Continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG)

Abstract

This invention discloses a continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG). The system includes: an intelligent finger cot terminal for acquiring fingertip PPG signals and adjusting the fingertip contact pressure in real time using a closed-loop control formula to maintain a constant pressure state on the finger blood vessels; a mobile computing terminal for receiving fingertip PPG signals and performing preprocessing operations on the fingertip PPG signals; and a cloud service platform for calculating the preprocessed fingertip PPG signals based on a preset blood pressure calculation model to obtain a continuous blood pressure waveform and estimate hemodynamic parameters based on the continuous blood pressure waveform. The preset blood pressure calculation model is specifically a pulse wave analysis model, a pulse wave propagation time model, or a deep learning mapping model. The hemodynamic parameters are specifically stroke volume and cardiac output. Through the mobile computing terminal and the cloud service platform, continuous arterial pressure waveform reconstruction and physiological parameter calculation are realized, and long-term monitoring accuracy can be maintained through a periodic calibration mechanism.

Description

Technical Field

[0001] This application relates to the field of blood pressure measurement technology, specifically to a continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG). Background Technology

[0002] With the increasing incidence of hypertension and cardiovascular diseases, long-term continuous monitoring of arterial blood pressure has become an important requirement for clinical management and remote health monitoring. Currently, the mainstream blood pressure measurement method is still the cuff-based oscillometric method. Although this method is widely used in hospitals and homes, it essentially only provides discrete blood pressure data and cannot capture the dynamic process of blood pressure changes over time, thus failing to meet the need for continuous blood pressure monitoring.

[0003] To achieve continuous non-invasive blood pressure measurement, existing technologies such as Finapres / CNAP incorporate the principle of volume compensation. By adjusting the external pressure applied to the fingertip in real time, the external pressure waveform is made consistent with the instantaneous arterial pressure waveform, thereby reconstructing a continuous arterial pressure curve. Although this technology has high measurement accuracy and clinical value, its reliance on high-speed pressure-driven devices and complex control algorithms results in a large system size, high power consumption, and high motion sensitivity, making it difficult to apply to true wearable scenarios. Furthermore, real-time tracking of instantaneous pressure amplifies mechanical hysteresis and noise interference, which is detrimental to long-term stable monitoring; the signal also drifts significantly when temperature changes, blood flow states change, or fingertip tissue characteristics change.

[0004] On the other hand, blood pressure models based on fingertip PPG signals have received widespread attention in recent years. Some methods attempt to directly map PPG to blood pressure values ​​using pulse wave propagation time, pulse wave profile analysis, or deep learning techniques. However, these methods generally rely on data-driven models, and their stability is greatly affected by ambient light interference, sensor position changes, and individual differences, making it difficult to achieve long-term consistent continuous monitoring. Therefore, there is currently a lack of wearable solutions that simultaneously possess miniaturized device structure, low power consumption, high stability, and long-term continuous blood pressure output capability. Summary of the Invention

[0005] This invention provides a continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) to solve the technical problems described in the background section. The system includes: The intelligent finger sleeve terminal is used to collect PPG signals from the fingertip and the control unit executes a closed-loop control formula to adjust the contact pressure of the fingertip in real time to maintain a constant pressure state on the blood vessels of the finger. A mobile computing terminal is used to receive the fingertip PPG signal and perform preprocessing operations on the fingertip PPG signal; The cloud service platform is used to calculate the preprocessed fingertip PPG signal based on a preset blood pressure calculation model, obtain a continuous blood pressure waveform, and estimate hemodynamic parameters based on the continuous blood pressure waveform. The preset blood pressure calculation model is specifically a pulse wave analysis model, a pulse wave propagation time model, or a deep learning mapping model, and the hemodynamic parameters are specifically stroke volume and cardiac output.

