PPG pulse wave principle blood pressure measuring system based on intelligent ring

Abstract

The invention discloses a PPG pulse wave principle blood pressure measurement system based on an intelligent ring, and relates to the field of artificial intelligence, and the system comprises a position layout module, a signal acquisition and preprocessing module, a calculation and modeling module, an error calibration module and a blood pressure measurement module. Through deep analysis of historical data and utilization of a deep learning model, a PPG optical sensor is adopted, blood flow changes are optically detected, noninvasive measurement is achieved, and discomfort caused by inflation of a traditional cuff is avoided. Through a low-power-consumption mode, all-weather monitoring of blood pressure changes can be achieved, and interference-free ambulatory blood pressure tracking is achieved. The intelligent ring is worn on the finger, the blood flow at the tail end is rich, the PPG signal is clearer and more stable, and the measurement precision is higher. The PPG signal quality of different finger parts is automatically detected through a multi-sensor signal quality adaptive selection algorithm, and the index finger with the optimal signal is selected for measurement, so that the accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a blood pressure measurement system based on the PPG pulse wave principle of a smart ring. Background Art

[0002] Currently, blood pressure measurement primarily relies on cuff-type blood pressure monitors, which use the Korotkoff sound method or the oscillometric method. However, these monitors present several challenges: Traditional blood pressure monitors require manual inflation of the cuff, making each measurement difficult to perform. Some users may experience elevated blood pressure readings in medical settings due to stress or anxiety, compromising diagnosis. Nocturnal blood pressure fluctuations are crucial for diagnosing cardiovascular disease, but traditional cuff-type blood pressure monitors can disrupt sleep when used at night, making them unsuitable for long-term monitoring. Single measurements fail to reflect dynamic changes in blood pressure, hindering hypertension management and personalized health assessments.

[0003] In recent years, blood pressure measurement methods based on PPG (photoplethysmography) and PTT (pulse transit time) have gradually developed. Their core principles include: PPG sensors detect changes in vascular volume using light of different wavelengths, thereby acquiring pulse wave signals. By calculating PTT (the time difference between blood pumping from the heart to the finger), blood pressure levels can be indirectly estimated. Incorporating parameters such as heart rate and blood oxygen saturation further improves the accuracy of blood pressure prediction. Although wearable devices such as smart bracelets and smartwatches have attempted to use PPG to measure blood pressure, technical bottlenecks remain. The wearer's position (e.g., wrist) is easily affected by hand movements, making the PPG signal susceptible to artifacts. Most current devices use simple empirical regression models and fail to incorporate individual physiological characteristics for personalized correction. Existing smartwatches often use intermittent measurement, making them incapable of 24-hour continuous blood pressure monitoring. Most smart devices fail to effectively integrate HRV analysis to analyze nighttime blood pressure fluctuations, making it difficult to provide a comprehensive health assessment. Summary of the Invention

[0004] In view of the problems that existing blood pressure measurement is not convenient for all-day monitoring, nighttime blood pressure monitoring is difficult, and personalized blood pressure analysis is difficult to achieve, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is to achieve non-invasive, convenient, and high-precision blood pressure monitoring around the clock.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a blood pressure measurement system based on the PPG pulse wave principle of a smart ring, which includes a position layout module for integrating a dual-wavelength LED light source and a photoelectric sensor on the inner ring surface of the smart ring, detecting the PPG signal quality of different finger parts through a signal quality adaptive selection algorithm of multiple photoelectric sensors, and selecting the index finger as the measurement position; a signal acquisition and preprocessing module for acquiring PPG waveforms through a high-sampling rate ADC, measuring blood oxygen saturation using a dual-wavelength light source to compensate for the influence of optical scattering, removing motion artifacts through adaptive Kalman filtering, and removing high-frequency noise in combination with wavelet transform; a calculation and modeling module The block is used to extract PPG feature points through signal processing algorithms. The feature points include peaks and troughs, rising edge slope and time center, and then calculate the pulse wave transmission time, and then use the linear regression model to predict blood pressure; the error calibration module is used to optimize blood pressure prediction based on historical data, and build a personalized blood pressure calibration curve based on the user's cuff blood pressure monitor measurement data to improve measurement accuracy; the blood pressure measurement module is used to sample once every 10 seconds with low power consumption in continuous monitoring mode; in fast measurement mode, high-density data is collected within 2 minutes, blood pressure is calculated and immediate feedback is provided; in sleep monitoring mode, PPG and HRV are combined to analyze nighttime blood pressure fluctuations and evaluate sleep health status.

[0008] As a preferred solution of the PPG pulse wave principle blood pressure measurement system based on the smart ring of the present invention, wherein: the signal quality adaptive selection algorithm of multiple photoelectric sensors is used to detect the PPG signal quality of different finger parts, and the index finger is selected as the measurement position, including:

[0009] The raw PPG signal collected by each photosensor goes through the following steps:

[0010] A high-pass filter was used to remove DC drift, and a 5 Hz low-pass filter was used to remove high-frequency noise. The amplitude of each channel signal was then normalized.

[0011] The PPG signal quality is scored in combination with the signal-to-noise ratio, and the signal-to-noise ratio formula is as follows:

[0012]

[0013] Among them, P signal is the power of the main frequency band of the PPG signal, P noise It is the power of the non-pulse wave frequency band. Set the threshold to ensure that the SNR is greater than 20dB.

