Method for monitoring and evaluating left ventricular outflow tract obstruction based on non-invasive pulse-by-pulse blood pressure

Abstract

The invention provides a left ventricular outflow tract obstruction monitoring and evaluating method based on non-invasive pulse-by-pulse blood pressure. The method comprises the following steps that the pulse-by-pulse systolic pressure and the left ventricular outflow tract pressure difference of a patient in a training set under the resting state and the Valsalva action are obtained; determining a first change value and a second change value according to the difference between the pace-by-pace systolic pressure and the left ventricular outflow tract pressure of the patient in the resting state and the Valsalva action; determining a linear quantitative relationship between the first change value and the second change value, and determining an optimal diagnosis threshold value for distinguishing obstruction and non-obstruction; acquiring the beat-to-beat systolic pressure of a patient to be monitored, and determining whether the patient has left ventricular outflow tract obstruction or not according to the optimal diagnosis threshold value; the method has the advantages that through four core technologies of high-resolution pulse-by-pulse blood pressure, standardized Valsalva excitation, millisecond-level synchronous triggering and a linear coupling algorithm, complex excitation gradient measurement is converted into instantaneous, non-invasive and remote digital indexes for the first time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of left ventricular outflow tract obstruction monitoring and evaluation, in particular to a left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure. BACKGROUND

[0002] Hypertrophic cardiomyopathy (HCM) is a serious heart disease that threatens human health. In particular, in the case of HCM accompanied by left ventricular outflow tract obstruction (LVOTO), the risk of serious adverse events such as sudden death, heart failure, and malignant arrhythmia in patients significantly increases, significantly affecting the prognosis of patients with the disease. Asymmetric hypertrophy of the myocardium, systolic anterior motion (SAM) of the mitral valve, and abnormal hypercontractility of the myocardium are the main pathophysiological abnormalities of hypertrophic cardiomyopathy and the main determinants of dynamic obstruction of the left ventricular outflow tract (LVOT). Studies have shown that the global prevalence of HCM is about 1%, and about one-third of HCM patients have resting outflow tract obstruction, and another one-third of HCM patients have occult LVOTO, i.e., LVOTO occurs when the cardiac load increases and the myocardial contractility strengthens. LVOTO significantly affects cardiac function and patient prognosis by hindering left ventricular ejection.

[0003] Currently, cardiac ultrasound Doppler measurement of left ventricular outflow tract pressure gradient (LVOTG) is the recognized "gold standard" method for evaluating LVOTO and is widely used in clinical practice. However, there are several problems with the cardiac ultrasound Doppler method for evaluating LVOTG: 1. The need for cardiac ultrasound hardware and software, high facility cost; 2. The need for long-term training of experienced professional cardiac ultrasound physicians for related index determination, high technical threshold; 3. The potential for a large clinical demand, while the medical resources for cardiac ultrasound are relatively scarce, and the waiting time for examination is long, with high labor and time costs; 4. The accuracy of the examination depends on the technical level and experience of the cardiac ultrasound physician, with potential errors and misdiagnosis risks. In summary, although the cardiac ultrasound Doppler method is accurate and feasible, its high technical threshold, labor cost, and economic cost limit its application, especially in areas with relatively scarce medical resources, making it difficult to ensure the timeliness and accuracy of cardiac ultrasound Doppler examination. Therefore, there is an urgent need for an accurate, convenient, efficient, and economical detection method to evaluate the LVOTO of HCM patients.

[0004] Systolic Blood Pressure (SBP) is a key indicator of cardiac afterload and systemic hemodynamic status. The correlation between SBP and Left Ventricular Outflow Tract Gradient (LVOTG) has not been systematically studied, especially in dynamic changes and clinical classification. Heart mechanical work-related parameters such as intraventricular pressure and cardiac output are correlated with SBP, which is the product of heart work. Previous studies have shown that HCM patients have myocardial changes, including hypertrophy, disordered arrangement, and interstitial fibrosis, which in turn lead to impaired diastolic function and elevated Left Ventricular Filling Pressure (LVFP). However, the relationship between LVOTG and SBP is not fully understood. As mentioned earlier, current clinical assessment of LVOTG relies mainly on echocardiography, which is complex, dependent on physician experience, and difficult to achieve continuous monitoring. If non-invasive blood pressure waveform analysis can be used to assist in determining LVOTG status, it will help to identify high-risk patients early, optimize treatment decisions (such as drug adjustment, surgical timing selection, etc.), and is particularly suitable for dynamic monitoring during follow-up. In addition, continuous non-invasive SBP monitoring is easy to integrate with wearable devices, achieving the goal of dynamic, real-time, and all-weather monitoring.

[0005] Current methods for evaluating LVOTO and their limitations:

[0006] 1. Diagnostic gold standard: LVOTG measured by Doppler echocardiography: 1) Requires a 1.7 MHz continuous wave (CW) probe and a high-end color ultrasound instrument with load ultrasound software, with a single machine price of 2-3 million yuan, making it difficult to popularize in primary hospitals. 2) The examination is completely dependent on experienced cardiac ultrasound physicians: The subjective factors such as technique angle, sampling line position, and Valsalva maneuver quality are large, and the measured values by different physicians for the same patient can differ by 15-20 mmHg, with limited repeatability. 3) Long process: reservation → load → offline measurement → report review, with an average time consumption of 45-60 minutes; during peak clinic hours or follow-up periods, the waiting time is often >2 weeks, making it easy to miss the optimal treatment window. 4) Cannot be continuously / remotely monitored: echocardiography cannot be worn for 24 hours, and patients need to repeatedly go back and forth to the hospital, with poor compliance; new drug (mavacamten) titration requires monthly review, doubling the cost of manpower and time. 5) High cost: one-time echocardiography charge 400-600 yuan, annual follow-up examination nearly 10,000 yuan, heavy burden on medical insurance and individuals.

[0007] 2. Invasive and image replacement means: 1) Cardiac catheter pressure measurement: invasive, radiation, high cost, only used for preoperative evaluation, and cannot be used repeatedly. 2) Cardiac MRI: can measure peak flow rate, but the excitation sequence is complex, the scanning time is long, the respiratory coordination requirement is high, and the body metal is contraindicated, and the popularization rate is much lower than that of ultrasound. 3) Traditional cuff blood pressure: single point, 30s interval, 1mmHg resolution, cannot capture the 1-2s instantaneous pressure drop in Valsalva II phase; and the resolution is only 1mmHg, the repeatability is poor, and there is no quantitative model with LVOTG. 4) Existing non-invasive hemodynamic monitoring (NICOM, ClearSight) focuses on cardiac output, and no quantitative conversion model with LVOTG has been established, and there is also a lack of standardized excitation protocol.