[0006] Preferably, the intelligent finger sleeve terminal specifically includes: A photoplethysmography (PPG) sensor is used to acquire the PPG signal of the fingertip, preferably including a red light source, an infrared light source, and a photodiode; the PPG sensor is used to acquire the PPG signal changes corresponding to different fingertip contact pressures during the voltage adjustment process, so as to provide an observation of changes in blood vessel volume. A pressure sensor is used to collect the fingertip contact pressure in real time and send the feedback signal of the fingertip contact pressure to the control unit; The control unit is used to receive the fingertip PPG signal and the feedback signal, generate a control quantity based on the closed-loop control formula, and output a drive signal to the micro-drive mechanism. A miniature drive mechanism is used to adjust the fingertip contact pressure based on the feedback signal to maintain a constant pressure state on the finger blood vessels.

[0007] Preferably, the closed-loop control formula is as follows: ; Where Pset(t) is the set pressure, Pavg(t) is the target average pressure, VPPG(t) is the real-time blood vessel volume, Vtarget is the target blood vessel volume, and Kp and Kd are the closed-loop control coefficients.

[0008] Preferably, the mobile computing terminal is specifically used for: The fingertip PPG signal was acquired using a 16-bit ADC at a sampling rate of 200Hz. The fingertip PPG signal is input into a high-pass filter to remove baseline drift, and the baseline-drift-removed fingertip PPG signal is input into a low-pass filter to remove high-frequency noise. An adaptive filtering method is used to process the fingertip PPG signal after removing high-frequency filtering noise, resulting in a preprocessed fingertip PPG signal.

[0009] Preferably, the mobile computing terminal further includes a user interface, which is specifically used for: It responds to user input commands and triggers a periodic calibration process, then displays the blood pressure monitoring results to the user.

[0010] Preferably, the periodic calibration process specifically includes: By periodically introducing calibration data from an upper arm electronic blood pressure monitor and dynamically updating the preset blood pressure calculation model using a periodic calibration formula, long-term stable, continuous, and accurate non-invasive blood pressure monitoring can be achieved. The periodic calibration formula is specifically as follows: ; Wherein, BPcorrected(t) is the blood pressure value calculated by continuous PPG mapping, BPcalib is the calibration value measured by the upper arm blood pressure monitor, and α is the dynamic calibration weight.

[0011] Preferably, the cloud service platform specifically includes: A blood pressure calculation unit is used to map the preprocessed fingertip PPG signal into a continuous upper arm arterial blood pressure waveform based on the preset blood pressure calculation model. The hemodynamic analysis unit is used to estimate hemodynamic parameters based on the continuous upper arm arterial blood pressure waveforms.

[0012] Preferably, the hemodynamic analysis unit is specifically used for: Detect feature points of continuous upper arm arterial blood pressure waveforms, divide the continuous upper arm arterial blood pressure waveforms into single cardiac cycles, and calculate the cycle length; Perform systolic pressure integral on the segmented blood pressure waveform and calculate the systolic area and diastolic area. Based on Windkessel's theory of the arterial system, an approximate relationship between stroke volume and the systolic integral of the pressure waveform is determined. The rate of change of systolic pressure was introduced, and the effect of the rate of increase of maximum pressure on stroke volume was incorporated into the enhanced model. The stroke volume is determined based on the enhanced model, and the cardiac output is determined based on the formula for calculating stroke volume and cardiac output.

[0013] Preferably, the enhancement model is specifically: ; Where SV is stroke volume, A s For the area during the contraction period, k s The individualized coefficient, k, is determined by the calibrated blood pressure value measured by the upper arm blood pressure monitor and vascular compliance. dp The weighting factor, obtained through calibration, reflects the contribution of contractile dynamics to the ejection volume.

[0014] Preferably, the formula for calculating cardiac output is as follows: ; CO is the cardiac output, HR=1 / T is the heart rate obtained from the continuous blood pressure waveform, and T is the cardiac cycle.