[0014] As a preferred solution of the PPG pulse wave principle blood pressure measurement system based on the smart ring of the present invention, wherein: the signal quality adaptive selection algorithm of multiple photoelectric sensors is used to detect the PPG signal quality of different finger parts, and the index finger is selected as the measurement position, and further includes:

[0015] Calculate the heart rate HR for each channel calc And estimate the baseline value HR with heart rate ref Compare:

[0016] ΔHR=|HR calc -HR ref |

[0017] Set the threshold to ensure the error ΔHR < 3bpm;

[0018] Then calculate the peak-valley detection consistency of 10 consecutive PPG cycles to ensure that the peak position deviation of adjacent cycles is less than 5%; then calculate the comprehensive quality score of all measurement points. The comprehensive quality score calculation formula is as follows:

[0019] Q=w1·SNR+w2·CC+w3·(1-ΔHR)

[0020] Among them, w1, w2, w3 are weights, SNR, CC, ΔHR are normalized to between 0 and 1, the measurement position with the highest score is selected, and the threshold Q is set threshold , make sure the selected signal is of good quality.

[0021] As a preferred solution of the blood pressure measurement system based on the PPG pulse wave principle of the smart ring described in the present invention, wherein: adaptive Kalman filtering is used to remove motion artifacts, and wavelet transform is combined to remove high-frequency noise, including:

[0022] A state space model of the PPG signal is established, and the expression of the state space model of the PPG signal is:

[0023] x k =Ax k-1 +w k

[0024] y k =Hx k +v k

[0025] Among them, x k is the true PPG state variable, A is the state transition matrix, wk is the process noise, yk is the PPG value measured by the sensor, H is the observation matrix, and vk is the measurement noise;

[0026] An adaptive adjustment strategy is used to estimate the signal change rate in real time. If the signal changes significantly, R is increased to reduce the filter's trust in the measured value. Motion status is detected, and high-frequency energy changes in accelerometer or PPG signals are used to dynamically adjust Q and R.

[0027] As a preferred solution of the blood pressure measurement system based on the PPG pulse wave principle of the smart ring described in the present invention, wherein: adaptive Kalman filtering is used to remove motion artifacts, and wavelet transform is combined to remove high-frequency noise, and further includes:

[0028] Perform discrete wavelet transform on the PPG signal x(t):

[0029]

[0030] Among them, c j is the approximate coefficient, retaining PPG information, d j is the detail coefficient, which contains noise information;

[0031] Then the detail coefficient d j Perform soft threshold denoising:

[0032]

[0033] Where λ is the threshold, σ is the noise standard deviation, and n is the data length;

[0034] The processed c j and The inverse wavelet transform is performed to obtain the denoised PPG signal, and the processed signal is further optimized by sliding mean filtering.

[0035] As a preferred solution of the PPG pulse wave principle blood pressure measurement system based on the smart ring of the present invention, wherein: PPG feature points are extracted by signal processing algorithm, pulse wave transit time is calculated, and then blood pressure is predicted using a linear regression model, including:

[0036] In the PPG waveform, the characteristic points include the peak value, which represents the peak of cardiac contraction, the valley value, which represents the state of cardiac relaxation, the maximum point of the first-order derivative, which represents the moment when the blood flow velocity reaches the maximum, and the inflection point of the second-order derivative, which represents the compliance of the blood vessels;

[0037] Compute the first-order derivative by differencing:

[0038]

[0039] The peak position of the first derivative corresponds to the time point when the blood flows fastest during the heart contraction;

[0040] Then calculate the second derivative of the PPG signal:

[0041]

[0042] The inflection point of the second-order derivative is used to evaluate vascular elasticity and assist in calculating pulse wave transmission time.

[0043] The adaptive threshold method was used to detect the systolic peak and diastolic trough of the PPG waveform.

[0044] Then calculate the pulse wave transit time:

[0045] PTT=T SP -T dPPGmax

[0046] Among them, T SP is the peak time of the PPG signal, T dPPGmax is the time of maximum value of the first-order derivative of PPG;

[0047] Through the linear regression model, systolic blood pressure (SBP) and diastolic blood pressure (DBP) were calculated based on PTT. The relationship between PTT and blood pressure satisfies the following:

[0048] BP=A+B·PTT+C·HR

[0049] Where BP is blood pressure, A, B, C are regression coefficients to be fitted, PTT is pulse wave transit time, and HR is an additional influencing factor;

[0050] Calculate PTT in real time and input PTT and HR into a linear model:

[0051] SBP=A1+B1·PTT+C1·HR

[0052] DBP=A2+B2·PTT+C2·HR

[0053] Systolic and diastolic blood pressures were calculated.

[0054] As a preferred solution of the PPG pulse wave principle blood pressure measurement system based on the smart ring of the present invention, it optimizes blood pressure prediction by combining historical data, constructs a personalized blood pressure calibration curve based on the user's cuff blood pressure monitor measurement data, and improves measurement accuracy, including:

[0055] The system records the measurement time, systolic and diastolic blood pressure, as well as the pulse wave transmission time and heart rate collected by the smart ring. It performs multiple measurements to ensure a balanced distribution of data at different times and under different conditions. The data is stored in the user's personal database, correlating measurements at different times with the user's age, gender, weight, and vascular condition information to form a personalized health profile.

