Pulse wave velocity measurement method and electronic device
By collecting and processing the multi-dimensional characteristics and individual characteristics of pulse wave signals on portable electronic devices, and using a random forest model for fusion prediction, the problems of low PWV measurement accuracy and high equipment configuration requirements in the prior art are solved, and portability is realized , continuous, highly accurate pulse wave velocity measurement.
Patent Information
- Application Number
- CN202410888612.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-04
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-04
AI Technical Summary
The existing pulse wave velocity (PWV) measurement methods based on wearable devices have low measurement accuracy, high equipment configuration requirements, and high wearability, so it is impossible to achieve continuous measurement of PWV anytime, anywhere.
A pulse wave velocity measurement method is adopted to collect pulse wave signals from a single target measurement site on a portable electronic device, extract multi-dimensional features, combine individual characteristics of the measured object, and use a pre-trained random forest model for feature fusion and Prediction, improve the accuracy and stability of PWV measurements.
It realizes high accuracy and stability PWV measurement under low configuration requirements on portable electronic devices. It is suitable for a single target measurement site and has a wide range of applications. It can realize continuous measurement of PWV anytime and anywhere, reducing measurement errors.
Smart Images

Figure CN118415601B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of terminal technology, and in particular to a pulse wave velocity measurement method and electronic equipment. Background Art
[0002] The degree of arteriosclerosis is closely related to the health of cardiovascular system. By evaluating the degree of arteriosclerosis, the cardiovascular health status can be evaluated. At present, pulse wave velocity (PWV) is a simple, effective, economical and non-invasive indicator for evaluating the degree of arteriosclerosis. In order to measure PWV anytime and anywhere, related technologies provide a variety of PWV measurement methods based on wearable devices. However, the measurement accuracy of these PWV measurement methods is low. Summary of the invention
[0003] The embodiments of the present application provide a pulse wave velocity measurement method and electronic device, which can improve the accuracy of PWV measurement.
[0004] In a first aspect, an embodiment of the present application provides a pulse wave velocity measurement method, comprising: collecting a pulse wave signal at a target measurement site on a subject; the number of target measurement sites is 1; extracting features of the pulse wave signal from multiple dimensions to obtain multi-dimensional features of the pulse wave signal; obtaining a target feature vector based on the multi-dimensional features and individual features of the subject; inputting the target feature vector into a pre-trained key feature classification model for processing to obtain a key feature vector related to the pulse wave velocity; fusing the target feature vector and the key feature vector to obtain a fused feature vector; inputting the fused feature vector into a pre-trained first pulse wave velocity prediction model for processing to obtain a first predicted pulse wave velocity of the subject; and determining the first predicted pulse wave velocity as the actual pulse wave velocity of the subject.
[0005] The object to be tested may be any animal body with vital signs, such as a human body.
[0006] The target measurement site can be any position on the measured body where an artery is distributed. For example, taking the measured body as a human body, since the human body has a radial artery distributed at the wrist, a carotid artery distributed at the neck, a brachial artery distributed at the upper arm, and a femoral artery distributed at the thigh, the target measurement site can be any position of the human body such as the wrist, neck, upper arm, and thigh.
[0007] Exemplarily, the portable electronic device can collect the pulse wave signal of the target measurement site on the measured body through the pulse wave sensor configured therein.
[0008] For example, a vascular health research APP may be installed on the portable electronic device, and a start measurement control may be configured in the vascular health research APP. Based on this, the portable electronic device may collect the pulse wave signal of the target measurement site on the measured body when detecting that the start measurement control in the vascular health research APP is triggered.
[0009] Exemplarily, the multiple dimensions may include a time domain dimension, a frequency domain dimension, and a time-frequency domain dimension.
[0010] The individual characteristics of the subject may be input into the portable electronic device by a user.
[0011] Exemplarily, the individual characteristics of the subject may include individual attribute characteristics and individual health characteristics of the subject.
[0012] The individual attribute characteristics of the subject may include, for example, gender, age, height, weight, and body mass index (BMI), etc. The individual health characteristics of the subject may include, for example, smoking status, drinking status, hypertension status, diabetes status, hypertension medication status, and diabetes medication status, etc.
[0013] The key feature classification model can be used to predict the actual status of key features related to pulse wave velocity.