[0008] 3. Special needs of occult LVOTO: About 1 / 3 of the global hypertrophic cardiomyopathy (HCM) patients are "occult obstruction", the resting gradient is normal, and only under Valsalva or drug excitation. Guidelines recommend rechecking excitation ultrasound every 3-6 months, but the above bottlenecks make the actual follow-up rate <50%, and a large number of high-risk patients are missed or delayed intervention.

[0009] In summary, the prior art of the present application is faced with the five major pain points of "expensive equipment, personnel shortage, long process, non-continuous, high cost", so there is an urgent need in the clinic for a digital replacement scheme that is instant, non-invasive, low cost, repeatable and highly coupled with ultrasound gradient. SUMMARY

[0010] In view of the above-mentioned shortcomings of the prior art, the purpose of the present application is to provide a left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure, which solves the five major problems of "expensive equipment, personnel shortage, long process, non-continuous, high cost" in the prior art.

[0011] To achieve the above-mentioned purposes and other related purposes, the present application provides the following technical scheme:

[0012] A left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure, the method comprising the following steps: obtaining the relevant information of a hypertrophic cardiomyopathy patient, and dividing the hypertrophic cardiomyopathy patient into a training set and a validation set according to the relevant information; obtaining the beat-by-beat systolic pressure of the patient in the training set under the resting state and Valsalva action recorded by a finger cuff blood pressure instrument in succession, and simultaneously obtaining the left ventricular outflow tract pressure difference under the resting state and after Valsalva excitation measured by continuous wave Doppler ultrasound; determining a first change value and a second change value according to the beat-by-beat systolic pressure of the patient under the resting state and Valsalva action and the left ventricular outflow tract pressure difference under the resting state and after Valsalva excitation;

[0013] determining a linear quantitative relationship between the first change value and the second change value, and determining an optimal diagnostic threshold for distinguishing obstruction from non-obstruction through ROC analysis and Youden index under the linear quantitative relationship; obtaining beat-to-beat systolic pressure of the patient to be monitored collected by the flexible MEMS pressure sensor array, and determining whether the patient has left ventricular outflow tract obstruction according to the optimal diagnostic threshold.

[0014] In an embodiment of the present application, the first change value and the second change value are determined according to the beat-to-beat systolic pressure and the left ventricular outflow tract pressure difference in the resting state and after Valsalva action, comprising: determining the difference between the average value of the beat-to-beat systolic pressure in the continuous N cardiac cycles after Valsalva action and the average value of the beat-to-beat systolic pressure in the continuous N cardiac cycles in the resting state, and taking it as the first change value; determining the difference between the peak value of the left ventricular outflow tract pressure difference after Valsalva action and the peak value of the left ventricular outflow tract pressure difference in the resting state, and taking it as the second change value.

[0015] In an embodiment of the present application, the difference between the peak value of the left ventricular outflow tract pressure difference after Valsalva action and the peak value of the left ventricular outflow tract pressure difference in the resting state is determined, and taken as the second change value, comprising: determining the difference between the peak value of the left ventricular outflow tract pressure difference after Valsalva action and the peak value of the left ventricular outflow tract pressure difference in the resting state according to the following formula: ; wherein, is the second change value, that is, the change value of the left ventricular outflow tract pressure difference; is the peak flow rate of the left ventricular outflow tract pressure difference in the resting state; is the peak flow rate of the left ventricular outflow tract pressure difference after Valsalva action.

[0016] In an embodiment of the present application, the linear quantitative relationship between the first change value and the second change value is determined, comprising: establishing the linear quantitative relationship between the first change value and the second change value through Pearson correlation analysis.

[0017] In an embodiment of the present application, the linear quantitative relationship between the first change value and the second change value is established through Pearson correlation analysis, comprising: performing Pearson correlation analysis on the first change value and the second change value of the patient in the training set, and calculating a correlation coefficient, wherein it is known from the correlation coefficient that there is a strong positive linear correlation between the first change value and the second change value; and establishing a linear regression model according to the correlation coefficient, wherein the linear quantitative relationship between the first change value and the second change value is obtained by fitting a linear equation through the least square method in establishing the linear regression model.

[0018] In one embodiment of the present invention, obtaining the linear quantitative relationship between the first change value and the second change value by fitting a linear equation using the least squares method in establishing the linear regression model includes: obtaining the linear quantitative relationship between the first change value and the second change value according to the following formula: ;in, The slope A slope of 1.066 indicates that per unit decline( Increase), The average increase was 1.066 mmHg. This is the first change value, and also the change in stroke-by-stroke systolic blood pressure after the Valsalva maneuver; It is a constant. The value is 3.87.

[0019] In one embodiment of the present invention, the determination of the optimal diagnostic threshold for distinguishing between obstruction and non-obstruction by ROC analysis and Youden index under the influence of the linear quantitative relationship includes: using the first change value of the patients in the training set as a test variable and the second change value as a state variable to plot an ROC curve, wherein the area under the ROC curve is used to evaluate the diagnostic accuracy.

[0020] The Youden index is defined as sensitivity + specificity - 1, and the threshold corresponding to the largest Youden index is the optimal diagnostic threshold. The sensitivity and specificity of the threshold corresponding to each of the first change values ​​are calculated, and the optimal diagnostic threshold for distinguishing between obstruction and non-obstruction is obtained based on the sensitivity and specificity.

[0021] In one embodiment of the present invention, after obtaining the linear quantitative relationship between the first change value and the second change value and the optimal diagnostic threshold for distinguishing between obstruction and non-obstruction, the validation set is used for extrapolation validation.

[0022] As described above, the method for monitoring and assessing left ventricular outflow tract obstruction based on non-invasive pulse-time blood pressure of the present invention has the following beneficial effects: The present invention, through four core technologies of "high-resolution pulse-time blood pressure + standardized Valsalva excitation + millisecond-level synchronous triggering + linear coupling algorithm", transforms the complex excitation gradient measurement into an instantaneous, non-invasive, and remote digital indicator for the first time. It also has dual functions of diagnosis and drug replacement endpoint, and has low hardware cost, zero operation threshold, and standardized quality control. It can be seamlessly integrated into existing clinical and remote follow-up processes. Attached Figure Description

[0023] Figure 1 This is an overall flowchart of the method for monitoring and assessing left ventricular outflow tract obstruction based on non-invasive pulse-fed blood pressure disclosed in this embodiment of the invention.

