and applies a hybrid computational approach that integrates physiological modeling with adaptive, data-driven estimation. The processor first maps time intervals and waveform features to underlying cardiovascular parameters by leveraging established relationships between pulse propagation characteristics and arterial pressure dynamics. This mapping is further refined by a regression-based computational layer that adjusts parameter estimates based on learned associations between input features and hemodynamic reference values. The estimation process is continuously updated using an adaptive calibration unit that incorporates patient-specific background information and dynamically adjusts internal coefficients to account for variations in vascular tone, compliance, and external conditions.The calibration process works iteratively, minimizing deviations between predicted trends and expected physiological patterns through parameter adjustment, thereby improving the long-term stability of the estimate.

[0042] An artifact detection unit operates in parallel to ensure signal integrity by evaluating signal quality indices derived from noise levels, waveform consistency, and motion-induced disturbances. If signal segments are identified as faulty, the system applies corrective measures such as interpolation, smoothing, or selective exclusion to prevent error propagation during the estimation process. The feedback controller interacts with the signal conditioning unit to dynamically adjust gain, filter bandwidth, and sampling parameters based on real-time signal quality assessment, thus maintaining optimal acquisition conditions under varying physiological and environmental conditions.

[0043] The processed hemodynamic parameters, including continuous estimates of blood pressure, stroke volume, and cardiac output, are stored in internal memory and transmitted to external devices or monitoring systems via a communication unit. The system ensures the temporal continuity of the estimated parameters, thus enabling the detection of rapid fluctuations and long-term trends. In embodiments where the sensor components are distributed across a portable structure and a base, the communication processor ensures precise temporal synchronization of the data streams through coordinated timestamping and alignment procedures.

[0044] The overall technical concept is designed for real-time operation with deterministic latency, ensuring that data acquisition, processing, and estimation are executed within predefined timeframes. The integration of multimodal signals enhances robustness by compensating for the limitations of individual sensor modalities. Deviations or impairments in one signal are balanced by complementary information from another signal. The adaptive nature of the computational process enables the system to maintain its accuracy even under dynamic physiological conditions such as movement, changes in posture, and varying levels of physical activity.Through this coordinated sequence of data acquisition, synchronization, feature extraction, adaptive estimation and feedback control, the system achieves continuous, non-invasive and reliable monitoring of hemodynamic parameters, suitable for both clinical and outpatient applications.

[0045] In one embodiment, the invention provides a hemodynamic monitoring device consisting of a multilayered structure. This structure comprises an upper sensor interface layer for contact with a test subject, a sensor embedding layer with optical emitters, photodetectors, and microelectromechanical accelerometers, a signal conditioning circuit with amplifiers, filters, and analog-to-digital converters, and a processing and communication layer with a microprocessor, memory unit, and wireless transmission circuitry. The device is positioned either as a wearable wristband for attachment to an extremity or as a pad integrated into the bed beneath the test subject to capture BCG signals through body micromovements. The structure ensures mechanical stability, noise isolation, and optimal sensor coupling for precise signal acquisition.

[0046] The photoplethysmographic acquisition unit comprises at least one light-emitting diode (LED) emitting light in a predefined wavelength range, including infrared and red spectrum, and at least one photodetector for measuring variations in reflected or transmitted light intensity corresponding to pulsating blood flow. The ballistocardiographic acquisition unit comprises one or more accelerometers or piezoelectric sensors for detecting micro-accelerations or displacement signals generated by cardiac body movements. The signals acquired by both modalities are transmitted to a signal processing unit comprising low-noise amplifiers, bandpass filters, and a digitization circuit for removing baseline deviations, motion artifacts, and high-frequency noise components.

[0047] The processed signals are fed to a synchronization processor configured to align characteristic waveform features such as systolic peaks, diastolic dips, and BCG-J wave components. The processor calculates time intervals, including pulse transit time and pulse arrival time, by correlating the onset of BCG events with corresponding PPG waveform features. Additionally, morphological parameters such as waveform amplitude, slope, area under the curve, and second derivative indices are extracted from the PPG signal, while kinetic energy features and waveform morphology are derived from the BCG signal.

[0048] A computational unit for estimating measured values ​​is configured to employ a hybrid modeling approach that combines physics-based hemodynamic equations with data-driven regression techniques. The unit uses extracted features as input for a trained model to estimate systolic and diastolic blood pressure, stroke volume, cardiac output, and arterial stiffness indices. The model is dynamically calibrated against patient-specific baseline parameters and continuously updated using adaptive filtering to compensate for physiological changes.

[0049] The system also includes an artifact detection and correction mechanism that calculates signal quality indices and either corrects segments with excessive noise or motion artifacts via interpolation or excludes them from analysis. A feedback control adjusts the sensor gain and filter parameters in real time to optimize signal acquisition.

[0050] In an alternative embodiment, the device is integrated into a mattress or chair structure, with the BCG sensors distributed on a flexible substrate to capture spatially resolved mechanical signals, while the PPG sensors are integrated into a peripheral, portable unit. The system establishes wireless synchronization between the two sensor components to ensure temporal coherence.

[0051] The communication unit is configured to transmit processed hemodynamic parameters to an external display device or remote server via wireless protocols such as Bluetooth Low Energy or WLAN. The system also supports data logging and cloud-based analysis for long-term monitoring and predictive assessment.

[0052] The present invention enables continuous, non-invasive monitoring of hemodynamic parameters with improved accuracy through multimodal signal integration. The fusion of PPG and BCG signals increases robustness against noise and motion artifacts, while adaptive modeling ensures personalized and dynamic measurement. The device's design allows for comfortable and discreet use in both clinical and home settings, thus contributing to the early detection and improved treatment of cardiovascular diseases.