[0014] The actual status of the key features related to the pulse wave velocity may include, for example, the actual status of the systolic pressure, the actual status of the diastolic pressure, the actual status of the systolic pressure and the diastolic pressure, and the actual status of the pulse wave velocity.
[0015] The key feature classification model can be obtained by training the first random forest model using the first sample set.
[0016] Exemplarily, the first random forest model may include several decision trees. Each piece of first sample data may include a target feature vector and a key feature vector corresponding to a historical measured object (eg, human body).
[0017] Specifically, when the first random forest model is trained using the first sample set, for each decision tree in the first random forest model, several sample data can be randomly selected from the first sample set as the training data of the decision tree, and the target feature vector in the training data of each decision tree is used as the input of the decision tree, and the key feature vector in the training data of each decision tree is used as the output of the decision tree. Since different decision trees in the first random forest model are trained using different first sample data, different decision trees can classify the target feature vector from different angles, thereby improving the accuracy of the classification result finally output by the first random forest model.
[0018] Exemplarily, the fusion method between the target feature vector and the key feature vector may be splicing, that is, the portable electronic device may splice the target feature vector and the key feature vector to obtain a fused feature vector.
[0019] The first pulse wave velocity prediction model can be used to predict the pulse wave velocity of the subject according to the fused feature vector.
[0020] The first pulse wave velocity prediction model may be obtained by training the second random forest model using the second sample set.
[0021] Exemplarily, the second random forest model may include several decision trees. Each piece of second sample data may include a fused feature vector and pulse wave velocity corresponding to a historical measured object (eg, human body).
[0022] Specifically, when the second random forest model is trained using the second sample set, for each decision tree in the second random forest model, several sample data can be randomly selected from the second sample set as the training data of the decision tree, and the fused feature vector in the training data of each decision tree is used as the input of the decision tree, and the pulse wave velocity in the training data of each decision tree is used as the output of the decision tree. Since different decision trees in the second random forest model are trained using different second sample data, different decision trees can classify the fused feature vector from different angles, thereby improving the accuracy of the pulse wave velocity finally predicted by the second random forest model.
[0023] The pulse wave velocity measurement method provided in the embodiment of the present application only needs to collect the pulse wave signal of a single target measurement site on the measured body, so the pulse wave velocity measurement method has low configuration requirements for portable electronic devices, and can be applied to portable electronic devices that are only equipped with one pulse wave sensor, and has a wide range of applications; and when measuring PWV, the user only needs to wear or place the portable electronic device on a single target measurement site of the measured body to achieve continuous measurement of PWV anytime and anywhere, and the measurement difficulty is relatively low. Since PWV is determined by the target feature vector, and the target feature vector integrates the multi-dimensional features of the pulse wave signal and the individual features of the measured body, the accuracy of PWV measurement can be improved. When determining PWV based on the target feature vector, a pre-trained key feature classification model is first used to obtain a key feature vector related to the pulse wave velocity, and then the PWV is determined based on a fusion feature vector of the target feature vector and the key feature vector. The key feature vector is closely related to the pulse wave velocity but cannot be obtained directly from the pulse wave signal. Therefore, by determining PWV by fusing the key feature vectors that are more difficult to obtain, the accuracy of PWV measurement can be further improved.
[0024] In an optional implementation of the first aspect, after obtaining the first predicted pulse wave velocity of the subject, the pulse wave velocity measurement method also includes: inputting the target feature vector into a pre-trained second pulse wave velocity prediction model for processing to obtain a second predicted pulse wave velocity of the subject; and determining the actual pulse wave velocity of the subject based on the first predicted pulse wave velocity and the second predicted pulse wave velocity.
[0025] The second pulse wave velocity prediction model may be used to predict the pulse wave velocity of the subject according to the target feature vector.
[0026] The second pulse wave velocity prediction model may be obtained by training the third random forest model using the third sample set.
[0027] Exemplarily, the third random forest model may include a plurality of decision trees. Each piece of third sample data may include a target feature vector and a pulse wave velocity corresponding to a historical measured object (eg, a human body).
[0028] Specifically, when the third random forest model is trained using the third sample set, for each decision tree in the third random forest model, a number of sample data can be randomly selected from the third sample set as the training data of the decision tree, and the target feature vector in the training data of each decision tree is used as the input of the decision tree, and the pulse wave velocity in the training data of each decision tree is used as the output of the decision tree. Since different decision trees in the third random forest model are trained using different third sample data, different decision trees can classify the target feature vector from different angles, thereby improving the accuracy of the pulse wave velocity finally predicted by the third random forest model.