[0056] The error between the blood pressure predicted by the smart ring and the actual blood pressure measured by the blood pressure cuff is calculated, and the deviation trend is observed. The measurement data for multiple time periods is compiled to analyze whether the error changes over time and with blood pressure levels. Based on historical data, the smart ring generates a blood pressure calibration curve for the user. The curve reflects the systematic deviation between the blood pressure predicted by the smart ring and the actual blood pressure, and the prediction result is then adjusted.

[0057] If the smart ring is found to systematically underestimate blood pressure within the hypertension range, the predicted value in this range will be adjusted upward in future measurements to ensure that it is closer to the actual blood pressure.

[0058] As a preferred solution of the blood pressure measurement system based on the PPG pulse wave principle of the smart ring described in the present invention, wherein:

[0059] In continuous monitoring mode, the system samples data every 10 seconds with low power consumption. In rapid measurement mode, high-density data is collected within 2 minutes, blood pressure is calculated, and immediate feedback is provided. In sleep monitoring mode, PPG and HRV are combined to analyze nighttime blood pressure fluctuations and assess sleep health, including:

[0060] In continuous measurement mode, intelligent data filtering is used. If the sampling error is large for three consecutive times, the abnormal data will be automatically discarded and uploaded to the smart ring for internal storage. Trend analysis will be performed every hour. The average blood pressure value is calculated every hour and a blood pressure curve for the whole day is drawn for the user to view. If the blood pressure is found to be continuously higher or lower than the user's baseline blood pressure, the smart ring will remind the user and ask whether a more accurate measurement is needed.

[0061] In the fast measurement mode, the user manually triggers this mode. The smart ring samples the PPG signal at a high frequency within 2 minutes, combines the pulse wave transmission time and heart rate data to calculate the blood pressure. Within 3 minutes, the sampling frequency is increased to 5-10 times per second to ensure measurement accuracy.

[0062] In sleep monitoring mode, samples are taken every 2 minutes for 2 hours before falling asleep to analyze the changes in the user's blood pressure during sleep; the sampling frequency is reduced to once every 10-15 minutes during deep sleep to reduce energy consumption; the sampling frequency is increased to once every 5 minutes from 5 to 7 in the morning; vascular tension is analyzed through PPG signals, and night-time blood pressure changes are calculated in combination with PTT, sympathetic and parasympathetic nerve activity are calculated, the night-time autonomic nervous system regulation ability is analyzed, and whether there are abnormal night-time blood pressure fluctuations is evaluated; if abnormal night-time blood pressure is found to be abnormally high, the smart ring will provide a health reminder in the morning; after waking up in the morning, the user will view the night-time blood pressure report through the mobile phone APP, including the night-time average blood pressure curve, blood pressure fluctuation trend and autonomic nervous system activity assessment. If an abnormal blood pressure pattern is detected, the system will provide targeted suggestions.

[0063] In a second aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the blood pressure measurement system based on the PPG pulse wave principle of the smart ring as described in the first aspect of the present invention are implemented.

[0064] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the blood pressure measurement system based on the PPG pulse wave principle of a smart ring as described in the first aspect of the present invention are implemented.

[0065] The beneficial effects of the present invention are as follows: the present invention adopts a PPG optical sensor to optically detect changes in blood flow, thereby achieving non-invasive measurement and avoiding the discomfort caused by traditional cuff inflation. Through low-power mode, blood pressure changes can be monitored around the clock to achieve interference-free dynamic blood pressure tracking. It is suitable for hypertensive patients, pregnant women, the elderly and people at high risk of cardiovascular disease who need long-term blood pressure monitoring. Traditional smart watches and bracelets are worn on the wrist, and the PPG signal is easily affected by hand movements, while smart rings are worn on the fingers, with rich blood flow at the end, and the PPG signal is clearer and more stable, and the measurement accuracy is higher. Through the multi-sensor signal quality adaptive selection algorithm, the PPG signal quality of different finger parts is automatically detected, and the index finger with the best signal is selected for measurement to improve accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0067] Figure 1 This is a structural diagram of the blood pressure measurement system based on the PPG pulse wave principle of the smart ring.

[0068] Figure 2 This is a computer device diagram of a blood pressure measurement system based on the PPG pulse wave principle of a smart ring. DETAILED DESCRIPTION

[0069] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0070] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0071] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0072] Example 1

[0073] Reference Figures 1 and 2 , which is the first embodiment of the present invention, provides a blood pressure measurement system based on the PPG pulse wave principle of a smart ring, including:

[0074] S1: A dual-wavelength LED light source and a photoelectric sensor are integrated on the inner surface of the smart ring. The PPG signal quality of different fingers is detected through the signal quality adaptive selection algorithm of multiple photoelectric sensors, and the index finger is selected as the measurement position.

[0075] Preferably, the raw PPG signal collected by each photosensor undergoes the following steps:

[0076] A high-pass filter was used to remove DC drift, and a 5 Hz low-pass filter was used to remove high-frequency noise. The amplitude of each channel signal was then normalized.

[0077] The PPG signal quality is scored in combination with the signal-to-noise ratio, and the signal-to-noise ratio formula is as follows:

[0078]

[0079] Among them, P signal is the power of the main frequency band of the PPG signal, P noise It is the power of the non-pulse wave frequency band. Set the threshold to ensure that the SNR is greater than 20dB.