[0015] Compared with the prior art, the present invention has the following beneficial effects: This invention discloses a continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG). The system includes: an intelligent fingertip terminal for acquiring fingertip PPG signals and, through a control unit, executing a closed-loop control formula to adjust the fingertip contact pressure in real time to maintain a constant pressure state on the finger's blood vessels; a mobile computing terminal for receiving the fingertip PPG signals and performing preprocessing operations on them; and a cloud service platform for calculating the preprocessed fingertip PPG signals based on a preset blood pressure calculation model to obtain a continuous blood pressure waveform and estimating hemodynamic parameters based on the continuous blood pressure waveform. The preset blood pressure calculation model is specifically a pulse wave analysis model, a pulse wave propagation time model, or a deep learning mapping model. The hemodynamic parameters are specifically stroke volume and cardiac output. Through the mobile computing terminal and the cloud service platform, continuous arterial pressure waveform reconstruction and physiological parameter calculation are achieved, and long-term monitoring accuracy can be maintained through a periodic calibration mechanism. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 The diagram shows a schematic of a continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) according to an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0019] In view of the problems of existing non-invasive blood pressure monitoring technologies, such as complex structure, high power consumption, significant long-term drift and insufficient stability, this invention is proposed.

[0020] Therefore, the technical problem to be solved by this invention is to propose a method and system for continuous non-invasive blood pressure and hemodynamic parameter monitoring based on fingertip PPG. This system acquires photoelectric pulse volume signals through a miniaturized intelligent fingertip terminal and employs a novel average pressure regulation strategy to maintain stable pressure on the fingertip vessels, thereby obtaining high-quality PPG signals. Combined with a mobile terminal and a cloud computing platform, this invention can achieve continuous arterial pressure waveform reconstruction and physiological parameter calculation, and can maintain long-term monitoring accuracy through a periodic calibration mechanism.

[0021] To address the aforementioned technical problems, the present invention provides a monitoring system comprising an intelligent finger cot terminal, a mobile computing terminal, and a cloud service platform. The intelligent finger cot terminal integrates a photoplethysmography (PPG) sensor, a pressure sensor, and a miniature drive mechanism to apply controlled external pressure to the fingertip and acquire high-quality PPG signals. Unlike existing real-time pressure tracking technologies such as CNAP, this invention employs an active pressure regulation method based on mean arterial pressure (MAP). Through closed-loop control, the pressure applied by the intelligent finger cot gradually approaches and stabilizes at a pressure level consistent with the user's MAP, thereby maintaining relatively uniform pressure on the fingertip vessels. Under this stable pressure condition, the nonlinear coupling between changes in fingertip vessel volume and external pressure is significantly reduced, resulting in reduced baseline drift, a more stable waveform, and a higher signal-to-noise ratio in the obtained PPG signal, providing a more reliable input signal for subsequent blood pressure calculation.

[0022] The acquired PPG signal is uploaded to a cloud service platform via a mobile computing terminal. The cloud-based blood pressure calculation unit maps the stable PPG waveform under pressure to a continuous upper arm arterial pressure waveform based on a preset pressure-light volume model, pulse wave analysis model, pulse wave propagation time model, or deep learning mapping model. The mapping model is calibrated using an upper arm electronic blood pressure monitor during initial system use to determine the model parameters corresponding to the user's systolic blood pressure, diastolic blood pressure, and mean arterial pressure. During long-term monitoring, the system can dynamically correct the model parameters based on intermittent upper arm calibration results to compensate for drift caused by changes in physiological state, sensor position, or differences in tissue characteristics.

[0023] After obtaining continuous arterial pressure waveforms, the hemodynamic analysis unit of this invention performs waveform profile analysis on the reconstructed blood pressure signal, including but not limited to valve opening point determination, peak value analysis, rate of rise estimation, and ejection time inference; thereby, advanced hemodynamic parameters such as stroke volume, cardiac output, peripheral resistance, and vascular compliance can be calculated. Based on continuous blood pressure and hemodynamic parameters, the intelligent intervention engine identifies abnormal blood pressure patterns in real time, including hypertensive emergency risk, hypoperfusion state, and abnormal vascular stiffness, and provides graded warnings and personalized health recommendations through mobile terminals.