[0024] Figure 2 A schematic diagram of the overall flow chart of the left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure disclosed in the embodiments of the present application is shown.

[0025] Figure 3 A curve graph of the beat-by-beat systolic pressure difference and the left ventricular outflow tract pressure difference of the patient synchronously collected in the left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure disclosed in the embodiments of the present application under the resting state and Valsalva action is shown.

[0026] Figure 4 A linear quantitative relationship graph between the change value of the beat-by-beat systolic pressure and the change value of the left ventricular outflow tract pressure difference after the Valsalva action in the left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure disclosed in the embodiments of the present application is shown.

[0027] Figure 5 An ROC diagnostic performance graph in the left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure disclosed in the embodiments of the present application is shown.

[0028] Figure 6 An independent predictive value graph corresponding to the best diagnostic threshold in the left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure disclosed in the embodiments of the present application is shown.

[0029] Figure 7 A subgroup stability graph corresponding to the best diagnostic threshold in the left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure disclosed in the embodiments of the present application is shown. DETAILED DESCRIPTION

[0030] The embodiments of the present application are described below through specific concrete examples, and other advantages and effects of the present application can be easily understood by those skilled in the art from the disclosure. It should be noted that the following examples and features in the examples can be combined with each other without conflict.

[0031] The present application relates to a left ventricular outflow tract obstruction monitoring and evaluation method based on non-invasive beat-by-beat blood pressure, the flow is as shown in Figure 1 , and specifically as follows:

[0032] Step 101, obtaining the related information of the hypertrophic cardiomyopathy patient, and dividing the hypertrophic cardiomyopathy patient into a training set and a validation set according to the related information.

[0033] Specifically, the training set includes 173 patients with hypertrophic cardiomyopathy (HCM), of which 49 patients are obstructive HCM (oHCM) and 124 patients are non-obstructive HCM (nHCM), and the validation set includes 161 patients with hypertrophic cardiomyopathy (HCM).

[0034] Step 102, acquiring the beat-by-beat systolic pressure of the patients in the training set recorded by the finger cuff blood pressure meter under resting state and Valsalva action, and acquiring the left ventricular outflow tract pressure difference under resting state and after Valsalva excitation by continuous wave Doppler ultrasound.

[0035] Specifically, the beat-by-beat systolic pressure (SBP) of the hypertrophic cardiomyopathy (HCM) patients in the training set is recorded by the finger cuff Finapres NOVA device (200Hz sampling) and the left ventricular outflow tract pressure difference (LVOTG) is measured by continuous wave Doppler ultrasound (180-220fps) simultaneously; the data collection includes the values under resting state and Valsalva action (maintaining 40mmHg pressure for 10-15 seconds), please refer to Figure 3 .

[0036] Step 103, determining the first change value and the second change value according to the beat-by-beat systolic pressure of the patient under resting state and Valsalva action and the left ventricular outflow tract pressure difference under resting state and after Valsalva excitation.

[0037] Specifically, Calculation of (change value of beat-by-beat systolic pressure after Valsalva action): take the average value of SBP of 5 consecutive cardiac cycles after Valsalva action, subtract the average value of SBP of 5 consecutive cardiac cycles under resting state, and thus obtain Since Valsalva action usually causes SBP to decrease, is negative, but the absolute value or negative value is often used in this paper (such as );

[0038] Calculation of (change value of left ventricular outflow tract pressure difference): take the peak value of LVOTG after Valsalva action minus the peak value of LVOTG under resting state, using the simplified Bernoulli equation: , wherein, is the peak flow velocity of left ventricular outflow tract pressure difference under resting state; is the peak flow velocity of left ventricular outflow tract pressure difference after Valsalva action.

[0039] Step 104, determining a linear quantitative relationship between the first change value and the second change value, and determining the optimal diagnostic threshold for distinguishing obstruction from non-obstruction by ROC analysis and Youden index under the action of the linear quantitative relationship.

[0040] Specifically, the linear quantitative relationship between the first change value and the second change value is established by Pearson correlation analysis, and includes the following steps: 1) Pearson correlation analysis of the values in the training set (173 patients) and calculation of the correlation coefficient The document reports (95% confidence interval 0.875-0.930, P<0.0001), indicating between them is a strong positive linear correlation; 2) Establish a linear regression model: by least squares fitting of a linear equation, we get , where the slope 1.066 represents a decrease (increase) of per unit , an average increase of 1.066 mmHg, and the intercept -3.87 is a constant term, and the determination coefficient , indicating that the model can explain 82% of the variation, please refer to Figure 4 ;

[0041] There is also a model validation after obtaining the linear quantitative relationship between the first change value and the second change value: extrapolation validation using the validation set (161 patients), residual analysis shows normal distribution and no heteroscedasticity, confirming the robustness of the model (validation set , AUC=0.93), the core of this process is to quantify the linear relationship between and by statistical correlation and regression analysis, thereby allowing to be used as a non-invasive surrogate indicator of .

[0042] More specifically, determining the optimal diagnostic threshold for distinguishing obstruction from non-obstruction by ROC analysis and Youden index under the action of the linear quantitative relationship includes the following steps:

[0043] 1) ROC curve drawing: using the training set data (173 patients), taking as the test variable, and taking the gold standard ​​​(Or clinical diagnosis oHCM vs nHCM) as a state variable, ROC curve is drawn, and the area under the ROC curve (AUC) is used to evaluate the diagnostic accuracy. In this paper, AUC=0.9919 (95% confidence interval 0.9836~1.000) is reported, indicating There is a very high discriminant ability for oHCM. For details, please refer to Figure 5 ;

[0044] 2) Youden index calculation: Youden index (J) is defined as sensitivity (Sensitivity) + specificity (Specificity) -1, and the threshold value corresponding to the maximum Youden index is the optimal diagnostic threshold value, because it balances sensitivity and specificity. The sensitivity and specificity of each possible threshold value are calculated: sensitivity: true positive rate (oHCM patients below the threshold value); specificity: true negative rate (nHCM patients above the threshold value), in this paper, when the threshold value is set to (i.e. decrease by 13 mmHg or more), the Youden index is maximum; therefore, the actual threshold value is decrease by 13 mmHg or more (i.e. or );

