Verification experimental system and method for measuring blood pressure using PWTT in high pressure environment

Abstract

The present invention provides a verification experimental system and method for measuring blood pressure using PWTT in a high-pressure environment, which synchronously collects electrocardiogram (ECG) signals, photoplethysmography (PWTT) signals and invasive blood pressure signals of animals under normal pressure and different high-pressure environments; a host computer analyzes the corresponding PWTT based on the ECG signals and photoplethysmography (PWTT) signals under normal pressure and different high-pressure environments, each PWTT corresponding to an invasive blood pressure value, and multiple groups of PWTTs and invasive blood pressure values constitute a feature data set; a regression model is trained using each PWTT in the feature data training set as input and the corresponding invasive blood pressure value as output; each PWTT in the feature data prediction set is input into the trained regression model to predict the corresponding non-invasive blood pressure value, and the non-invasive blood pressure value corresponding to each PWTT is compared with the invasive blood pressure value. The comparison result shows that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value. The experiment verifies that the blood pressure value measured by the PWTT method in a high-pressure environment is effective and accurate.

Description

Technical Field

[0001] The present invention relates to the field of verification experiment technology, and in particular to a verification experiment system and method for measuring blood pressure using a PWTT method in a high-pressure environment, which is used to verify the accuracy of blood pressure values measured by the PWTT method in a high-pressure environment (below 10 MPa). Background Art

[0002] PWTT (Pulse Wave Transit Time) is a popular non-invasive blood pressure detection method. It is a non-contact measurement method suitable for measurements on special populations. This method mainly relies on the propagation time of the pulse wave. By measuring the time it takes for the pulse wave generated by the heartbeat to be reflected back to the receiver through the blood vessel wall, the blood pressure value is calculated. The main advantage of the PWTT blood pressure detection method is its non-contact and non-invasive characteristics, which makes it more widely used in special environments and special populations. Especially in high-pressure environments, traditional electronic cuff blood pressure monitors are limited by the influence of high pressure, and the normal use of their automatic inflation function is affected. The greater the pressure, the greater the impact.

[0003] Under high pressure, blood pressure may decrease due to vasodilation. However, PWTT, as an indicator of arteriosclerosis, is affected by vascular tone: increased vascular tone shortens PWTT, while decreased vascular tone prolongs it. Therefore, PWTT may change accordingly under high pressure.

[0004] There are currently no clear studies on the specific performance of PWTT in high-pressure environments. Although PWTT may show some changes in high-pressure environments, there is currently insufficient research evidence to support this. Summary of the Invention

[0005] In view of the problems and shortcomings of the prior art, the present invention provides a verification experimental system and method for measuring blood pressure using a PWTT in a high-pressure environment.

[0006] The present invention solves the above technical problems through the following technical solutions:

[0007] The present invention provides a verification experimental system for measuring blood pressure using a PWTT in a high-pressure environment. The system is characterized in that it includes a three-lead electrode sheet, a transmissive PPG sensor, a tracheal cannula, a 12MPa animal pressure chamber system including an animal pressure chamber, an electrocardiogram signal acquisition module, a pulse wave signal acquisition module, an invasive blood pressure acquisition module including a pressure sensor, and a host computer.

[0008] The first electrode of the three-lead electrode sheet is attached to the animal's left upper limb, the second electrode sheet is attached to the animal's right upper limb, and the third electrode sheet is attached to the animal's left lower limb. The transmissive PPG sensor is fixed to the animal's right lower limb. The endotracheal tube is inserted into the animal's carotid artery, and the outer end of the endotracheal tube is directly connected to an invasive blood pressure acquisition module. The animal is placed in an animal compression chamber and is anesthetized. The electrocardiogram signal acquisition module, the pulse wave signal acquisition module, and the invasive blood pressure acquisition module are placed outside the animal compression chamber and are in communication with a host computer.