[0024] Combination Figure 1 The present invention proposes a continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG), specifically as follows: The intelligent finger sleeve terminal is used to collect PPG signals from the fingertip and the control unit executes a closed-loop control formula to adjust the contact pressure of the fingertip in real time to maintain a constant pressure state on the blood vessels of the finger. A mobile computing terminal is used to receive the fingertip PPG signal and perform preprocessing operations on the fingertip PPG signal; The cloud service platform is used to calculate the preprocessed fingertip PPG signal based on a preset blood pressure calculation model, obtain a continuous blood pressure waveform, and estimate hemodynamic parameters based on the continuous blood pressure waveform. The preset blood pressure calculation model is specifically a pulse wave analysis model, a pulse wave propagation time model, or a deep learning mapping model, and the hemodynamic parameters are specifically stroke volume and cardiac output.

[0025] In a preferred embodiment of this application, the intelligent finger sleeve terminal specifically includes: A photoplethysmography (PPG) sensor is used to acquire the PPG signal of the fingertip, preferably including a red light source, an infrared light source, and a photodiode; and is used to acquire the PPG signal changes corresponding to different fingertip contact pressures during the pressure adjustment process to characterize changes in vascular volume. A pressure sensor is used to collect the fingertip contact pressure in real time and send the feedback signal of the fingertip contact pressure to the control unit; The control unit is used to receive the PPG signal and the feedback signal, generate a control quantity based on the closed-loop control formula, and output a drive signal to the micro-drive mechanism. A miniature drive mechanism is used to adjust the fingertip contact pressure based on the feedback signal to maintain a constant pressure state on the finger blood vessels.

[0026] In a preferred embodiment of this application, the closed-loop control formula is specifically as follows: ; Where Pset(t) is the set pressure, Pavg(t) is the target average pressure, VPPG(t) is the real-time blood vessel volume, Vtarget is the target blood vessel volume, and Kp and Kd are the closed-loop control coefficients.

[0027] Specifically, the smart finger cot terminal integrates a photoplethysmography (PPG) sensor, a pressure sensor, and a miniature drive mechanism to apply controlled external pressure to the fingertip and acquire high-quality PPG signals. Unlike existing real-time pressure tracking technologies such as CNAP, this invention employs an active pressure regulation method based on mean arterial pressure (MAP). Through closed-loop control, the pressure applied by the smart finger cot gradually approaches and stabilizes at a pressure level consistent with the user's MAP, thereby maintaining a relatively uniform pressure state on the fingertip vessels. Under this stable pressure condition, the nonlinear coupling between changes in fingertip vessel volume and external pressure is significantly reduced, resulting in less baseline drift, a more stable waveform, and a higher signal-to-noise ratio in the obtained PPG signal, providing a more reliable input signal for subsequent blood pressure calculation.

[0028] Specifically, the intelligent finger cot terminal is used to collect PPG signals from the fingertip. It incorporates red and infrared light sources, photodiodes, a miniature air pump, a pressure sensor, and a control unit. It actively adjusts finger pressure through volume compensation or flattening tension methods to maintain a constant mean vascular pressure and acquire high-quality, low-drift PPG signals. The mobile computing terminal is responsible for receiving data, providing the user interface, and triggering periodic calibration processes. The cloud service platform includes a blood pressure calculation unit, a hemodynamic analysis unit, and an intelligent intervention engine, enabling continuous blood pressure calculation, hemodynamic parameter analysis, and abnormal pattern recognition.

[0029] Specifically, the system uses a closed-loop control micro-pump to maintain the finger's blood vessels in a state of average pressure, actively adjusting the pressure to ensure a stable PPG signal. Unlike the real-time pressure tracking of existing CNAP systems, this invention employs an average pressure active tracking strategy, controlling the average pressure rather than instantaneous pressure to simulate the "uniform compression" process of the blood vessels, thereby improving the accuracy of blood pressure measurement. The closed-loop control formula is: ; Where Pset(t) is the set pressure, Pavg(t) is the target average pressure, VPPG(t) is the real-time blood vessel volume, Vtarget is the target blood vessel volume, and Kp and Kd are closed-loop control coefficients. This embodiment achieves rapid response and pressure stability by fine-tuning the proportional and differential parameters, balancing measurement accuracy and comfort.

[0030] Specifically, the closed-loop control formula is used to describe the overall closed-loop pressure regulation control logic of the smart finger sleeve terminal. The control module in the terminal executes the following: the control unit reads the photoplethysmography pulse wave signal and pressure feedback signal, calculates the control quantity, and drives the micro air pump to adjust the fingertip contact pressure in real time to keep the blood vessels in a state of average pressure.