[0045] Threshold determination: in the training set, , the sensitivity is 100% (95% confidence interval 92.7~100%), the specificity is 95.2% (95% confidence interval 89.8~97.8%), and the Youden index is 0.952 (i.e. 1.00+0.952-1=0.952); the positive likelihood ratio is 20.7, further confirming the diagnostic value of the threshold value. In the independent validation set (161 patients), the threshold value maintains high diagnostic performance (AUC=0.93, sensitivity 100%, specificity 95.2%), indicating that the threshold value is robust, and this process ensures The threshold value has the optimal sensitivity and specificity in distinguishing oHCM and nHCM, and is suitable for clinical screening;

[0046] Please also refer to Figure 6 and Figure 7 In Figure 6 , multivariate logistic regression shows that only is the strongest independent predictor of oHCM (OR significantly>1, CI does not contain 1); age, hypertension, ACEI / ARB / ARNI, calcium channel blockers, blockers, family history and gender are not statistically significant; in Figure 7In the absence of LVOT obstruction, the association between the presence of LVOT obstruction and the presence of oHCM was consistent, regardless of whether ACEI / ARB / ARNI, calcium antagonists or blockers were used, whether hypertension or family history was present, and whether the patient was >60 years or ≤60 years, male or female, The association with oHCM was consistent throughout, suggesting that this index is not affected by clinical background, drug treatment and demographic characteristics, and is highly generalizable.

[0047] Step 105, acquiring the beat-by-beat systolic pressure of the patient to be monitored collected by the flexible MEMS pressure sensor array, and confirming whether the patient has left ventricular outflow tract obstruction according to the optimal diagnostic threshold.

[0048] It needs to be explained that the "beat-by-beat blood pressure-excitation coupling" full-link solution is:

[0049] 1. Signal acquisition layer: 1) Dual-channel synchronous hardware: Finapres NOVA: 200Hz beat-by-beat SBP / DBP / MAP, resolution 0.1mmHg, delay <5ms; Continuous wave Doppler 1.7MHz probe: 180-220fps peak flow velocity curve; Three-lead ECG 2kHz trigger; Self-made Arduino photoelectric trigger box: 1ms pulse on R-wave rising edge, write blood pressure and ultrasound time stamp simultaneously, eliminate phase drift; 2) Airway standardization excitation device: medical PVC mouthpiece + mechanical pressure gauge (0-100mmHg, accuracy ±2mmHg); 40mmHg standard line visualization, patients maintain for 10-15s, ensure that the phase II pressure drop is repeatable;

[0050] 2. Algorithm feature layer: 1) Automatic phase II identification: 5-point moving average on 200Hz SBP sequence → first-order difference >15mmHg / s for ≥3 beats → mark phase II trough; Take the difference between the average of 5 beats before and after the trough and the average of 5 beats at rest ; 2) Ultrasound side flow velocity processing: same cycle R-wave gated ensemble average ≥10 beats; Peak flow velocity automatic picking (findpeaks), simplified Bernoulli: ; 3) Linear coupling model: training set of 173 cases to establish: , ; validation set of 161 cases to extrapolate, residual normal, no heteroscedasticity, confirm the robustness of the model;

[0051] 3. Decision threshold layer: ROC AUC 0.991, Youden maximum point : sensitivity 100% (95% CI 92.7~100), specificity 95.2% (89.8~97.8), positive likelihood ratio 20.7; Multivariate logistic verification: OR is independent and significant, traditional confounding factors (age, all the rest (e.g. blockers, hypertension, etc.) are ruled out;

[0052] 4. Efficacy monitoring extension layer: 1) Drug response definition: SBP response: ≥13mmHg decrease from baseline without symptomatic hypotension; echocardiographic response: resting LVOTG<30mmHg + Valsalva LVOTG<30mmHg + LVEF>50%; Kaplan-Meier shows that SBP response curve almost overlaps with the composite response of echocardiography, with a median of 4 weeks earlier to meet the standard; 2) Wearable implant: MEMS flexible finger cuff 24h endurance, nRF52840 BLE broadcast , real-time alarm on mobile phone <-13mmHg event; 50 cases comparison: difference of 0.6±3.2mmHg with standard Finapres, ICC=0.93, meet ISO 81060-2;

[0053] 5. Quality control features: every case of pre-Allen test + necessary 8MHz handheld Doppler to exclude upper limb arterial occlusion; double-blind operation: blood pressure collector does not know the ultrasound results, and the ultrasound doctor does not know the blood pressure curve; abnormal waveform is automatically excluded: PVC>5%, signal loss>1s are re-sampled to ensure effective period ≥10 beats;

[0054] 6. Output and interface features: one-key generation of structured report: , , ROC confidence, response state; HL7 FHIR format is automatically written into the hospital electronic medical record, supporting CDSS to trigger subsequent referral / drug titration; RESTful API is provided to allow remote follow-up platform to pull historical trends by patient ID, please refer to Figure 2 .

[0055] It should be further pointed out that the method further comprises the following steps:

[0056] 1. Acquisition step: record the beat-by-beat systolic blood pressure (SBP) of the patient at rest and under Valsalva action with a finger cuff Finopres, sampling frequency 200Hz; simultaneously measure the left ventricular outflow tract pressure difference (LVOTG) at rest and after provocation with continuous wave Doppler ultrasound, frame frequency 180-220fps, and superimpose a 2kHz ECG trigger time marker to realize R-wave gated ensemble averaging of ≥10 cardiac cycles;

[0057] 2. Calculation step: subtract the average value at rest from the average value of the continuous 5 SBP after Valsalva action to obtain ; subtract the resting LVOTG from the provoked LVOTG to obtain ; establish and linear quantitative relationship between the two, and the maximum Youden index is determined The area under the ROC curve is greater than or equal to 0.90 for the best diagnostic threshold to distinguish oHCM from non-obstructive HCM (nHCM);

[0058] 3. Efficacy monitoring step: for oHCM patients receiving mavacamten treatment, the first day of taking medicine is zero, steps 1 and 2 are repeated in the follow-up period of ≥12 weeks, if If the SBP decreases by ≥13 mmHg and there is no symptomatic hypotension, it is determined that the SBP responds; if both the resting and Valsalva LVOTG are <30 mmHg and the LVEF is >50%, it is determined that the hemodynamic composite responds; the cumulative incidence of SBP response and LVOTG response is compared by the Kaplan-Meier method to achieve non-invasive, real-time and repeatable evaluation of the efficacy of mavacamten;

[0059] 4. Quality control step: before measurement, the patency of the upper limb artery is evaluated by physical examination combined with Allen test and Doppler ultrasound if necessary to exclude arterial obstruction; the entire process is independently analyzed by blind method to ensure that NIBBP and ultrasound data do not interfere with each other, thereby overcoming the limitations of traditional ultrasound on operator dependence and accessibility.