[0080] Preferably, calculate the heart rate HR of each channel calc And estimate the baseline value HR with heart rate ref Compare:

[0081] ΔHR=|HR calc -HR ref |

[0082] Set the threshold to ensure the error ΔHR < 3bpm;

[0083] Then calculate the peak-valley detection consistency of 10 consecutive PPG cycles to ensure that the peak position deviation of adjacent cycles is less than 5%; then calculate the comprehensive quality score of all measurement points. The comprehensive quality score calculation formula is as follows:

[0084] Q=w1·SNR+w2·CC+w3·(1-ΔHR)

[0085] Among them, w1, w2, w3 are weights, SNR, CC, ΔHR are normalized to between 0 and 1, the measurement position with the highest score is selected, and the threshold Q is set threshold , make sure the selected signal is of good quality.

[0086] Furthermore, the PPG signal quality is scored based on the calculated SNR value to ensure that the SNR is ≥ 20 dB to guarantee signal reliability.

[0087] SNR>30dB indicates a high-quality signal and can be used directly for blood pressure calculation; 20dB≤SNR≤30dB indicates an acceptable signal; SNR<20dB indicates poor signal quality and requires changing the optical wavelength or adjusting the sensor position.

[0088] If the SNR is lower than the set threshold of 20dB, the system will adjust the light intensity to increase the LED light source power, optimize the measurement position to detect multiple finger PPG signals, and select the optimal signal to alert the user.

[0089] Further, Example 1: Index finger signal quality is the best: HR calc =72bpm, and HR ref =73bpm error ΔHR=1bpm, meeting the threshold condition; peak deviation = 2.8%, less than 5%, indicating that the signal is stable; SNR = 28dB, normalized score 0.9, calculated Q = 0.4×0.9+0.3×0.97+0.3×(1-1 / 3)=0.87, meeting Q threshold (0.8), select the index finger as the measurement position; Example 2: The signal quality of the middle finger is poor: HR calc =75bpm, and HR ref =72bpm error ΔHR=3bpm, just reaching the maximum threshold, peak deviation = 7%, exceeding 5%, not meeting the requirements, SNR = 22dB, normalized score 0.7, calculated Q = 0.4×0.7+0.3×0.8+0.3×(1-3 / 3)=0.64, lower than Q threshold (0.8), the middle finger channel is not selected.

[0090] S2: The PPG waveform is acquired through a high-sampling-rate ADC, and the blood oxygen saturation is measured using a dual-wavelength light source to compensate for the influence of optical scattering. Motion artifacts are removed through adaptive Kalman filtering, and high-frequency noise is removed in combination with wavelet transform.

[0091] Preferably, a state space model of the PPG signal is established, and the expression of the PPG signal state space model is:

[0092] x k =Ax k-1 +w k

[0093] y k =Hx k +v k

[0094] Among them, x k is the true PPG state variable, A is the state transition matrix, wk is the process noise, yk is the PPG value measured by the sensor, H is the observation matrix, and vk is the measurement noise;

[0095] An adaptive adjustment strategy is used to estimate the signal change rate in real time. If the signal changes significantly, R is increased to reduce the filter's trust in the measured value. Motion status is detected, and high-frequency energy changes in accelerometer or PPG signals are used to dynamically adjust Q and R.

[0096] Preferably, the PPG signal x(t) is subjected to discrete wavelet transform:

[0097]

[0098] Among them, c j is the approximate coefficient, retaining PPG information, d j is the detail coefficient, which contains noise information;

[0099] Then the detail coefficient d j Perform soft threshold denoising:

[0100]

[0101] Where λ is the threshold, σ is the noise standard deviation, and n is the data length;

[0102] The processed c j and The inverse wavelet transform is performed to obtain the denoised PPG signal, and the processed signal is further optimized by sliding mean filtering.

[0103] Furthermore, the rate of change of the PPG signal (ΔPPG) is calculated:

[0104] ΔPPG k =|y k -y k-1 |

[0105] If the signal change rate is large (ΔPPG>threshold), it means that the signal fluctuation is large, increase R k , reduce the trust in the measured value; if the signal change rate is small (ΔPPG < threshold), it means the signal is stable, reduce R k , increasing trust in the measured values;

[0106] Use the high-frequency energy changes of the accelerometer or PPG signal to detect the motion state and calculate the high-frequency energy of the PPG signal:

[0107]

[0108] If the high-frequency energy E HF Increase, indicating a stronger state of motion, increasing Q k , to adapt to the drastic changes in PPG signals:

[0109]

[0110] Where Q0 is the initial process noise covariance, β is the adjustment coefficient;

[0111] Ten subjects were selected to collect PPG signals in resting state (sitting), light activity (walking) and vigorous exercise (running):

[0112] state ΔPPG (mV) High frequency energy EHF R changes Q changes Rest 2.1 10.5 <![CDATA[1.0×R0]]> <![CDATA[1.0×Q0]]> Light activity 5.7 25.8 <![CDATA[1.5×R0]]> <![CDATA[1.8×Q0]]> vigorous exercise 12.3 58.2 <![CDATA[3.2×R0]]> <![CDATA[4.5×Q0]]>

[0113] Example 1: Resting state (stable signal)

[0114] Set the initial value R0 = 0.1, Q0 = 0.01; calculate the signal change rate ΔPPG = 2.1mV; calculate the high-frequency energy E HF =10.5; Since the signal is stable, keep R=1.0×R0,Q=1.0×Q0;

[0115] Example 2: Light activity (walking)

[0116] Calculate the signal change rate ΔPPG = 5.7mV; calculate the high-frequency energy E HF =25.8; adjust the parameters R = 1.5 × R0, Q = 1.8 × Q0, indicating that the Kalman filter begins to reduce its trust in the measured values ​​and increase its reliance on historical states;

[0117] Example 3: Vigorous exercise (running)

[0118] Calculate the signal change rate ΔPPG = 12.3m; calculate the high-frequency energy E HF =58.2; adjusting parameters R = 3.2×R0, Q = 4.5×Q0, which shows that the filter greatly reduces the trust in the measurement value and enhances the adaptability of state estimation.