[0053] The present invention relates to the field of biomedical instrumentation and physiological monitoring systems, in particular a system for the continuous, non-invasive determination of hemodynamic parameters by integrated processing of photoplethysmographic and ballistocardiographic signals. The invention further relates to portable and surface-mounted monitoring devices with synchronized sensor, signal processing, and computing units for real-time acquisition of cardiovascular dynamics, including blood pressure, cardiac output, stroke volume, and vascular characteristics.

[0054] The drawing and the preceding description illustrate embodiments. Those skilled in the art will recognize that one or more of the described elements can be combined to form a single functional element. Alternatively, certain elements can be divided into several functional elements. Elements of one embodiment can be added to another. For example, the process flows described here can be modified and are not limited to the manner described herein. Furthermore, the actions of a flowchart need not be performed in the sequence shown; nor do all actions necessarily need to be carried out. Actions that do not depend on other actions can be performed in parallel with the other actions. The scope of protection of the embodiments is in no way limited by these specific examples. Numerous variations, whether explicitly stated in the description or not, such as...Differences in structure, dimensions, and materials are possible. The scope of protection of the embodiments is at least as comprehensive as described by the following claims.

[0055] The advantages, other benefits, and problem solutions have been described above with reference to specific embodiments. However, the advantages, benefits, problem solutions, and any components that can effect or enhance an advantage, benefit, or solution are not to be construed as critical, necessary, or essential features or components of the claims. REFERENCES 100 A system for the continuous estimation of hemodynamic parameters by means of the integration of photoplethysmographic and ballistocardiographic signals. 102 Photoplethysmographic recording unit 104 Ballistocardiographic recording unit 106 Signal processing unit 108 Synchronization processor 110 Feature Extraction Processor 112 Hemodynamic Estimation Processor

Claims

[1] A system for the continuous estimation of hemodynamic parameters using photoplethysmographic and ballistocardiographic signal integration, wherein the system comprises: a photoplethysmographic detection unit configured to emit optical radiation into biological tissue and detect variations in reflected or transmitted light corresponding to pulsating changes in blood volume; a ballistocardiographic detection unit with at least one motion-sensitive sensor for detecting micro-accelerations or displacements of the subject's body caused by the ejection activity of the heart; a signal processing unit operationally connected to the photoplethysmographic acquisition unit and the ballistocardiographic acquisition unit, wherein the signal processing unit comprises an amplification circuit, a filter circuit and an analog-to-digital conversion circuit configured to generate processed digital signals; a synchronization processor configured to align the processed photoplethysmographic and ballistocardiographic signals in time based on identifiable waveform features; a feature extraction processor configured to derive temporal, morphological, and frequency-domain features from the aligned signals; and a hemodynamic estimation processor configured to calculate one or more hemodynamic parameters such as blood pressure, stroke volume, cardiac output and vascular compliance based on the extracted features. [2] System according to claim 1, wherein the photoplethysmographic detection unit comprises a plurality of light-emitting diodes operating at different wavelengths and at least one photodetector arranged to detect variations in light absorption corresponding to arterial pulsations, and wherein the unit further comprises a control circuit configured to modulate the emission intensity and sampling frequency to optimize signal quality under different physiological conditions. [3] System according to claim 1, wherein the ballistocardiographic detection unit comprises at least one microelectromechanical accelerometer embedded in a structural mount configured to maintain mechanical coupling with the subject's body, and wherein the accelerometer is configured to detect multi-axis motion signals corresponding to body movements caused by the heart. [4] System according to claim 1, wherein the signal processing unit comprises a cascaded filter arrangement comprising a bandpass filter for isolating physiological frequency components and a notch filter for suppressing mains disturbances, and wherein the amplification circuit comprises a low-noise amplifier for amplifying weak signal components prior to digitization. [5] System according to claim 1, wherein the synchronization processor is configured to identify reference points including systolic peaks in the photoplethysmographic signal and corresponding mechanical peaks in the ballistocardiographic signal and to calculate time intervals representing physiological delays between the cardiac ejection phase and the arrival of the peripheral pulse. [6] System according to claim 1, wherein the feature extraction processor is configured to calculate the slope of the waveform, the amplitude variation, the pulse width, the area under the waveform and higher-order derivatives from the photoplethysmographic signal and is further configured to calculate the peak-to-peak intervals, the energy distribution and the waveform morphology from the ballistocardiographic signal. [7] System according to claim 1, wherein the hemodynamic estimation processor comprises a computing unit configured to implement a hybrid estimation approach combining physiological modeling and data-driven regression techniques, and wherein the processor is further configured to generate continuous estimates of systolic and diastolic blood pressure without the use of an inflatable cuff. [8] System according to claim 1, further comprising an adaptive calibration unit configured to update estimation parameters based on patient-specific baseline data and continuously improve estimation accuracy by compensating for fluctuations in vascular properties and physiological condition. [9] System according to claim 1, further comprising an artifact detection unit configured to calculate signal quality indices based on noise levels, motion disturbances and waveform consistency, and to selectively exclude or correct damaged signal sections using interpolation or filtering techniques. [10] System according to claim 1, wherein the photoplethysmographic acquisition unit and the ballistocardiographic acquisition unit are integrated into a portable structure comprising a flexible housing for attachment to an extremity, and wherein the structural arrangement ensures stable sensor positioning and consistent signal acquisition.

Images