[0031] In a preferred embodiment of this application, the mobile computing terminal is specifically used for: The fingertip PPG signal was acquired using a 16-bit ADC at a sampling rate of 200Hz. The fingertip PPG signal is input into a high-pass filter to remove baseline drift, and the baseline-drift-removed fingertip PPG signal is input into a low-pass filter to remove high-frequency noise. An adaptive filtering method is used to process the fingertip PPG signal after removing high-frequency filtering noise, resulting in a preprocessed fingertip PPG signal.

[0032] In a preferred embodiment of this application, the mobile computing terminal further includes a user interface, which is specifically used for: It responds to user input commands and triggers a periodic calibration process, then displays the blood pressure monitoring results to the user.

[0033] In a preferred embodiment of this application, the periodic calibration process specifically includes: By periodically introducing calibration data from an upper arm electronic blood pressure monitor and dynamically updating the preset blood pressure calculation model using a periodic calibration formula, long-term stable, continuous, and accurate non-invasive blood pressure monitoring can be achieved. The periodic calibration formula is specifically as follows: ; Wherein, BPcorrected(t) is the blood pressure value calculated by continuous PPG mapping, BPcalib is the calibration value measured by the upper arm blood pressure monitor, and α is the dynamic calibration weight.

[0034] Specifically, this solution employs a 16-bit ADC with a sampling rate of 200 Hz. The acquired signal is first filtered through a high-pass filter (0.5Hz cutoff) to remove baseline drift, and then through a low-pass filter (15Hz cutoff) to remove high-frequency noise. Simultaneously, to suppress slight finger movements and ambient light interference, an adaptive filtering method is used to process the signal, ensuring the stability of the continuous blood pressure waveform. The preprocessed PPG signal is uploaded to a cloud platform via a mobile terminal, providing high-quality input for blood pressure calculation and hemodynamic parameter analysis.

[0035] Specifically, the acquired PPG signal is uploaded to a cloud service platform via a mobile computing terminal. The cloud-based blood pressure calculation unit maps the stable PPG waveform under pressure to a continuous upper arm arterial pressure waveform based on a preset pressure-light volume model, pulse wave analysis model, pulse wave propagation time model, or deep learning mapping model. The mapping model is calibrated using an upper arm electronic blood pressure monitor during initial system use to determine the model parameters corresponding to the user's systolic blood pressure, diastolic blood pressure, and mean arterial pressure. During long-term monitoring, the system can dynamically correct the model parameters based on intermittent upper arm calibration results to compensate for drift caused by changes in physiological state, sensor position, or differences in tissue characteristics.

[0036] In a preferred embodiment of this application, the cloud service platform specifically includes: A blood pressure calculation unit is used to map the preprocessed fingertip PPG signal into a continuous upper arm arterial blood pressure waveform based on the preset blood pressure calculation model. The hemodynamic analysis unit is used to estimate hemodynamic parameters based on the continuous upper arm arterial blood pressure waveforms.

[0037] In a preferred embodiment of this application, the hemodynamic analysis unit is specifically used for: Detect feature points of continuous upper arm arterial blood pressure waveforms, divide the continuous upper arm arterial blood pressure waveforms into single cardiac cycles, and calculate the cycle length; Perform systolic pressure integral on the segmented blood pressure waveform and calculate the systolic area and diastolic area. Based on Windkessel's theory of the arterial system, an approximate relationship between stroke volume and the systolic integral of the pressure waveform is determined. The rate of change of systolic pressure was introduced, and the effect of the rate of increase of maximum pressure on stroke volume was incorporated into the enhanced model. The stroke volume is determined based on the enhanced model, and the cardiac output is determined based on the formula for calculating stroke volume and cardiac output.

[0038] In a preferred embodiment of this application, the enhancement model specifically refers to: ; Where SV is stroke volume, A s For the area during the contraction period, k s The individualized coefficient, k, is determined by the calibrated blood pressure value measured by the upper arm blood pressure monitor and vascular compliance. dp The weighting factor, obtained through calibration, reflects the contribution of contractile dynamics to the ejection volume.