[0060] Further, the present application is based on the following principles: Valsalva action instantaneously remodels hemodynamics through the "chest-circulation coupling" mechanism, and its pathological amplified version provides a quantifiable diagnostic window for hypertrophic obstructive cardiomyopathy (oHCM); action initiation, forced closed glottis expiration, pleural pressure rises by about 20-40 mmHg, chest organs are subjected to "external pressure clamping", transmural pressure of the heart and large vessels decreases sharply, venous collapse, right atrial pressure rises sharply, venous return is "externally blocked"; although the absolute pressure in the heart chamber increases, the effective filling (preload) decreases sharply, the Frank-Starling mechanism drives the stroke volume to decrease sharply, the aortic pressure first appears a false peak due to external pressure on large vessels (phase I), and then rapidly falls (phase II) after the output fails; pressure drop triggers baroreceptor reflex, heart rate compensates for the surge, forming a "pressure drop-heart rate acceleration" mirror reaction; the moment of exhalation recovery, the external pressure is removed, the aortic pressure is temporarily "vacuum" down (phase III), and then the return blood volume suddenly fills, the stroke volume rebounds, accompanied by the increase of systemic vascular resistance after sympathetic excitation, and the blood pressure appears a super-peak (phase IV), the heart rate slows down accordingly, and a typical W-shaped curve is completed;

[0061] This study establishes "Valsalva action under each beat Real-time monitoring is a widely applicable dynamic indicator used to differentiate between obstructive hypertrophic cardiomyopathy (oHCM) and non-obstructive hypertrophic cardiomyopathy (nHCM). A training set of 173 cases showed that the area under the ROC curve (AUC) for diagnosing oHCM with a Valsalva-induced systolic blood pressure drop of <–13 mmHg was 0.99. This performance remained at AUC 0.93 in an independent validation cohort of 161 cases, unaffected by clinical subgroups. High-resolution synchronous recording confirmed this. With Doppler It exhibits a near-perfect linear correlation. A follow-up of 84 patients treated with mavacamten for 10 months showed that monthly... With Valsalva Synchronous changes ( Furthermore, it reaches the therapeutic standard earlier than resting pressure gradient or NYHA classification; Achieving the target (<–13 mmHg) can reduce the risk of LVOTO by 46-48%, which is equivalent to the traditional Valsalva-LVOTG <30 mmHg endpoint; no serious adverse events occurred in 574 consecutive Valsalva operations, which confirms that the test is instantaneous, does not require ultrasound equipment, and saves costs, thus having immediate translational value;

[0062] Current guidelines rely on specialist ultrasound to assess the excitation pressure gradient, hindering timely diagnosis and dosage optimization; while the NIBBP devices already available in primary healthcare can... Complete obstruction screening, treatment guidance and risk prognosis, significantly lower the entry threshold, and have high safety, suitable for elderly or high symptom burden patients; Valsalva maneuver usually induces a typical biphasic arterial pressure response, which is W-shaped curve: early rise in stage I, drop below baseline in stage II, rebound overshoot in stage III / IV, and decrease in stroke volume and blood pressure in stage II due to reduced venous return and reduced preload caused by positive thoracic pressure;

[0063] This study found that the response was significantly amplified in oHCM patients: the average in stage II of the oHCM group The blood pressure reached 31 mmHg, while the nHCM group only reached 3 mmHg; synchronous Increased by 30 times; regression equation , suggesting that for every 1 mmHg decrease in SBP, LVOTG increased by approximately 1 mmHg, consistent with the hydraulic model of a collapsible elastic conduit: the more the arterial pressure drops, the more severe the dynamic obstruction mediated by SAM; the mechanism lies in the fact that the hypertrophied and poorly compliant ventricle in oHCM is on the steep portion of the Frank-Starling curve, so a slight decrease in preload can cause a disproportionate decrease in stroke volume and SBP; at the same time, the left ventricular cavity is reduced, making the anterior leaflet of the mitral valve closer to the interventricular septum, exacerbating LVOTO; the decrease in afterload further increases the driving pressure difference required in the stenosis segment, ultimately leading to a transient 1:1 coupling between LVOTG and SBP decrease; the ROC-AUC in the training cohort was nearly perfect (0.99), 0.93 and 0.81 in the validation cohort, respectively;

[0064] The decrease in correlation coefficient was not a biological drift, but rather a result of the measurement scheme: NIBBP and Doppler were collected synchronously in the training group, eliminating phase differences and observer differences; the correlation value decreased due to physiological variations in the validation group, but the hemodynamic coupling remained; the pre-set threshold value had a sensitivity of ≥90% and a specificity of ≥87% in both cohorts, indicating that the index was robust under real-world measurement conditions, and the AUC and The slight decrease in AUC was methodological noise, which actually strengthened its clinical transferability; the proportion of oHCM in the training cohort was 28% (49 / 173), and it increased to 70% (113 / 161) in the validation cohort; such a design can maximize specificity and reduce false positives when defining the threshold, and it can test the positive predictive value under the true prevalence in the validation phase; the results of AUC, cut-off point, and sensitivity / specificity remained unchanged, proving that this biomarker reflects the intrinsic pathophysiology of oHCM rather than the effect of sample composition, and it has full-spectrum clinical universality;

[0065] In oHCM patients, the above-mentioned "phase II pressure drop" is significantly amplified: interventricular septal hypertrophy + anterior mitral valve movement (SAM) makes the left ventricular outflow tract a "collapsible elastic tube", and any decrease in preload causes the intracavity diameter to further decrease, with LVOTG rising exponentially; at the same time, the hypertrophied myocardium is already on the steep portion of the Frank-Starling curve, and stroke volume is extremely sensitive to reduced filling, causing SBP to "dive", thus producing a 1:1 hydraulic coupling of "1 mmHg SBP drop ≈ 1 mmHg LVOTG rise", making the Valsalva phase II pressure drop amplitude a "real-time reading" of LVOTO; when (II minimum mean - resting mean) < -13 mmHg, it triggers the oHCM diagnostic threshold, achieving "one breath hold, one second judgment of obstruction";

[0066] Mavacamten treatment time sequence analysis showed that is a more dynamic and equally sensitive efficacy indicator than traditional parameters; with The cumulative incidence of improved defined response almost overlaps with Valsalva-LVOTG definition (Kaplan-Meier curve), and is significantly better than resting LVOTG or NYHA classification; its physiological significance lies in that resting LVOTG can be normalized early, and sustained provable obstruction is easily underestimated; symptom improvement lags behind hemodynamic benefit due to patient adaptation and subjective factors; and Valsalva-induced Directly reflecting the elimination of dynamic obstruction under stress, it is a more integrated indicator of the physiological effect of drugs, and thus can detect the effect of drugs earlier Significant hemodynamic effect of myosin inhibitors;

[0067] In summary, The application of surpasses diagnosis and extends to the field of treatment, and is an effective, sensitive and prognostic mavacamten efficacy surrogate endpoint, and the strong correlation of Valsalva LVOTG and clinical practicability support its use as a key monitoring tool in clinical trials and daily practice, and future research needs to be prospectively verified Is the ultrasound-guided titration scheme more efficient, cost-effective and outcome-oriented than the standard ultrasound-guided scheme? From "chest-circulation coupling" to "one breath-hold, one second of obstruction judgment".