[0029] The pulse wave velocity measurement method provided in the embodiment of the present application can reduce the measurement error caused by the prediction error of the key eigenvector, thereby maintaining the accuracy of PWV measurement at a high level and improving the stability of PWV measurement, by directly determining the second predicted pulse wave velocity of the subject according to the target eigenvector, and then combining the first predicted pulse wave velocity and the second predicted pulse wave velocity to determine the actual pulse wave velocity of the subject.
[0030] In an optional implementation of the first aspect, determining the actual pulse wave velocity of the subject based on a first predicted pulse wave velocity and a second predicted pulse wave velocity includes: when the first predicted pulse wave velocity and the second predicted pulse wave velocity are both greater than a first velocity threshold, or are both equal to the first velocity threshold, or are both less than the first velocity threshold, determining the first predicted pulse wave velocity as the actual pulse wave velocity; when one of the first predicted pulse wave velocity and the second predicted pulse wave velocity is greater than or equal to the first velocity threshold and the other is less than the first velocity threshold, determining the second predicted pulse wave velocity as the actual pulse wave velocity.
[0031] In the pulse wave velocity measurement method provided in the embodiment of the present application, when the first predicted pulse wave velocity and the second predicted pulse wave velocity are both greater than the first velocity threshold, or are both equal to the first velocity threshold, or are both less than the first velocity threshold, it indicates that the prediction of the key feature vector is not wrong, and the more accurate first predicted pulse wave velocity is used as the actual pulse wave velocity; and when one of the first predicted pulse wave velocity and the second predicted pulse wave velocity is greater than or equal to the first velocity threshold and the other is less than the first velocity threshold, it indicates that the prediction of the key feature vector may be wrong, and the more accurate second predicted pulse wave velocity is used as the actual pulse wave velocity; thereby improving the overall accuracy of PWV measurement.
[0032] In an optional implementation of the first aspect, the multi-dimensional features include time domain features, frequency domain features, and time-frequency domain features; correspondingly, feature extraction is performed on the pulse wave signal from multiple dimensions to obtain multi-dimensional features of the pulse wave signal, including: feature extraction of the pulse wave signal from the time domain dimension to obtain the time domain features of the pulse wave signal; feature extraction of the pulse wave signal from the frequency domain dimension to obtain the frequency domain features of the pulse wave signal; feature extraction of the pulse wave signal from the time-frequency domain dimensions to obtain the time-frequency domain features of the pulse wave signal.
[0033] In an optional implementation of the first aspect, feature extraction is performed on the pulse wave signal from the time domain dimension to obtain the time domain features of the pulse wave signal, including: extracting physiological related features, waveform features and heart rate related features of the pulse wave signal from the pulse wave signal; the physiological related features include the main wave height, main wave peak time, heart cycle, normalized peak time, ejection time, reflection index, vascular sclerosis index and subendocardial myocardial vitality rate of the pulse wave; the waveform features include one or more preset percentage heights of the main wave height of the pulse wave signal, the start time and end time corresponding to each preset percentage height, the waveform width corresponding to each preset percentage height and the K spectrum value of the pulse wave signal; the heart rate related features include the heart rate of the subject and one or more preset heart rate variability indicators.
[0034] In an optional implementation of the first aspect, feature extraction is performed on the pulse wave signal from the frequency domain dimension to obtain the frequency domain features of the pulse wave signal, including: performing Fourier transform processing on the pulse wave signal to obtain a frequency domain signal of the pulse wave signal; extracting the frequency domain features of the pulse wave signal from the frequency domain signal; the frequency domain features include the energy proportion of the pulse wave signal within one or more preset frequency ranges.
[0035] In an optional implementation of the first aspect, feature extraction is performed on the pulse wave signal from the time-frequency domain dimension to obtain the time-frequency domain features of the pulse wave signal, including: performing wavelet transform processing on the pulse wave signal to obtain detail coefficients of the pulse wave signal at multiple different wavelet decomposition levels; determining the time-frequency domain features of the pulse wave signal based on the detail coefficients at multiple different wavelet decomposition levels; the time-frequency domain features include the average, standard deviation, energy and entropy of the detail coefficients of the pulse wave signal at all wavelet decomposition levels.