[0119] S3: The PPG feature points are extracted through signal processing algorithms. The feature points include peaks and troughs, rising edge slope and time center. The pulse wave transmission time is then calculated, and the blood pressure is predicted using a linear regression model.

[0120] Preferably, in the PPG waveform, the characteristic points include a peak value, representing the peak of cardiac contraction, a valley value, representing the cardiac diastolic state, a maximum point of the first-order derivative, representing the moment when the blood flow velocity reaches the maximum, and a second-order derivative inflection point, representing vascular compliance;

[0121] Compute the first-order derivative by differencing:

[0122]

[0123] The peak position of the first derivative corresponds to the time point when the blood flows fastest during the heart contraction;

[0124] Then calculate the second derivative of the PPG signal:

[0125]

[0126] The inflection point of the second-order derivative is used to evaluate vascular elasticity and assist in calculating pulse wave transmission time.

[0127] The adaptive threshold method was used to detect the systolic peak and diastolic trough of the PPG waveform.

[0128] Then calculate the pulse wave transit time:

[0129] PTT=T SP -T dPPGmax

[0130] Among them, T SP is the peak time of the PPG signal, T dPPGmax is the time of maximum value of the first-order derivative of PPG;

[0131] Through the linear regression model, systolic blood pressure (SBP) and diastolic blood pressure (DBP) were calculated based on PTT. The relationship between PTT and blood pressure satisfies the following:

[0132] BP=A+B·PTT+C·HR

[0133] Where BP is blood pressure, A, B, C are regression coefficients to be fitted, PTT is pulse wave transit time, and HR is an additional influencing factor;

[0134] Calculate PTT in real time and input PTT and HR into a linear model:

[0135] SBP=A1+B1·PTT+C1·HR

[0136] DBP=A2+B2·PTT+C2·HR

[0137] Systolic and diastolic blood pressures were calculated.

[0138] Furthermore, a smartwatch with a PPG sampling rate of 50Hz and a standard blood pressure monitor were selected. The subjects included 10 healthy volunteers and 30 sets of data were measured. The measurement scenarios included resting state, light exercise, and intense exercise. The experimental data are as follows:

[0139]

[0140]

[0141] Through linear regression we get:

[0142] SBP=140-0.2×PTT+0.15×HR

[0143] DBP=85-0.1×PTT+0.08×HR

[0144] Example 1: Resting state measurement values ​​PTT = 170ms, HR = 65bpm, calculate blood pressure:

[0145] SBP=140-0.2×170+0.15×65=118mmHg

[0146] DBP=85-0.1×170+0.08×65=78mmHgmmHg;

[0147] Example 2: Intense exercise measurement values ​​PTT = 120ms, HR = 110bpm, calculate blood pressure:

[0148] SBP=140-0.2×120+0.15×110=135mmHg

[0149] DBP=85-0.1×120+0.08×110=90mmHg.

[0150] S4: Optimize blood pressure predictions based on historical data, build personalized blood pressure calibration curves based on user cuff blood pressure monitor measurement data, and improve measurement accuracy.

[0151] Preferably, the measurement time, systolic blood pressure, diastolic blood pressure, and pulse wave transmission time and heart rate collected by the smart ring are recorded, and multiple measurements are taken to ensure that the data is evenly distributed at different times and states; the data is stored in the user's personal database, and the measurements at different times are associated with the user's age, gender, weight, and vascular condition information to form a personalized health record;

[0152] The error between the blood pressure predicted by the smart ring and the actual blood pressure measured by the blood pressure cuff is calculated, and the deviation trend is observed. The measurement data for multiple time periods is compiled to analyze whether the error changes over time and with blood pressure levels. Based on historical data, the smart ring generates a blood pressure calibration curve for the user. The curve reflects the systematic deviation between the blood pressure predicted by the smart ring and the actual blood pressure, and the prediction result is then adjusted.

[0153] If the smart ring is found to systematically underestimate blood pressure within the hypertension range, the predicted value in this range will be adjusted upward in future measurements to ensure that it is closer to the actual blood pressure.

[0154] Furthermore, historical measurement data was collected and a deviation curve was fitted between the blood pressure predicted by the smart ring and the actual blood pressure:

[0155] ΔBP=f(PTT,HR,Time,State)

[0156] This curve is used to reflect systematic deviations and make corrections;

[0157] If the predicted value in the hypertension range is too low, adjust the model to improve the predicted value in the hypertension range; if the predicted value in the hypotension range is too high, reduce the predicted value in the hypotension range;

[0158] The adjusted blood pressure calculation formula is as follows:

[0159] BP adj =BP ring +ΔBP corr

[0160] Among them, ΔBP corr is the correction value, which comes from the personalized calibration curve;

[0161] Example 1: Error Calculation

[0162] The calculated data is as follows:

[0163]

[0164] Error trend: In the low blood pressure range (08:00, PTT=170), the smart ring underestimated the blood pressure (-4 mmHg); in the high blood pressure range (22:00, PTT=140), the smart ring overestimated the blood pressure (+2 mmHg). The error showed a trend of increasing with decreasing PTT.