[0068] More specifically, the present application has the following characteristic points: 1. Clinical pain point driven: About 1 / 3 of HCM patients worldwide have occult LVOTO, which can only be exposed by provocation test, but ultrasound load-dependent equipment, experts and long time are needed, and the accessibility of primary hospitals is poor, leading to missed diagnosis or delayed treatment of high-risk patients; myosin inhibitors such as mavacamten need to monitor the provable gradient dynamically to guide the dosage, and traditional ultrasound follow-up has high frequency, high cost and limited patient compliance; there is an urgent need for an "instant, non-invasive, economical and repeatable" surrogate index that can reflect LVOTO changes in any scenario simultaneously;

[0069] 2. Physiological basis - Valsalva chest-circulation coupling: tightly closed vocal cords forced expiration → intrathoracic pressure rises 20-40 mmHg → venous return "chest external shunt" → left ventricular preload drops sharply → hypertrophied and high-systolic HCM ventricle slides along the steep segment of the Frank-Starling curve, and each stroke volume decreases sharply; preload ↓ + mitral valve-septal geometry close → SAM aggravation → LVOT dynamic obstruction momentary exacerbation; at the same time, the afterload decreases synchronously, forming a 1:1 hydraulic coupling of "1 mmHg SBP drop ≈ 1 mmHg LVOTG rise"; this amplified phase II pressure drop is the "real-time mirror" of LVOTO, which can be quantified by beat-by-beat systolic pressure SBP;

[0070] 3. Preliminary findings and mechanism verification: Small-sample preliminary experiments showed for the first time that the depth of the Valsalva II stage SBP trough was correlated with that of synchronous CW-LVOTG. A linear correlation was observed only in oHCM; simultaneous ultrasound confirmed that SAM displacement, LVOT instantaneous area, and... The near overlap suggests that the pressure drop is not simply a reflection of fluctuations, but a direct consequence of mechanical obstruction; establishing " Can be used as The scientific hypothesis of "real-time substitute variables";

[0071] 4. Technical Feasibility: The finger-type Finapers has been approved by the FDA, and its 200Hz beat-by-beat sampling can accurately capture the <1s trough of Phase II blood pressure; the open-source Arduino trigger box achieves millisecond-level alignment of ECG, blood pressure and ultrasound, solving the pain point of traditional "phase drift"; ROC training and verification showed AUC 0.99→0.93, ICC>0.91, proving that the method is robust and reproducible.

[0072] 5. Conversion Value: At the same threshold ( It takes into account both diagnostic and therapeutic endpoints, and can detect the response of Mavacaitai 4-8 weeks earlier than resting LVOTG or NYHA, naturally meeting the regulatory requirements of "surrogate endpoints"; the hardware cost is less than 1 / 10 of high-end load ultrasound, the operation is performed by nurses, and it is suitable for primary care, emergency departments, and wearable remote follow-up, filling the gap in the entire chain of "screening-titering-long-term management".

[0073] In summary, this invention, based on the quantifiable pathophysiological circuit of "Valsalva intrathoracic pressure - sudden decrease in preload - exacerbation of SAM", transforms the originally complex and experience-dependent ultrasound provocation test into a non-invasive blood pressure indicator that provides results in seconds with just one breath-hold. This solves the problem of clinical accessibility and meets the regulatory requirements of dynamic endpoints in drug trials, thus completing a closed-loop innovation from scientific discovery to practical technology.

[0074] In practical applications, the present invention specifically includes the following steps:

[0075] Step 1: Equipment Preparation and Subject Screening: 1.1 Hardware: Finapres NOVA finger blood pressure monitor (200Hz sampling), GE Vivid E95 ultrasound machine (continuous wave Doppler 1.7MHz, 180-220fps), three-lead ECG (2kHz trigger); 1.2 Subjects: Age ≥18 years, diagnosed with HCM, resting LVEF ≥55%, negative Allen test of upper limb arteries, excluding severe valvular disease, previous Septal reduction, atrial fibrillation, and upper limb arterial occlusion; 1.3 Environment: Constant temperature 22-24℃, subjects lie supine for ≥10 minutes, cuff and finger cot on the same side of the upper limb, ECG electrodes attached in the standard position;

[0076] Step 2: Synchronous Signal Acquisition: 2.1 Resting Period: Record a 1-minute stable blood pressure waveform, using the last 10 sinus cycle R waves as time anchors; simultaneously acquire resting LVOTG (continuous wave Doppler sampling line placed at the LVOT maximum flow velocity level, average of three measurements); 2.2 Valsalva Provocation: Instruct the subject to hold the sealed connector in their mouth, keeping the glottis closed, and forcefully exhale to maintain a pressure of 40 mmHg (pressure gauge feedback in real time) for 10-15 seconds; acquire data synchronously with Finapres and ultrasound, covering complete phases I-IV; 2.3 Data Post-processing: Matlab script R-wave gating ≥10 cardiac cycles ensemble average, automatically identify the lowest point in phase II, calculate the difference between the average SBP of the 5 beats after Valsalva and the average SBP of the 5 beats at rest → ; Calculate the difference between the peak Valsalva LVOTG value and the resting temperature on the ultrasound side → ;

[0077] Step 3: Threshold calibration and quality control: 3.1 Training set (n=173) is used for ROC fitting, and the maximum point of Youden exponent is determined. <–13 mmHg is the oHCM interpretation threshold, AUC 0.991; 3.2 Blind retesting on the validation set (n=161), sensitivity 100%, specificity 95%; Bland-Altman display. and The mean difference was -0.8 mmHg, and the 95% consensus limit was ±6.2 mmHg, meeting the clinically acceptable standard; 3.3 Quality control: the Finapres zero point was calibrated before each acquisition; ultrasound was independently double-blindly measured by two physicians with ≥5 years of HCM experience, and a third party was introduced for arbitration when the difference was >10%; upper limb arterial patency was confirmed by 8MHz handheld Doppler ultrasound when Allen test was abnormal;