[0009] The ECG signal acquisition module is used to collect ECG signals of animals under normal pressure and different high pressure environments through three-lead electrodes and transmit them to the host computer;

[0010] The pulse wave signal acquisition module is used to synchronously collect the photoplethysmographic pulse wave signals of the animal under normal pressure and different high pressure environments through a transmissive PPG sensor and transmit them to the host computer;

[0011] The invasive blood pressure acquisition module is used to synchronously acquire the invasive blood pressure signals of the animal under normal pressure and different high pressure environments through the pressure sensor and transmit them to the host computer;

[0012] The host computer is used to analyze the pulse wave transit time PWTT under normal pressure and different high-pressure environments based on the R wave and photoplethysmography signal in the electrocardiogram signal under normal pressure and different high-pressure environments, each pulse wave transit time PWTT corresponds to an invasive blood pressure value in the invasive blood pressure signal, and multiple groups of pulse wave transit times PWTT and invasive blood pressure values constitute a feature data set, a portion of which is used as a feature data training set, and the remaining portion is used as a feature data prediction set;

[0013] The host computer is also used to train and learn the constructed regression model using each pulse wave transit time PWTT in the feature data training set as input and the corresponding invasive blood pressure value as output;

[0014] The host computer is also used to input each pulse wave transmission time PWTT in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compare the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and output the comparison result. The comparison result is that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

[0015] The present invention also provides a verification experimental method for measuring blood pressure using a PWTT in a high-pressure environment, which is characterized in that it is implemented using the above-mentioned verification experimental system, and the verification experimental method includes the following steps:

[0016] S1. Experimental Preparation: Anesthetize the animal and record its height and weight; secure the animal to the operating table, insert the endotracheal tube into the animal's carotid artery via arterial puncture, connect the outer end of the endotracheal tube to an invasive blood pressure acquisition module for monitoring the animal's arterial blood pressure, attach the first electrode of the three-lead electrode sheet to the animal's left upper limb, the second electrode to the animal's right upper limb, and the third electrode to the animal's left lower limb, and attach the transmissive PPG sensor to the animal's right lower limb; and place the animal in the animal pressurized chamber.

[0017] S2, Normal Pressure: The animal pressurization chamber is at normal pressure. When the animal is in a stable state, the ECG signal acquisition module collects the animal's ECG signal through a three-lead electrode and transmits it to the host computer. The pulse wave signal acquisition module synchronously collects the animal's photoplethysmography signal through a transmissive PPG sensor and transmits it to the host computer. The invasive blood pressure acquisition module synchronously collects the animal's invasive blood pressure signal through a pressure sensor and transmits it to the host computer.

[0018] S3. Pressurization: The animal pressurization chamber is pressurized to a target high pressure value using a 12 MPa animal pressurization chamber system. When the animal is in a stable state, the ECG signal acquisition module acquires the animal's ECG signal through a three-lead electrode and transmits it to a host computer. The pulse wave signal acquisition module synchronously acquires the animal's photoelectric volume pulse wave signal through a transmissive PPG sensor and transmits it to the host computer. The invasive blood pressure acquisition module synchronously acquires the animal's invasive blood pressure signal through a pressure sensor and transmits it to the host computer. The target high pressure values are 1ATA, 2ATA, 3ATA, 4ATA, 5ATA, 6ATA, 7ATA, 8ATA, 9ATA, and 10ATA, respectively.

[0019] S4. Feature Extraction: The host computer analyzes the pulse wave transit time (PWTT) at normal pressure and at different target high pressure values based on the R wave and photoplethysmography signal in the electrocardiogram signal at normal pressure and at different target high pressure values. Each pulse wave transit time (PWTT) corresponds to an invasive blood pressure value in the invasive blood pressure signal. Multiple sets of pulse wave transit times (PWTT) and invasive blood pressure values constitute a feature data set, a portion of which serves as a feature data training set, and the remaining portion serves as a feature data prediction set.

[0020] S5. Model training: The host computer uses each pulse wave transit time PWTT in the feature data training set as input and the corresponding invasive blood pressure value as output to train and learn the constructed regression model;

[0021] S6. Model verification: The host computer inputs each pulse wave transmission time PWTT in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compares the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and outputs the comparison result. The comparison result shows that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is valid and accurate.

[0022] The positive progress effect of the present invention is:

[0023] Experimental studies have shown that PWTT detection technology can meet the requirements of real-time and continuous blood pressure detection under a net water pressure of 10MPa, effectively monitor changes in human blood pressure, and better ensure the safety of workers in special environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a structural diagram of a verification experimental system for measuring blood pressure using a PWTT in a high-pressure environment according to a preferred embodiment of the present invention.