[0165] Example 2: Calibrated Prediction

[0166] Calculate the correction value from historical data:

[0167] ΔBP corr =0.1×(130-PTT)

[0168] After correction:

[0169] BP adj =BP ring +ΔBP corr

[0170] Calibrated data

[0171] time PTT (ms) Original predicted SBP Calibration value Calibrated SBP Real SBP 08:00 170 118 4 122 122 12:00 160 126 2 128 128 18:00 150 132 -2 130 130 22:00 140 138 -2 136 136

[0172] After calibration, the predicted values ​​are closer to the true values.

[0173] S5: In continuous monitoring mode, low power consumption is used to sample once every 10 seconds. In rapid measurement mode, high-density data is collected within 2 minutes, blood pressure is calculated, and immediate feedback is provided. In sleep monitoring mode, PPG and HRV are combined to analyze nighttime blood pressure fluctuations and assess sleep health.

[0174] Preferably, in the continuous measurement mode, intelligent data screening is used. If the sampling error is large for three consecutive times, the abnormal data will be automatically discarded and the data will be uploaded to the smart ring for internal storage and trend analysis every 1 hour. The blood pressure average is calculated every hour and a blood pressure curve for the whole day is drawn for the user to view. If the blood pressure is found to be continuously higher or lower than the user's baseline blood pressure, the smart ring will remind the user and prompt whether a more accurate measurement is needed.

[0175] In the fast measurement mode, the user manually triggers this mode. The smart ring samples the PPG signal at a high frequency within 2 minutes, combines the pulse wave transmission time and heart rate data to calculate the blood pressure. Within 3 minutes, the sampling frequency is increased to 5-10 times per second to ensure measurement accuracy.

[0176] In sleep monitoring mode, samples are taken every 2 minutes for 2 hours before falling asleep to analyze the changes in the user's blood pressure during sleep; the sampling frequency is reduced to once every 10-15 minutes during deep sleep to reduce energy consumption; the sampling frequency is increased to once every 5 minutes from 5 to 7 in the morning; vascular tension is analyzed through PPG signals, and night-time blood pressure changes are calculated in combination with PTT, sympathetic and parasympathetic nerve activity are calculated, the night-time autonomic nervous system regulation ability is analyzed, and whether there are abnormal night-time blood pressure fluctuations is evaluated; if abnormal night-time blood pressure is found to be abnormally high, the smart ring will provide a health reminder in the morning; after waking up in the morning, the user will view the night-time blood pressure report through the mobile phone APP, including the night-time average blood pressure curve, blood pressure fluctuation trend and autonomic nervous system activity assessment. If an abnormal blood pressure pattern is detected, the system will provide targeted suggestions.

[0177] Furthermore, in the continuous measurement mode, the smart ring compares the predicted blood pressure value of each sample with the actual blood pressure value. If the sampling error is greater than the set threshold (for example, more than ±5mmHg) for three consecutive times, it is regarded as abnormal data and automatically discarded. If the sampling error is greater than the set threshold for three consecutive times, the data is discarded and resampled. The abnormal data is marked and stored in the internal storage of the smart ring for subsequent trend analysis. The average blood pressure is calculated once an hour, and a blood pressure curve for the whole day is generated for the user to view. If the measurement data shows that the blood pressure is continuously higher or lower than the user's baseline blood pressure, the smart ring will remind the user and prompt whether a more accurate measurement is needed (for example, using a cuff blood pressure monitor).

[0178] The PPG signal is sampled at high frequency within a short period of time to calculate blood pressure and improve measurement accuracy. This is manually triggered by the user and lasts for 2-3 minutes. Within 2 minutes, multiple PPG cycles are recorded at 5-10 times per second. PTT and HR are calculated, and a dynamic filtering algorithm is used to remove noise. Within 3 minutes, multiple PTT and HR averages are combined to improve accuracy. PTT is calculated over different time windows, and data from stable time periods is used for regression analysis. The deviation between the blood pressure predicted by the smart ring and the blood pressure cuff is calculated. If the error is greater than 5 mmHg, the regression model is updated to improve prediction accuracy.

[0179] Sleep monitoring mode analyzes nighttime blood pressure fluctuations and assesses autonomic nervous system activity. Blood pressure is sampled every two minutes during the two hours before bedtime (pre-sleep preparation period) to analyze blood pressure fluctuations during sleep. During deep sleep, the sampling frequency is reduced to every 10-15 minutes to reduce energy consumption. In the morning, between 5 and 7 a.m., the sampling frequency is increased to every five minutes. If blood pressure fluctuates abnormally during the night (>10 mmHg), a health reminder will be issued in the morning.

[0180] This embodiment also provides a computer device, which is suitable for the blood pressure measurement system based on the PPG pulse wave principle of a smart ring, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the blood pressure measurement system based on the PPG pulse wave principle of the smart ring proposed in the above embodiment.

[0181] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0182] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, a blood pressure measurement system based on the PPG pulse wave principle of a smart ring as proposed in the above embodiment is implemented.