[0078] Step 4: Implementation of drug efficacy monitoring: 4.1 Baseline: oHCM patients should complete steps 1-3 before starting mavacamten. and 4.2 Follow-up: Return to the hospital at weeks 4, 8, and 12, repeating the same procedure; blood samples were collected simultaneously to measure mavacamten concentration, and NYHA classification and LVEF were recorded; 4.3 Endpoint determination: a) SBP response: a) Decrease in blood pressure ≥13 mmHg from baseline without subjective symptoms of hypotension; b) Hemodynamic composite response: resting LVOTG < 30 mmHg + Valsalva LVOTG < 50 mmHg + LVEF > 50%; 4.4 Statistical analysis: Kaplan-Meier comparison of the difference between SBP response and ultrasound composite response time, Log-rank P < 0.05 was considered significant; Pearson follow-up and Correlation Verify the validity of the substitution;

[0079] Step 5: Reproducibility and Safety: 5.1 Short-term repeatability: 30 subjects were retested by different operators within 24 hours, ΔSBP intra-class correlation coefficient (ICC) = 0.94; 5.2 Long-term repeatability: 20 subjects were retested after 4 weeks. ICC=0.91 ICC=0.93; 5.3 Safety: No sustained arrhythmia, syncope or sudden drop in blood pressure (SBP < 80 mmHg) occurred in 574 consecutive Valsalva trials; only 2 cases of transient premature ventricular contractions occurred, which resolved spontaneously after 30 seconds of rest;

[0080] Step 6: Hardware Simplification and Wearable Expansion: 6.1 Connect the Finapres signal output port to the Raspberry Pi 4 microprocessor and run the open-source Python script (uploaded to GitHub, commit ID xxxxx) to achieve... Real-time automatic reading and Bluetooth push to mobile app; 6.2 finger sleeve replaced with flexible MEMS pressure sensor array (utility model applied for 20232xxxxxx), weight <30g, can record continuously for 24 hours; preliminary comparison of 50 cases shows that it is comparable to standard Finapers. The deviation is 0.6±3.2 mmHg, which meets the requirements of ISO81060-2;

[0081] Through the above steps, colleagues can reproduce this invention in any center equipped with basic ultrasound and non-invasive blood pressure equipment, without the need for additional high-value consumables, to complete non-invasive screening of LVOTO in HCM patients and dynamic evaluation of mavacamten efficacy; in the screening cohort collected simultaneously, the Valsalva maneuver instantly outlines the "watershed" of hemodynamics: the SBP curve of oHCM patients shows a steep trough missing in nHCM, while the baseline heart rate shows no difference between groups; at rest, the LVOTOG of oHCM is significantly elevated (26.35 vs 6.10 mmHg), After excitation, the difference further widened to 47.89 mmHg (55.72 vs 7.83 mmHg). ),correspond The blood pressure reached 31.43 mmHg, more than 30 times that of the oHCM group; accompanied by this rapid increase was a simultaneous "plunge" in the SBP—the oHCM group. The average blood pressure was 31.10 mmHg, while the nHCM was only -3.44 mmHg, resulting in a difference effect size as high as 0.76.

[0082] Linear regression revealed a strong coupling of both in a near 1:1 ratio: , , , suggesting that each 1 mmHg SBP drop brings a 1 mmHg gradient rise; the ROC curve AUC built on this relationship was near perfect (0.992), taking a cut-off of 1.0 yielded 100% sensitivity and 95% specificity, positive likelihood ratio 20.7, true positive rate 98%, and true negative rate 95%; the multivariate model confirmed this threshold as the only independent predictor of oHCM, and its diagnostic efficacy was not affected by any subgroup, including age, gender, hypertension, blockers, calcium antagonists, or family history, thus establishing the widespread feasibility and pathophysiological universality of "one breath-hold, one-second judgment of obstruction", the association of and in the screening cohort: hemodynamic coupling recorded synchronously.

[0083] In summary, the present application has the following advantages: 1. The device is extremely simple: only a flexible MEMS pressure array is needed, without the need for high-end load ultrasound software or three-dimensional modules, which can be deployed in primary hospitals, outpatient clinics, and even wearable scenarios; 2. Zero threshold for operation: a nurse / technician can complete the operation independently after 5 minutes of training; the system automatically identifies Valsalva phase II and calculates the results, eliminating the dependence on experienced ultrasound physicians and solving the "shortage of personnel" problem; 3. Instantaneous results: from the end of blowing to the pop-up window on the screen, ≤3s, achieving "bedside-real-time" interpretation, without the need for patients to wait for offline measurement and report transfer, significantly shortening the time of medical treatment; 4. Economic cost drops sharply: no additional consumables are needed (only a disposable mouthpiece, about 0.3 dollars); the total time of one examination is <2 minutes, and the labor cost is less than 1 / 5 of that of ultrasound load, saving more than 80% of the cost of large-scale screening or follow-up per year;

[0084] 5. Accuracy and repeatability surpass traditional methods: synchronous 200Hz beat-by-beat sampling + R-wave gated ensemble averaging, eliminating respiratory artifacts; training-validation AUC 0.99→0.93, ICC>0.91, not lower than or even better than manual measurement by experienced physicians; 6. High safety: 574 consecutive Valsalva without serious adverse events; finger cuff measurement without arterial puncture, suitable for anticoagulation, elderly, renal dysfunction, and other contraindications for invasive monitoring populations; 7. Dynamic monitoring and remote expansion: the same index ( ) both for diagnosis and for mavacamten efficacy tracking, finding treatment response 4-8 weeks ahead of NYHA classification or resting LVOTG; matched wearable MEMS finger cuff, realizing 24h gradient trend uploading at home, truly completing the closed-loop management from hospital to home; 8, guideline compatibility: not overthrowing the existing gold standard, but providing "plug and play" quantitative supplement; data can be seamlessly written into electronic medical record, directly supporting the existing guideline intervention decision for a provoked gradient <30mmHg;

[0085] In summary, the present application converts the LVOTO evaluation originally relying on high-end equipment and experts into a standardized action that can be completed instantly, economically and safely in any medical scene, providing a new scalable and replicable solution for HCM screening, drug efficacy evaluation and remote follow-up; the detection is instantaneous, does not require ultrasonic equipment and saves costs, and has instant conversion value; the existing guideline relies on specialized ultrasonic evaluation of provoked pressure difference, hindering timely diagnosis and dose optimization; and the NIBBP equipment already equipped in primary care can be used to complete obstruction screening, treatment guidance and risk prognosis, significantly reduce the access threshold, and has high safety, and is suitable for elderly or high symptom load patients;

[0086] The application of the present application goes beyond diagnosis and extends to the treatment field, and is an effective, sensitive and prognostic value mavacamten efficacy surrogate endpoint, its strong correlation with Valsalva LVOTG and clinical practicability support its use as a key monitoring tool for clinical trials and daily practice; future research needs to prospectively verify whether the titration scheme guided by the present application is more optimal than the standard ultrasonic guided scheme in terms of efficiency, cost and outcome.