[0025] Figure 2 Flowchart of a verification experimental method for measuring blood pressure using a PWTT in a high-pressure environment according to a preferred embodiment of the present invention. DETAILED DESCRIPTION

[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0027] Example 1

[0028] like Figure 1 As shown, this embodiment provides a verification experimental system for measuring blood pressure using PWTT in a high-pressure environment, which includes a three-lead electrode sheet 10, a transmissive PPG sensor 20, a tracheal tube, a 12MPa animal pressurization chamber system including an animal pressurization chamber, an electrocardiogram signal acquisition module 30, a pulse wave signal acquisition module 40, an invasive blood pressure acquisition module 50 including a pressure sensor 51, and a host computer 60.

[0029] The first electrode 11 of the three-lead electrode sheet 10 is attached to the rabbit's left upper limb, the second electrode 12 is attached to the rabbit's right upper limb, and the third electrode 13 is attached to the rabbit's left lower limb. The transmissive PPG sensor 20 is fixed to the rabbit's right lower limb. The endotracheal tube is inserted into the rabbit's carotid artery. The outer end of the endotracheal tube is directly connected to the invasive blood pressure acquisition module 50. The rabbit is placed in an animal compression chamber and is in an anesthetized state. The ECG signal acquisition module 30, the pulse wave signal acquisition module 40, and the invasive blood pressure acquisition module 50 are placed outside the animal compression chamber and all communicate with the host computer 60.

[0030] Perform tracheal intubation on the rabbit's neck to maintain normal breathing during the experiment to avoid tracheal obstruction caused by anesthesia.

[0031] The experiment collected the following three signals: ① Electrocardiogram (ECG) signals from the rabbit's four limbs using a three-lead system. ② Transmitted pulse wave signals from the rabbit's right lower limb using a transmissive PPG sensor. ③ Invasive blood pressure (IBP) signals from the rabbit's carotid artery were collected using arterial puncture. This invasive blood pressure signal is the gold standard for blood pressure measurement. The first two signals were continuously collected by a multi-channel physiological parameter acquisition device, and the third was collected by an invasive blood pressure monitoring device.

[0032] The invasive blood pressure acquisition module 50 uses the 3.5F single pressure catheter invasive blood pressure detection device (Millar, USA) of Ed Instrument International Trading (Shanghai) Co., Ltd., and the electrocardiogram signal acquisition module 30 and the pulse wave signal acquisition module 40 use the multi-channel physiological parameter acquisition instrument (PWTT detector) developed by Tianjin University of Technology.

[0033] The ECG signal acquisition module 30 is used to collect ECG signals of the rabbit under normal pressure and different high pressure environments (such as 1ATA, 2ATA, etc.) through the three-lead electrode sheet 10 and transmit them to the host computer 60.

[0034] The pulse wave signal acquisition module 40 is used to synchronously collect the photoplethysmography signals of the rabbit under normal pressure and different high pressure environments through the transmissive PPG sensor 20 and transmit them to the host computer 60.

[0035] The invasive blood pressure acquisition module 50 is used to synchronously acquire the invasive blood pressure signals of the rabbit under normal pressure and different high pressure environments through the pressure sensor 51 and transmit them to the host computer 60.

[0036] The upper computer 60 is used to analyze the pulse wave conduction time PWTT under normal pressure and different high-pressure environments based on the R wave and photoelectric volumetric pulse wave signal in the electrocardiogram signal under normal pressure and different high-pressure environments. Specifically, the time difference between the peak value of the photoelectric volumetric pulse wave signal at normal pressure and the nearest R wave peak value in the electrocardiogram signal is analyzed as the pulse wave conduction time PWTT under normal pressure, and the time difference between the peak value of the photoelectric volumetric pulse wave signal under different high-pressure environments and the nearest R wave peak value in the electrocardiogram signal is analyzed as the pulse wave conduction time PWTT under different high-pressure environments; each pulse wave conduction time PWTT corresponds to an invasive blood pressure value in the invasive blood pressure signal, and multiple groups of pulse wave conduction times PWTT and invasive blood pressure values constitute a feature data set, a part of which is used as a feature data training set, and the rest is used as a feature data prediction set.

[0037] The host computer 60 is also used to train and learn the constructed regression model BP=a×PWTT+b (a<0) by taking each pulse wave transmission time PWTT in the feature data training set as input and the corresponding invasive blood pressure value as output to output the values of variables a and b.