[0183] In summary, the present invention adopts a PPG optical sensor to achieve non-invasive measurement by optically detecting changes in blood flow, thus avoiding the discomfort caused by traditional cuff inflation. Through low-power mode, blood pressure changes can be monitored around the clock, and dynamic blood pressure tracking without interference can be achieved. It is suitable for hypertensive patients, pregnant women, the elderly and people at high risk of cardiovascular disease who need long-term blood pressure monitoring. Traditional smart watches and bracelets are worn on the wrist, and the PPG signal is easily affected by hand movements, while smart rings are worn on the fingers, with rich blood flow at the end, and the PPG signal is clearer and more stable, and the measurement accuracy is higher. Through the multi-sensor signal quality adaptive selection algorithm, the PPG signal quality of different finger parts is automatically detected, and the index finger with the best signal is selected for measurement to improve accuracy.

[0184] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A blood pressure measurement system based on the PPG pulse wave principle of a smart ring, characterized by: include: The position layout module is used to integrate a dual-wavelength LED light source and a photoelectric sensor on the inner surface of the smart ring. The module uses an adaptive signal quality selection algorithm based on multiple photoelectric sensors to detect the PPG signal quality of different fingers and selects the index finger as the measurement location. The signal acquisition and preprocessing module is used to acquire PPG waveforms through a high-sampling-rate ADC, measure blood oxygen saturation using a dual-wavelength light source to compensate for optical scattering, remove motion artifacts through adaptive Kalman filtering, and remove high-frequency noise in combination with wavelet transform; The calculation and modeling module is used to extract PPG feature points through signal processing algorithms, including peaks and troughs, rising edge slopes, and time centers, calculate pulse wave transit time, and then use linear regression models to predict blood pressure; The error calibration module is used to optimize blood pressure predictions by combining historical data and constructing a personalized blood pressure calibration curve based on the user's cuff blood pressure monitor measurement data to improve measurement accuracy; The blood pressure measurement module is used in continuous monitoring mode, with low power consumption and sampling every 10 seconds. In rapid measurement mode, it collects high-density data within 2 minutes, calculates blood pressure and provides immediate feedback. In sleep monitoring mode, it combines PPG and HRV to analyze nighttime blood pressure fluctuations and assess sleep health.

2. The blood pressure measurement system based on the PPG pulse wave principle of the smart ring according to claim 1, characterized in that: The method of detecting the PPG signal quality of different finger positions using a signal quality adaptive selection algorithm of multiple photoelectric sensors and selecting the index finger as the measurement position includes: The raw PPG signal collected by each photosensor goes through the following steps: A high-pass filter was used to remove DC drift, and a 5 Hz low-pass filter was used to remove high-frequency noise. The amplitude of each channel signal was then normalized. The PPG signal quality is scored in combination with the signal-to-noise ratio, and the signal-to-noise ratio formula is as follows: Among them, P signal is the power of the main frequency band of the PPG signal, P noise It is the power of the non-pulse wave frequency band. Set the threshold to ensure that the SNR is greater than 20dB.

3. The blood pressure measurement system based on the PPG pulse wave principle of a smart ring as claimed in claim 1, characterized in that: The method of detecting the PPG signal quality of different finger positions by using a signal quality adaptive selection algorithm of multiple photoelectric sensors and selecting the index finger as the measurement position also includes: Calculate the heart rate HR for each channel calc And estimate the baseline value HR with heart rate ref Compare: ΔHR=|HR calc -HR ref | Set the threshold to ensure the error ΔHR < 3bpm; Then calculate the peak-valley detection consistency of 10 consecutive PPG cycles to ensure that the peak position deviation of adjacent cycles is less than 5%; then calculate the comprehensive quality score of all measurement points. The comprehensive quality score calculation formula is as follows: Q=w1·SNR+w2·CC+w3·(1-ΔHR) Among them, w1, w2, w3 are weights, SNR, CC, ΔHR are normalized to between 0 and 1, the measurement position with the highest score is selected, and the threshold Q is set threshold , make sure the selected signal is of good quality.

4. The blood pressure measurement system based on the PPG pulse wave principle of a smart ring as claimed in claim 1, characterized in that: The method of removing motion artifacts by adaptive Kalman filtering and removing high-frequency noise by combining wavelet transform includes: A state space model of the PPG signal is established, and the expression of the state space model of the PPG signal is: x k =Ax k-1 +w k y k =Hx k +v k Among them, x k is the real PPG state variable, A is the state transfer matrix, wk is the process noise, yk is the PPG value measured by the sensor, H is the observation matrix, and vk is the measurement noise; An adaptive adjustment strategy is used to estimate the signal change rate in real time. If the signal changes significantly, R is increased to reduce the filter's trust in the measured value. Motion status is detected, and high-frequency energy changes in accelerometer or PPG signals are used to dynamically adjust Q and R.

5. The blood pressure measurement system based on the PPG pulse wave principle of the smart ring as claimed in claim 1, characterized in that: The method of removing motion artifacts by adaptive Kalman filtering and removing high-frequency noise by combining wavelet transform also includes: Perform discrete wavelet transform on the PPG signal x(t): Among them, c j is the approximate coefficient, retaining PPG information, d j is the detail coefficient, which contains noise information; Then the detail coefficient d j Perform soft threshold denoising: Where λ is the threshold, σ is the noise standard deviation, and n is the data length; The processed c j and The inverse wavelet transform is performed to obtain the denoised PPG signal, and the processed signal is further optimized by sliding mean filtering.