[0036] In a second aspect, an embodiment of the present application provides a portable electronic device, comprising: one or more processors, and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the electronic device to execute a pulse wave velocity measurement method as any implementation of the first aspect above.
[0037] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, which includes instructions. When the instructions are executed on an electronic device, the electronic device executes a pulse wave velocity measurement method as implemented in any of the above-mentioned first aspects.
[0038] In a fourth aspect, an embodiment of the present application provides a computer executable program product. When the computer executable program product is run on an electronic device, the electronic device executes the pulse wave velocity measurement method of any implementation of the first aspect.
[0039] In a fifth aspect, an embodiment of the present application provides a chip system, which is applied to an electronic device. The chip system includes one or more processors, and the one or more processors are used to call computer instructions so that the electronic device executes a pulse wave velocity measurement method such as any implementation method of the first aspect mentioned above.
[0040] It can be understood that the beneficial effects of the second to fifth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] Figure 1 is a schematic diagram of a PWV measurement method;
[0042] Figure 2 A schematic diagram of a wearable device equipped with multiple PPG sensors when measuring cfPWV;
[0043] Figure 3 A schematic diagram of a scenario in which a smart watch equipped with a PPG sensor and an ECG sensor measures PWV;
[0044] Figure 4 A schematic diagram of a measurement process of a PWV measurement method applied to a single portable electronic device provided in an embodiment of the present application;
[0045] Figure 5 A schematic diagram of a measurement process of a PWV measurement method applied to a single portable electronic device provided in another embodiment of the present application;
[0046] Figure 6 A schematic diagram of a measurement process of a PWV measurement method applied to multiple portable electronic devices provided in an embodiment of the present application;
[0047] Figure 7 A schematic architecture diagram of a hardware system of a portable electronic device provided in an embodiment of the present application;
[0048] Figure 8 A schematic architecture diagram of a software system of a portable electronic device provided in an embodiment of the present application;
[0049] Fig. 9 A schematic flow chart of a pulse wave velocity measurement method provided in an embodiment of the present application;
[0050] Fig.10 A schematic diagram of a user interface involved in a process of measuring PWV through a smart watch provided in an embodiment of the present application;
[0051] Fig.11 A specific implementation flow chart of S902 in a pulse wave velocity measurement method provided in an embodiment of the present application;
[0052] Fig.12 A waveform diagram of a pulse wave signal in a heartbeat cycle provided in an embodiment of the present application;
[0053] Fig.13 A schematic diagram of waveform characteristics of a pulse wave signal within a heartbeat cycle provided in an embodiment of the present application;
[0054] Fig.14 A schematic diagram of the NN interval of a pulse wave signal provided in an embodiment of the present application;
[0055] Fig.15A schematic diagram of a target feature vector fusion process provided in an embodiment of the present application;
[0056] Fig.16 A schematic flow chart of a pulse wave velocity measurement method provided in another embodiment of the present application;
[0057] Fig.17 A specific implementation flow chart of S909 in a pulse wave velocity measurement method provided in an embodiment of the present application;
[0058] Fig.18 A schematic diagram of a process for determining an actual pulse wave velocity provided in an embodiment of the present application;
[0059] Fig.19 A schematic diagram of a PWV measurement result provided in an embodiment of the present application;
[0060] Fig. 20 A schematic flow chart of a pulse wave velocity measurement method provided in yet another embodiment of the present application;
[0061] Fig.21 A schematic diagram of a user interface involved in a process of measuring PWV through a smart watch and a mobile phone provided in an embodiment of the present application;
[0062] Fig. 22 A schematic diagram of a display interface for PWV measurement results provided in an embodiment of the present application. DETAILED DESCRIPTION
[0063] It should be noted that the terms used in the implementation methods of the embodiments of the present application are only used to explain the specific embodiments of the present application, and are not intended to limit the present application. In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, unless otherwise specified, "multiple" means two or more than two, "at least one" and "one or more" mean one, two or more than two.
[0064] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, it is defined that the "first" and "second" features may explicitly or implicitly include one or more of the features.
[0065] References to "one embodiment" or "some embodiments" etc. described in this specification mean that a particular feature, structure or characteristic described in conjunction with the embodiment is included in one or more embodiments of the present application. Thus, the phrases "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear at different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.