[0038] The upper computer 60 is also used to input each pulse wave transmission time PWTT in the feature data prediction set into a trained regression model (i.e., a regression model in which the values of variables a and b have been determined) to predict the corresponding non-invasive blood pressure value, and compare the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and output the comparison result. The comparison result is that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

[0039] like Figure 2 As shown, this embodiment also provides a verification experimental method for measuring blood pressure using a PWTT in a high-pressure environment, which is implemented using the above-mentioned verification experimental system. The verification experimental method includes the following steps:

[0040] Step 101, Experimental Preparation: Anesthetize the rabbit and record its height and weight; secure the rabbit to the operating table, insert a endotracheal tube into the rabbit's carotid artery through arterial puncture, connect the outer end of the endotracheal tube to the invasive blood pressure acquisition module 50 to monitor the rabbit's arterial blood pressure, attach the first electrode 11 of the three-lead electrode 10 to the rabbit's left upper limb, the second electrode 12 to the rabbit's right upper limb, and the third electrode 13 to the rabbit's left lower limb; and secure the transmissive PPG sensor 20 to the rabbit's right lower limb; and place the rabbit in an animal pressurized chamber.

[0041] Step 102, Normal Pressure: The animal pressurization chamber is at normal pressure. When the rabbit is in a stable state, the ECG signal acquisition module 30 collects the rabbit's ECG signal via the three-lead electrode 10 and transmits it to the host computer 60. The pulse wave signal acquisition module 40 synchronously collects the rabbit's photoplethysmography signal via the transmissive PPG sensor 20 and transmits it to the host computer 60. The invasive blood pressure acquisition module 50 synchronously collects the rabbit's invasive blood pressure signal via the pressure sensor 51 and transmits it to the host computer 60.

[0042] After the experimental preparation phase is completed, wait for 10-15 minutes, and start recording normal pressure data for 5 minutes after the data stabilizes. That is, when the rabbit is in a stable state, start recording the invasive blood pressure, electrocardiogram, pulse wave and other signals at this time. These data are normal pressure data.

[0043] Step 103, pressurization: Use a 12 MPa animal pressurization chamber system to pressurize the animal pressurization chamber to a target high pressure value. When the rabbit is in a stable state, the ECG signal acquisition module 30 acquires the rabbit's ECG signal through the three-lead electrode 10 and transmits it to the host computer 60. The pulse wave signal acquisition module 40 synchronously acquires the rabbit's photoplethysmography signal through the transmissive PPG sensor 20 and transmits it to the host computer 60. The invasive blood pressure acquisition module 50 synchronously acquires the rabbit's invasive blood pressure signal through the pressure sensor 51 and transmits it to the host computer 60. The target high pressure values are 1ATA, 2ATA, 3ATA, 4ATA, 5ATA, 6ATA, 7ATA, 8ATA, 9ATA and 10ATA, respectively.

[0044] Step 104, feature extraction: The host computer 60 analyzes the pulse wave transit time PWTT at normal pressure and different target high pressure values based on the R wave and photoplethysmographic signal in the electrocardiogram signal at normal pressure and different target high pressure values. Each pulse wave transit time PWTT corresponds to an invasive blood pressure value in the invasive blood pressure signal. Multiple groups of pulse wave transit times PWTT and invasive blood pressure values constitute a feature data set, a portion of which is used as a feature data training set, and the remaining portion is used as a feature data prediction set.

[0045] In step 104, the time difference between the peak value of the photoplethysmography signal at normal pressure and different target high pressure values and the nearest R wave peak value in the electrocardiogram signal is analyzed as the pulse wave transmission time PWTT at normal pressure and different target high pressure values.

[0046] Step 105, model training: The host computer 60 uses each pulse wave transmission time PWTT in the feature data training set as input and the corresponding invasive blood pressure value as output to train and learn the constructed regression model BP = a × PWTT + b (a < 0) to output the values of variables a and b.

[0047] Step 106, model verification: The upper computer 60 inputs each pulse wave transmission time PWTT in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compares the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and outputs the comparison result. The comparison result shows that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

[0048] During the experiment, ECG leads were used to collect the rabbit's ECG signals at its left upper limb, right upper limb, and left lower limb, the photoplethysmography method was used to collect the rabbit's photoplethysmography signal at its right lower limb, and the invasive blood pressure method was used to collect the rabbit's invasive blood pressure at its carotid artery. The synchronous collection of ECG signals and photoplethysmography signals can facilitate the subsequent extraction of characteristic parameters, and the collection of invasive blood pressure signals can serve as the "gold standard" of the actual blood pressure value, and will be used as a reference value for modeling training and error analysis in the future.