6. The blood pressure measurement system based on the PPG pulse wave principle of a smart ring as claimed in claim 1, characterized in that: The method of extracting PPG feature points through signal processing algorithm, calculating pulse wave transit time, and then predicting blood pressure using linear regression model includes: In the PPG waveform, the characteristic points include the peak value, which represents the peak of cardiac contraction, the valley value, which represents the state of cardiac relaxation, the maximum point of the first-order derivative, which represents the moment when the blood flow velocity reaches the maximum, and the inflection point of the second-order derivative, which represents the compliance of the blood vessels; Compute the first-order derivative by differencing: The peak position of the first derivative corresponds to the time point when the blood flows fastest during the heart contraction; Then calculate the second derivative of the PPG signal: The inflection point of the second-order derivative is used to evaluate vascular elasticity and assist in calculating pulse wave transmission time. The adaptive threshold method was used to detect the systolic peak and diastolic trough of the PPG waveform. Then calculate the pulse wave transit time: PTT=T SP -T dPPGmax Among them, T SP is the peak time of the PPG signal, T dPPGmax is the time of maximum value of the first-order derivative of PPG; Through the linear regression model, systolic blood pressure (SBP) and diastolic blood pressure (DBP) were calculated based on PTT. The relationship between PTT and blood pressure satisfies the following: BP=A+B·PTT+C·HR Where BP is blood pressure, A, B, C are regression coefficients to be fitted, PTT is pulse wave transit time, and HR is an additional influencing factor; Calculate PTT in real time and input PTT and HR into a linear model: SBP=A1+B1·PTT+C1·HR DBP=A2+B2·PTT+C2·HR Systolic and diastolic blood pressures were calculated.

7. The blood pressure measurement system based on the PPG pulse wave principle of a smart ring as claimed in claim 1, characterized in that: The method of optimizing blood pressure prediction by combining historical data and constructing a personalized blood pressure calibration curve based on the user's blood pressure cuff measurement data to improve measurement accuracy includes: The system records the measurement time, systolic and diastolic blood pressure, as well as the pulse wave transmission time and heart rate collected by the smart ring. It performs multiple measurements to ensure a balanced distribution of data at different times and under different conditions. The data is stored in the user's personal database, correlating measurements at different times with the user's age, gender, weight, and vascular condition information to form a personalized health profile. The error between the blood pressure predicted by the smart ring and the actual blood pressure measured by the blood pressure cuff is calculated, and the deviation trend is observed. The measurement data for multiple time periods is compiled to analyze whether the error changes over time and with blood pressure levels. Based on historical data, the smart ring generates a blood pressure calibration curve for the user. The curve reflects the systematic deviation between the blood pressure predicted by the smart ring and the actual blood pressure, and the prediction result is then adjusted. If the smart ring is found to systematically underestimate blood pressure within the hypertension range, the predicted value in this range will be adjusted upward in future measurements to ensure that it is closer to the actual blood pressure.

8. The blood pressure measurement system based on the PPG pulse wave principle of a smart ring as claimed in claim 1, characterized in that: In the continuous monitoring mode, low power consumption is used to sample blood pressure once every 10 seconds; in the rapid measurement mode, high-density data is collected within 2 minutes, blood pressure is calculated, and immediate feedback is provided. In sleep monitoring mode, PPG and HRV are combined to analyze nighttime blood pressure fluctuations and assess sleep health, including: In continuous measurement mode, intelligent data filtering is used. If the sampling error is large for three consecutive times, the abnormal data will be automatically discarded and uploaded to the smart ring for internal storage. Trend analysis will be performed every hour. The average blood pressure value is calculated every hour and a blood pressure curve for the whole day is drawn for the user to view. If the blood pressure is found to be continuously higher or lower than the user's baseline blood pressure, the smart ring will remind the user and ask whether a more accurate measurement is needed. In the fast measurement mode, the user manually triggers this mode. The smart ring samples the PPG signal at a high frequency within 2 minutes, combines the pulse wave transmission time and heart rate data to calculate the blood pressure. Within 3 minutes, the sampling frequency is increased to 5-10 times per second to ensure measurement accuracy. In sleep monitoring mode, samples are taken every 2 minutes for 2 hours before falling asleep to analyze the changes in the user's blood pressure during sleep; the sampling frequency is reduced to once every 10-15 minutes during deep sleep to reduce energy consumption; the sampling frequency is increased to once every 5 minutes from 5 to 7 in the morning; vascular tension is analyzed through PPG signals, and night-time blood pressure changes are calculated in combination with PTT, sympathetic and parasympathetic nerve activity are calculated, the night-time autonomic nervous system regulation ability is analyzed, and whether there are abnormal night-time blood pressure fluctuations is evaluated; if abnormal night-time blood pressure is found to be abnormally high, the smart ring will provide a health reminder in the morning; after waking up in the morning, the user will view the night-time blood pressure report through the mobile phone APP, including the night-time average blood pressure curve, blood pressure fluctuation trend and autonomic nervous system activity assessment. If an abnormal blood pressure pattern is detected, the system will provide targeted suggestions.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the blood pressure measurement system based on the PPG pulse wave principle of the smart ring according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the blood pressure measurement system based on the PPG pulse wave principle of a smart ring according to any one of claims 1 to 7 are implemented.

Images