[0039] In a preferred embodiment of this application, the formula for calculating cardiac output is specifically as follows: ; CO is the cardiac output, HR=1 / T is the heart rate obtained from the continuous blood pressure waveform, and T is the cardiac cycle.

[0040] Specifically, after obtaining continuous arterial pressure waveforms, the hemodynamic analysis unit of this invention performs waveform contour analysis on the reconstructed blood pressure signal, including but not limited to valve opening point determination, peak value analysis, rate of rise estimation, and ejection time inference; thereby, advanced hemodynamic parameters such as stroke volume, cardiac output, peripheral resistance, and vascular compliance can be calculated. Based on continuous blood pressure and hemodynamic parameters, the intelligent intervention engine identifies abnormal blood pressure patterns in real time, including hypertensive emergency risk, hypoperfusion state, and abnormal vascular stiffness, and provides graded warnings and personalized health suggestions through mobile terminals.

[0041] Specifically, this embodiment calculates hemodynamic parameters using continuous blood pressure waveforms, and calculates stroke volume (SV) and cardiac output (CO) by analyzing the pressure change characteristics of a single cardiac cycle. The blood pressure waveform of this invention is derived from the volume compensation mechanism of mean pressure steady state, which has a stable baseline, complete waveform, and no low-frequency drift, making the estimation of hemodynamic parameters based on pressure waveforms highly reliable.

[0042] First, by detecting characteristic points such as the pacemaker, systolic peak, and dicrotic notch of the pressure waveform, the continuous blood pressure waveform is divided into individual cardiac cycles, and the cycle length is calculated. Then, the systolic pressure integral is performed on the segmented pressure waveform, and the systolic area and diastolic area are calculated. The systolic pressure integral can be expressed as: ; Where t0 is the start of the systolic phase and td is the time when the dicrotic notch appears.

[0043] According to Windkessel's theory of the arterial system, stroke volume is approximately proportional to the systolic integral of the pressure waveform, which can be expressed as: ; Where ks is the user-calibrated individualization coefficient, which is determined by the calibrated blood pressure value measured by the upper arm blood pressure monitor and vascular compliance.

[0044] To enhance accuracy, this invention further incorporates the systolic pressure change rate to reflect ventricular ejection dynamics. The effect of the maximum pressure rise rate (dP / dt)max on stroke volume can be incorporated into the enhancement model. ; Among them, kdp is a weighting factor obtained by calibration, which reflects the contribution of contractile dynamics to the ejection volume.

[0045] After obtaining the stroke volume, cardiac output (CO) can be further calculated, defined as: ; Where HR=1 / T is the heart rate obtained from the continuous blood pressure waveform, and T is the cardiac cycle.

[0046] To improve computational robustness, this embodiment uses a sliding window to perform parameter statistics on multiple consecutive cardiac cycles, and reduces the influence of noise and abnormal beats by using moving average or median filtering, thereby obtaining stable and continuous SV and CO curves.

[0047] In summary, this embodiment estimates SV and CO by analyzing the systolic integral of the arterial pressure waveform, the rate of pressure change, and the cardiac cycle. It fully utilizes the high-quality blood pressure waveform under volume compensation of this invention to provide a core calculation method for continuous hemodynamic monitoring.

[0048] Furthermore, the intelligent intervention engine can combine continuous blood pressure and hemodynamic parameters, using rule-based algorithms or deep learning models to identify abnormal blood pressure patterns, such as hypertension risk or hypotension trends, and provide tiered warnings and personalized health recommendations via mobile devices. The system can also record historical blood pressure trends and hemodynamic parameters and upload them to the cloud for remote doctor assessment and chronic disease management, enhancing the practical application value of wearable health monitoring.

[0049] Furthermore, the smart finger sleeve features a miniaturized, low-power design, combined with a flexible wearing structure, ensuring user comfort during extended wear. Closed-loop pressure regulation and periodic calibration strategies enable the system to maintain the stability and accuracy of continuous blood pressure measurements even under changes in finger posture, slight movement disturbances, or changes in the external environment, improving system fault tolerance and reliability. This design is suitable for home, remote monitoring, and mobile wearable health management scenarios.