[0049] Through the rabbit experiment, we know that high-voltage environment has an impact on ECG, but high-voltage environment does not change the rabbit's R wave. Therefore, the R wave of ECG can be used to calculate PWTT.

[0050] The PWTT was calculated by combining the real-time invasive blood pressure values of rabbits with the ECG and pulse wave data under high pressure measured by the test equipment. A model was constructed using PWTT to predict blood pressure and analyze the relationship between the blood pressure values measured by the PWTT method and pressure.

[0051] Invasive blood pressure calculation: After the rabbit is anesthetized, its hair is cleaned. Tracheal intubation is performed and a catheter is placed through puncture into a blood vessel in the neck. The outer end of the catheter is directly connected to a pressure sensor. Because fluids transmit pressure, the pressure within the blood vessels is transmitted to the external pressure sensor through the liquid in the catheter. This allows for the acquisition of a dynamic waveform of real-time pressure changes within the blood vessels. Using specific calculation methods, the systolic and diastolic pressures, as well as the mean arterial pressure, of the blood vessels at the measured site can be obtained.

[0052] Example 2

[0053] On the basis of Example 1, the host computer 60 is used to analyze the pulse wave transmission time PWTT under normal pressure and different high-pressure environments based on the R wave and photoelectric volume pulse wave signal in the electrocardiogram signal under normal pressure and different high-pressure environments, and analyze the heart rate variability HRV under normal pressure and different high-pressure environments based on the electrocardiogram signal under normal pressure and different high-pressure environments. Each pulse wave transmission time PWTT and heart rate variability HRV corresponds to an invasive blood pressure value in the invasive blood pressure signal. Multiple groups of pulse wave transmission time PWTT, heart rate variability HRV and invasive blood pressure values constitute a feature data set, a part of which is used as a feature data training set, and the rest is used as a feature data prediction set.

[0054] The host computer 60 is also used to train and learn the regression model constructed using the machine learning method by taking each pulse wave transmission time PWTT and the corresponding heart rate variability HRV in the feature data training set as input and the corresponding invasive blood pressure value as output.

[0055] The upper computer 60 is also used to input each pulse wave transmission time PWTT and the corresponding heart rate variability HRV in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compare the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and output the comparison result. The comparison result is that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

[0056] In step 104, the host computer 60 analyzes the pulse wave transit time PWTT at normal pressure and different target high pressure values based on the R wave and photoelectric volume pulse wave signal in the electrocardiogram signal at normal pressure and different target high pressure values, and analyzes the heart rate variability HRV at normal pressure and different target high pressure values based on the electrocardiogram signal at normal pressure and different target high pressure values. Each pulse wave transit time PWTT and heart rate variability HRV corresponds to an invasive blood pressure value in the invasive blood pressure signal. Multiple groups of pulse wave transit time PWTT, heart rate variability HRV and invasive blood pressure values constitute a feature data set, a part of which is used as a feature data training set, and the rest is used as a feature data prediction set.

[0057] In step 105, the host computer 60 uses each pulse wave transmission time PWTT and the corresponding heart rate variability HRV in the feature data training set as input and the corresponding invasive blood pressure value as output to train and learn the regression model constructed using the machine learning method.

[0058] In step 106, the host computer 60 inputs each pulse wave transmission time PWTT and the corresponding heart rate variability HRV in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compares the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and outputs the comparison result. The comparison result shows that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

[0059] Example 3

[0060] On the basis of Example 2, the host computer 60 is further used to obtain ECG signals of rabbits of different heights and weights under normal pressure and different high-pressure environments, photoplethysmography signals under normal pressure and different high-pressure environments, and invasive blood pressure signals under normal pressure and different high-pressure environments. The pulse wave transit times (PWTT) under normal pressure and different high-pressure environments are respectively analyzed based on the R waves and photoplethysmography signals in the ECG signals of rabbits of different heights and weights under normal pressure and different high-pressure environments. The heart rate variability (HRV) under normal pressure and different high-pressure environments is analyzed based on the ECG signals of rabbits of different heights and weights under normal pressure and different high-pressure environments. Each pulse wave transit time (PWTT) and heart rate variability (HRV) of rabbits of different heights and weights corresponds to an invasive blood pressure value in the invasive blood pressure signal. Multiple groups of heights, weights, pulse wave transit times (PWTT), heart rate variability (HRV), and invasive blood pressure values constitute a feature data set, a portion of which is used as a feature data training set, and the remaining portion is used as a feature data prediction set.