[0087] The above embodiments only exemplarily illustrate the principles and effects of the present application, and are not used to limit the present application. All equivalent modifications or changes made by those skilled in the art without departing from the spirit and technical thought disclosed by the present application should still be covered by the claims of the present application.​

Claims

1. A method for monitoring and assessing left ventricular outflow tract obstruction based on non-invasive beat-by-beat blood pressure, characterized in that, The method comprises the following steps: Obtaining relevant information of a patient with hypertrophic cardiomyopathy, and dividing the patient with hypertrophic cardiomyopathy into a training set and a validation set according to the relevant information; Obtaining beat-by-beat systolic pressure of the patient in the training set recorded by a finger cuff blood pressure meter under a resting state and a Valsalva action, and simultaneously obtaining a left ventricular outflow tract pressure difference under the resting state and after the Valsalva action measured by continuous wave Doppler ultrasound; Determining a first change value and a second change value according to the beat-by-beat systolic pressure of the patient under the resting state and the Valsalva action and the left ventricular outflow tract pressure difference under the resting state and after the Valsalva action; Determining a linear quantitative relationship between the first change value and the second change value, and determining an optimal diagnostic threshold for distinguishing obstruction from non-obstruction through ROC analysis and Youden index under the action of the linear quantitative relationship; Obtaining beat-by-beat systolic pressure of a patient to be monitored collected by a flexible MEMS pressure sensor array, and confirming whether the patient has left ventricular outflow tract obstruction according to the optimal diagnostic threshold.

2. The method for monitoring and assessing left ventricular outflow tract obstruction based on non-invasive beat-by-beat blood pressure according to claim 1, characterized in that: The determination of the first change value and the second change value according to the beat-by-beat systolic pressure of the patient under the resting state and the Valsalva action and the left ventricular outflow tract pressure difference under the resting state and after the Valsalva action comprises: Determining a difference value between an average value of beat-by-beat systolic pressure of N consecutive cardiac cycles after the Valsalva action and an average value of beat-by-beat systolic pressure of N consecutive cardiac cycles under the resting state, and taking the difference value as the first change value; Determining a difference value between a peak value of the left ventricular outflow tract pressure difference after the Valsalva action and a peak value of the left ventricular outflow tract pressure difference under the resting state, and taking the difference value as the second change value.

3. The method for monitoring and assessing left ventricular outflow tract obstruction based on non-invasive beat-by-beat blood pressure according to claim 2, characterized in that: The determination of the difference value between the peak value of the left ventricular outflow tract pressure difference after the Valsalva action and the peak value of the left ventricular outflow tract pressure difference under the resting state, and taking the difference value as the second change value, comprises: Determining the difference value between the peak value of the left ventricular outflow tract pressure difference after the Valsalva action and the peak value of the left ventricular outflow tract pressure difference under the resting state according to the following formula: ； wherein, is the second change value, i.e. the change value of the left ventricular outflow tract pressure difference; is the peak flow rate of the left ventricular outflow tract pressure difference in the resting state; is the peak flow rate of the left ventricular outflow tract pressure difference after the Valsalva maneuver.

4. The method of monitoring and assessing left ventricular outflow tract obstruction based on beat-to-beat non-invasive blood pressure of claim 1, wherein: The determination of the linear quantitative relationship between the first change value and the second change value comprises: Establishing the linear quantitative relationship between the first change value and the second change value through Pearson correlation analysis.

5. The method for monitoring and assessing left ventricular outflow tract obstruction based on non-invasive beat-by-beat blood pressure according to claim 4, characterized in that: The establishment of the linear quantitative relationship between the first change value and the second change value through Pearson correlation analysis comprises: Performing Pearson correlation analysis on the first change value and the second change value of the patient in the training set, and calculating a correlation coefficient, wherein it is known from the correlation coefficient that there is a strong positive linear correlation between the first change value and the second change value; Establishing a linear regression model according to the correlation coefficient, and fitting a linear equation to obtain the linear quantitative relationship between the first change value and the second change value through the least square method in establishing the linear regression model.

6. The method for monitoring and assessing left ventricular outflow tract obstruction based on non-invasive beat-by-beat blood pressure according to claim 5, wherein: The fitting of the linear equation to obtain the linear quantitative relationship between the first change value and the second change value through the least square method in establishing the linear regression model comprises: A linear quantitative relationship between the first change value and the second change value is obtained according to the following formula: ； wherein, is the slope, , the slope 1.066 represents an average increase of 1.066 mmHg per unit decrease (increase), decrease (increase), 1.066 mmHg; is the first change value, also the change in systolic blood pressure beat by beat after the Valsalva maneuver; is a constant, has a value of 3.

87.

7. The method of monitoring and assessing left ventricular outflow tract obstruction based on beat-to-beat non-invasive blood pressure of claim 1, wherein: The optimal diagnostic threshold for distinguishing obstruction from non-obstruction is determined by ROC analysis and Youden index under the action of the linear quantitative relationship, and the optimal diagnostic threshold comprises: The first change value of the patient in the training set is taken as a test variable, and the second change value is taken as a state variable, so as to draw an ROC curve, wherein the area under the ROC curve is used to evaluate the diagnostic accuracy; The Youden index is defined as sensitivity + specificity - 1, and the threshold corresponding to the maximum Youden index is the optimal diagnostic threshold. The sensitivity and specificity of each threshold corresponding to the first change value are calculated, and the optimal diagnostic threshold for distinguishing obstruction from non-obstruction is obtained according to the sensitivity and specificity.

8. The method of monitoring and assessing left ventricular outflow tract obstruction based on beat-to-beat non-invasive blood pressure of claim 1, wherein: After obtaining the linear quantitative relationship between the first change value and the second change value and the optimal diagnostic threshold for distinguishing obstruction from non-obstruction, both are extrapolated and verified using the verification set.

Images