[0066] Arteriosclerosis refers to a state in which the wall of an artery becomes thicker and harder, thus losing its elasticity. The degree of arteriosclerosis is closely related to cardiovascular health. For example, when the degree of arteriosclerosis reaches a certain threshold, it may induce cardiovascular diseases such as coronary heart disease or ischemic stroke. Therefore, accurately assessing the degree of arteriosclerosis can achieve an accurate assessment of cardiovascular health. At present, pulse wave velocity (PWV) is a simple, effective, economical and non-invasive indicator for assessing the degree of arteriosclerosis. Among them, the pulse wave can refer to the pulse pressure wave generated by the heart every time it pumps blood into the artery, which is transmitted along the wall of the artery. PWV can refer to the speed at which the pulse pressure wave is transmitted along the wall of the artery. Generally, the higher the degree of arteriosclerosis, the less elastic the artery is and the greater the PWV; conversely, the lower the degree of arteriosclerosis, the greater the elasticity of the artery is and the smaller the PWV is.
[0067] In a specific application, PWV can be calculated by measuring the distance between any two measurement sites on the artery and the conduction time of the pulse wave between the two measurement sites. The calculation formula can be: PWV=L / Δt; where L is the distance between any two measurement sites on the artery, and Δt is the conduction time of the pulse wave between the two measurement sites. For example, please refer to Figure 1 , is a schematic diagram of a PWV measurement method. Assume Figure 1 11 is a section of the artery. The two measurement sites corresponding to PWV are measurement site A and measurement site B. The distance between measurement site A and measurement site B is 10 cm. The time required for the pulse wave to be transmitted from measurement site A to measurement site B is 10 milliseconds. Then PWV can be 10 cm / 10 milliseconds=10 m / s.
[0068] Depending on the measurement site corresponding to PWV, PWV may include carotid-femoral artery PWV (cfPWV), carotid-radial artery PWV (crPWV) and brachial-ankle artery PWV (baPWV), etc. Among them, cfPWV refers to the PWV measured when the two measurement sites are the carotid artery and the femoral artery, which can be used to assess the degree of hardening of large central arteries. Exemplarily, large central arteries may include the aorta and its important branches, such as the carotid artery or the femoral artery. Large central arteries are mainly responsible for transporting blood from the heart to other major parts of the body, such as the brain, upper limbs or lower limbs. crPWV refers to the PWV measured when the two measurement sites are the carotid artery and the brachial artery, which can be used to assess the degree of hardening of the upper limb arteries. baPWV refers to the PWV measured when the two measurement sites are the brachial artery and the ankle artery, which can be used to assess the degree of hardening of the upper and lower limb arteries. As mentioned above, since cfPWV can assess the degree of hardening of large central arteries, and large central arteries are key nodes of the cardiovascular system, used to connect the heart with other major parts of the body, the health of large central arteries can reflect the health of the entire cardiovascular system. Based on this, cfPWV is widely considered to be an independent assessment factor for cardiovascular health, that is, the value of cfPWV can achieve a relatively accurate assessment of cardiovascular health.
[0069] The traditional PWV measurement method usually uses professional medical PWV measurement equipment to measure PWV. For example, when measuring cfPWV, the traditional measurement method is to use a medical cfPWV meter to measure the conduction time of the pulse wave between the carotid artery and the femoral artery, and then calculate the cfPWV based on the above calculation formula PWV=L / Δt. However, this traditional PWV measurement method has technical problems such as expensive measurement equipment, complicated operation, and professional operation. It is usually only applicable to PWV measurement in specific occasions such as hospitals or laboratories, and cannot achieve PWV measurement anytime and anywhere. In view of this, the relevant technology provides a variety of PWV measurement methods based on wearable devices (such as smart watches), so that users can measure PWV anytime and anywhere by wearing wearable devices with PWV measurement function.