[0061] The host computer 60 is also used to train and learn the regression model constructed using the machine learning method by taking the height, weight, pulse wave transmission time PWTT, and heart rate variability HRV in the feature data training set as input and the corresponding invasive blood pressure value as output.

[0062] The upper computer 60 is also used to input the height, weight, pulse wave transmission time PWTT, and heart rate variability HRV in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compare the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and output the comparison result. The comparison result is that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

[0063] In step 104, the host computer 60 obtains the ECG signals at normal pressure and at different target high pressure values, the photoplethysmography signals at normal pressure and at different target high pressure values, and the invasive blood pressure signals at normal pressure and at different target high pressure values of the rabbits of different heights and weights. The pulse wave transit time (PWTT) at normal pressure and at different target high pressure values is analyzed based on the R wave and photoplethysmography signals in the ECG signals of the rabbits of different heights and weights. The heart rate variability (HRV) at normal pressure and at different target high pressure values is analyzed based on the ECG signals of the rabbits of different heights and weights. Each pulse wave transit time (PWTT) and heart rate variability (HRV) of the rabbits of different heights and weights corresponds to an invasive blood pressure value in the invasive blood pressure signal. Multiple sets of heights, weights, pulse wave transit times (PWTT), heart rate variability (HRV), and invasive blood pressure values constitute a feature data set, a portion of which serves as a feature data training set, and the remaining portion serves as a feature data prediction set.

[0064] In step 105, the host computer 60 uses the height, weight, pulse wave transit time PWTT, and heart rate variability HRV in the feature data training set as input and the corresponding invasive blood pressure value as output to train the regression model constructed using the machine learning method;

[0065] In step 106, the host computer 60 inputs each height, weight, pulse wave transmission time PWTT, and heart rate variability HRV in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compares the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and outputs the comparison result. The comparison result shows that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

[0066] Although specific embodiments of the present invention have been described above, those skilled in the art will appreciate that these are merely illustrative and that the scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, and such changes and modifications are intended to fall within the scope of the present invention.

Claims

1. A verification experimental system for measuring blood pressure using PWTT in a high-pressure environment, characterized by: It includes three-lead electrodes, a transmissive PPG sensor, a tracheal tube, a 12MPa animal pressure chamber system including an animal pressure chamber, an ECG signal acquisition module, a pulse wave signal acquisition module, an invasive blood pressure acquisition module including a pressure sensor, and a host computer; The first electrode of the three-lead electrode sheet is attached to the animal's left upper limb, the second electrode is attached to the animal's right upper limb, and the third electrode is attached to the animal's left lower limb. The transmissive PPG sensor is fixed to the animal's right lower limb. The endotracheal tube is inserted into the animal's carotid artery, and the outer end of the endotracheal tube is directly connected to an invasive blood pressure acquisition module. The animal is placed in an animal compression chamber and anesthetized. The electrocardiogram signal acquisition module, the pulse wave signal acquisition module, and the invasive blood pressure acquisition module are placed outside the animal compression chamber and are in communication with a host computer. The ECG signal acquisition module is used to collect ECG signals of animals under normal pressure and different high pressure environments through three-lead electrodes and transmit them to the host computer; The pulse wave signal acquisition module is used to synchronously collect the photoplethysmography signals of the animal under normal pressure and different high pressure environments through a transmissive PPG sensor and transmit them to the host computer; The invasive blood pressure acquisition module is used to synchronously acquire the invasive blood pressure signals of the animal under normal pressure and different high pressure environments through the pressure sensor and transmit them to the host computer; The host computer is used to analyze the pulse wave transit time PWTT under normal pressure and different high-pressure environments based on the R wave and photoplethysmography signal in the electrocardiogram signal under normal pressure and different high-pressure environments, each pulse wave transit time PWTT corresponds to an invasive blood pressure value in the invasive blood pressure signal, and multiple groups of pulse wave transit times PWTT and invasive blood pressure values constitute a feature data set, a portion of which is used as a feature data training set, and the remaining portion is used as a feature data prediction set; The host computer is also used to train and learn the constructed regression model using each pulse wave transit time PWTT in the feature data training set as input and the corresponding invasive blood pressure value as output; The host computer is also used to input each pulse wave transmission time PWTT in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compare the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and output the comparison result. The comparison result is that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