[0050] In summary, this invention discloses a continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG), belonging to the fields of biomedical engineering, wearable devices, and intelligent health monitoring technology. The system includes: an intelligent fingertip terminal for acquiring fingertip PPG signals; a mobile computing terminal for data reception, interactive display, and triggering calibration processes; and a cloud service platform for performing core computing tasks.

[0051] The core of this invention lies in the following: First, the system utilizes a smart finger sleeve terminal to collect PPG signals in real time, and performs active pressure regulation through a built-in pressure sensor and a micro-drive mechanism to ensure that the finger blood vessels are under uniform pressure, thereby obtaining high-quality, low-drift PPG signals. After the collected signals are uploaded to a cloud service platform via a mobile computing terminal, the blood pressure calculation unit maps the processed PPG signals into continuous upper arm arterial blood pressure waveforms based on a preset blood pressure calculation model (including but not limited to pulse wave analysis models, pulse wave propagation time models, or deep learning mapping models).

[0052] To maintain long-term monitoring accuracy, this invention introduces an upper-arm electronic blood pressure monitor for periodic calibration. By calibrating the user's systolic blood pressure, diastolic blood pressure, and mean arterial pressure, the model parameters are dynamically updated to compensate for signal drift, changes in physiological state, and individual differences. Subsequently, the hemodynamic analysis unit extracts advanced hemodynamic parameters such as stroke volume (SV) and cardiac output (CO) based on the obtained continuous blood pressure waveform using pulse wave contour analysis. Finally, the intelligent intervention engine identifies abnormal blood pressure patterns based on real-time continuous blood pressure and hemodynamic characteristics, and issues graded warnings and personalized suggestions to the user via mobile terminal.

[0053] This invention enables non-invasive, continuous, high-precision, and long-term monitoring of blood pressure and hemodynamic parameters, and is suitable for wearable health monitoring, chronic disease management, and telemedicine scenarios.

[0054] Through the above technical solutions, this invention realizes a low-power, high-precision, and continuously usable non-invasive blood pressure monitoring system on a small fingertip device, solving the technical bottleneck of existing technologies that struggle to balance device portability, system stability, and continuous blood pressure monitoring. This invention not only provides stable continuous blood pressure waveforms but also allows for the calculation of key hemodynamic parameters, enabling its wide application in wearable health monitoring, cardiovascular disease management, chronic disease follow-up, and telemedicine. Finally, this invention breaks through the dependence of existing volumetric compensation technologies on instantaneous pressure synchronization in its pressure regulation strategy. By adopting an average pressure stabilization control method, it constructs a novel blood pressure measurement paradigm with higher fault tolerance, lower mechanical complexity, and insensitivity to noise, demonstrating significant innovation in both its technical approach and system structure.

[0055] Those skilled in the art will understand that the modules in the device can be distributed within the device of the implementation scenario as described, or they can be located in one or more devices different from this implementation scenario, with corresponding changes. The modules of the above-mentioned implementation scenario can be combined into one module, or they can be further divided into multiple sub-modules.

[0056] The serial numbers of the present invention mentioned above are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenarios.

[0057] The above-disclosed examples are only a few specific implementation scenarios of the present invention. However, the present invention is not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. A continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG), characterized in that, The system includes: The intelligent finger sleeve terminal is used to collect PPG signals from the fingertip and adjust the contact pressure of the fingertip in real time through a closed-loop control formula to maintain a constant pressure state on the blood vessels of the finger. A mobile computing terminal is used to receive the fingertip PPG signal and perform preprocessing operations on the fingertip PPG signal; The cloud service platform is used to calculate the preprocessed fingertip PPG signal based on a preset blood pressure calculation model, obtain a continuous blood pressure waveform, and estimate hemodynamic parameters based on the continuous blood pressure waveform. The preset blood pressure calculation model is specifically a pulse wave analysis model, a pulse wave propagation time model, or a deep learning mapping model, and the hemodynamic parameters are specifically stroke volume and cardiac output.

2. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 1, characterized in that, The intelligent finger sleeve terminal specifically includes: A photoplethysmography (PPG) sensor is used to collect the PPG signal of the fingertip photoplethysmography pulse wave. The photoplethysmography sensor includes a red light source, an infrared light source, and a photodiode. A pressure sensor is used to collect the fingertip contact pressure in real time and send the feedback signal of the fingertip contact pressure to the control unit; The control unit is used to receive the fingertip PPG signal and the feedback signal, generate a control quantity based on the closed-loop control formula, and output a drive signal to the micro-drive mechanism. The miniature drive mechanism, including a miniature air pump and a pressure-applying structure connected thereto, is used, under the drive of the control unit, to adjust the fingertip contact pressure based on the feedback signal in order to maintain a constant pressure state on the finger blood vessels.

3. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 2, characterized in that, The closed-loop control formula is as follows: ； Where Pset(t) is the set pressure, Pavg(t) is the target average pressure, VPPG(t) is the real-time blood vessel volume, Vtarget is the target blood vessel volume, and Kp and Kd are the closed-loop control coefficients.

4. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 1, characterized in that, The mobile computing terminal is specifically used for: The fingertip PPG signal was acquired using a 16-bit ADC at a sampling rate of 200Hz. The fingertip PPG signal is input into a high-pass filter to remove baseline drift, and the baseline-drift-removed fingertip PPG signal is input into a low-pass filter to remove high-frequency noise. An adaptive filtering method is used to process the fingertip PPG signal after removing high-frequency filtering noise, resulting in a preprocessed fingertip PPG signal.

5. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 1, characterized in that, The mobile computing terminal also includes a user interface, which is specifically used for: It responds to user input commands and triggers a periodic calibration process, then displays the blood pressure monitoring results to the user.

6. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 5, characterized in that, The periodic calibration process is as follows: By periodically introducing calibration data from an upper arm electronic blood pressure monitor and dynamically updating the preset blood pressure calculation model using a periodic calibration formula, long-term stable, continuous, and accurate non-invasive blood pressure monitoring can be achieved. The periodic calibration formula is specifically as follows: ； Wherein, BPcorrected(t) is the blood pressure value calculated by continuous PPG mapping, BPcalib is the calibration value measured by the upper arm blood pressure monitor, and α is the dynamic calibration weight.

7. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 1, characterized in that, The cloud service platform specifically includes: A blood pressure calculation unit is used to map the preprocessed fingertip PPG signal into a continuous upper arm arterial blood pressure waveform based on the preset blood pressure calculation model. The hemodynamic analysis unit is used to estimate hemodynamic parameters based on the continuous upper arm arterial blood pressure waveforms.

8. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 7, characterized in that, The hemodynamic analysis unit is specifically used for: Detect feature points of continuous upper arm arterial blood pressure waveforms, divide the continuous upper arm arterial blood pressure waveforms into single cardiac cycles, and calculate the cycle length; Perform systolic pressure integral on the segmented blood pressure waveform and calculate the systolic area and diastolic area. Based on Windkessel's theory of the arterial system, an approximate relationship between stroke volume and the systolic integral of the pressure waveform is determined. The rate of change of systolic pressure was introduced, and the effect of the rate of increase of maximum pressure on stroke volume was incorporated into the enhanced model. The stroke volume is determined based on the enhanced model, and the cardiac output is determined based on the formula for calculating stroke volume and cardiac output.

9. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 7, characterized in that, The enhancement model is specifically as follows: ； Where SV is stroke volume, A s For the area during the contraction period, k s The individualized coefficient, k, is determined by the calibrated blood pressure value measured by the upper arm blood pressure monitor and vascular compliance. dp The weighting factor, obtained through calibration, reflects the contribution of contractile dynamics to the ejection volume.

10. The continuous non-invasive blood pressure monitoring system based on fingertip photoplethysmography (PPG) as described in claim 9, characterized in that, The formula for calculating cardiac output is as follows: ； CO is the cardiac output, HR=1 / T is the heart rate obtained from the continuous blood pressure waveform, and T is the cardiac cycle.

Images