2. The verification experimental system for measuring blood pressure using PWTT in a high-pressure environment according to claim 1, characterized in that: The host computer is used to analyze the pulse wave transmission time PWTT under normal pressure and different high-pressure environments based on the R wave and photoelectric volume pulse wave signal in the electrocardiogram signal under normal pressure and different high-pressure environments: the time difference between the peak value of the photoelectric volume pulse wave signal under normal pressure and different high-pressure environments and the nearest R wave peak value in the electrocardiogram signal is analyzed as the pulse wave transmission time PWTT under normal pressure and different high-pressure environments.

3. The verification experimental system for measuring blood pressure using PWTT in a high pressure environment according to claim 1, characterized in that: The regression model is ( ); The host computer is further configured to train and learn the constructed regression model using each pulse wave transit time PWTT in the feature data training set as input and the corresponding invasive blood pressure value as output, so as to output the values of variables a and b; The host computer is further used to input each pulse wave transit time PWTT in the feature data prediction set into a regression model with determined values of variables a and b to predict the corresponding non-invasive blood pressure value.

4. The verification experimental system for measuring blood pressure using PWTT in a high pressure environment according to claim 1, characterized in that: The host computer is used to analyze the pulse wave transit time PWTT under normal pressure and different high-pressure environments based on the R wave and photoplethysmogram signal in the electrocardiogram signal under normal pressure and different high-pressure environments, and analyze the heart rate variability HRV under normal pressure and different high-pressure environments based on the electrocardiogram signal under normal pressure and different high-pressure environments. Each pulse wave transit time PWTT and heart rate variability HRV corresponds to an invasive blood pressure value in the invasive blood pressure signal. Multiple groups of pulse wave transit time PWTT, heart rate variability HRV and invasive blood pressure values constitute a feature data set, a part of which is used as a feature data training set, and the rest is used as a feature data prediction set; The host computer is further used to train and learn the regression model constructed using the machine learning method using each pulse wave transit time PWTT and the corresponding heart rate variability HRV in the feature data training set as input and the corresponding invasive blood pressure value as output; The host computer is also used to input each pulse wave transmission time PWTT and the corresponding heart rate variability HRV in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compare the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and output the comparison result. The comparison result is that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

5. The verification experimental system for measuring blood pressure using PWTT in a high pressure environment according to claim 4, characterized in that: The host computer is further used to obtain electrocardiogram signals under normal pressure and in different high-pressure environments, photoelectric volumetric pulse wave signals under normal pressure and in different high-pressure environments, and invasive blood pressure signals under normal pressure and in different high-pressure environments of animals of different body lengths and weights; the pulse wave transit time PWTT under normal pressure and in different high-pressure environments is respectively analyzed based on the R wave and photoelectric volumetric pulse wave signal in the electrocardiogram signals under normal pressure and in different high-pressure environments of animals of different body lengths and weights; the heart rate variability HRV under normal pressure and in different high-pressure environments is analyzed based on the electrocardiogram signals under normal pressure and in different high-pressure environments of animals of different body lengths and weights; each pulse wave transit time PWTT and heart rate variability HRV of animals of different body lengths and weights corresponds to an invasive blood pressure value in the invasive blood pressure signal; multiple groups of body lengths, weights, pulse wave transit times PWTT, heart rate variability HRV and invasive blood pressure values constitute a feature data set, a portion of which is used as a feature data training set, and the remaining portion is used as a feature data prediction set; The host computer is also used to train and learn the regression model constructed using the machine learning method by taking each height, weight, pulse wave transit time PWTT, and heart rate variability HRV in the feature data training set as input and the corresponding invasive blood pressure value as output; The host computer is also used to input the height, weight, pulse wave transmission time PWTT, and heart rate variability HRV in the feature data prediction set into the trained regression model to predict the corresponding non-invasive blood pressure value, and compare the non-invasive blood pressure value corresponding to each pulse wave transmission time PWTT with the invasive blood pressure value corresponding to this pulse wave transmission time PWTT, and output the comparison result. The comparison result is that the non-invasive blood pressure value predicted under high pressure is equal to the invasive blood pressure value under this high pressure. The experiment verifies that the blood pressure value measured by the PWTT method under high pressure environment is effective and accurate.

Images