[0070] For example, a PWV measurement method based on a wearable device is to collect photoplethysmography (PPG) signals of multiple measurement sites on an arterial blood vessel through the wearable device, and then determine the PWV based on the PPG signals of the multiple measurement sites through a portable electronic device connected to the wearable device. For example, see Figure 2 , is a schematic diagram of a wearable device equipped with multiple PPG sensors when measuring cfPWV. Figure 2 As shown, when measuring cfPWV, the PPG signal of the measurement site on the carotid artery and the PPG signal of the measurement site on the femoral artery can be collected by the wearable device 21, and then the cfPWV can be determined based on the PPG signal of the measurement site on the carotid artery and the PPG signal of the measurement site on the femoral artery by the mobile phone 22 connected to the wearable device. However, since the PWV measurement method needs to collect PPG signals of multiple measurement sites, the wearable device is required to be equipped with multiple PPG sensors corresponding to different measurement sites. For example, please continue to refer to Figure 2 When measuring cfPWV, the wearable device 21 is required to be equipped with a PPG sensor 211 corresponding to the carotid artery and a PPG sensor 212 corresponding to the femoral artery. In addition, when using the corresponding wearable device to measure PWV, the user is required to fit the PPG sensors corresponding to different measurement sites to the corresponding measurement sites. For example, please continue to refer to Figure 2 When measuring cfPWV, the user needs to fit the PPG sensor 211 corresponding to the carotid artery to the measurement site on the carotid artery, and fit the PPG sensor 212 corresponding to the femoral artery to the measurement site on the femoral artery. This will increase the difficulty of wearing the wearable device and reduce the convenience of PWV measurement.
[0071] Exemplarily, another PWV measurement method based on a wearable device is to synchronously collect electrocardiogram (ECG) signals and PPG signals at a single measurement site on an arterial blood vessel through a wearable device, and then determine the PWV measurement based on the ECG signal and the PPG signal at a single measurement site on the arterial blood vessel. However, since this PWV measurement method requires synchronous collection of two different types of signals, ECG signals and PPG signals, the wearable device is required to have ECG signal collection functions and PPG signal collection functions, that is, the wearable device is required to be equipped with an ECG sensor (such as a contact electrode) for collecting ECG signals and a PPG sensor for collecting PPG signals. It can be seen that this PWV measurement method has high performance requirements for wearable devices and is only applicable to wearable electronic devices equipped with ECG sensors and PPG sensors, and has a small scope of application. In addition, when the user uses the corresponding wearable device to measure PWV, the user needs to fit the PPG sensor on the wearable device to a single measurement site on the artery, and fit the fingertip of one hand to the ECG sensor on the wearable device, so as to achieve synchronous acquisition of the ECG signal and the PPG signal of the single measurement site on the artery. For example, taking the wearable device as a smart watch, please refer to Figure 3, is a schematic diagram of a scenario in which a smart watch equipped with an ECG sensor and a PPG sensor measures PWV. Assuming that the ECG sensor 311 is set on the side wall of the dial of the smart watch, and the PPG sensor 312 is set on the bottom of the dial of the smart watch, when the user uses the smart watch to measure PWV, the smart watch needs to be correctly worn on the wrist of one hand (for example, the right hand) so that the PPG sensor 312 at the bottom of the dial fits the artery at the wrist, and the fingertips of the fingers of the other hand (for example, the left hand) need to fit on the ECG sensor 311 on the side wall of the dial, so that the wearable device can synchronously collect the ECG signal and the PPG signal at the wrist. However, in daily life, users need to use both hands to carry out many daily activities, that is, the fingertips of the users' fingers cannot always fit on the ECG sensor on the wearable device. Therefore, the PWV measurement method cannot continuously and synchronously collect ECG signals and PPG signals, and thus cannot achieve continuous measurement of PWV.
[0072] In addition, the related art also provides some PWV measurement methods applied to wearable devices, which determine PWV based on the PPG signal of a single target measurement site. However, the measurement accuracy of these PWV measurement methods is low.
[0073] In view of this, in order to solve the technical problems of the wearable device-based PWV measurement method, such as the difficulty of wearing the wearable device, the small scope of application, the inability to achieve continuous measurement of PWV, and the low accuracy of PWV measurement, the embodiment of the present application provides a PWV measurement method, which only needs to collect the pulse wave signal of a single target measurement site on the measured body, so the configuration requirements of the portable electronic device are low, and it can be applied to a portable electronic device that is only equipped with one pulse wave sensor, and the scope of application is wide; and when measuring PWV, the user only needs to wear or place the portable electronic device on a single target measurement site of the measured body, and the continuous measurement of PWV can be achieved anytime and anywhere, and the measurement difficulty is low. The PWV measurement method extracts the multi-dimensional features of the pulse wave signal, obtains the target feature vector according to the multi-dimensional features and the individual features of the measured body, and determines the PWV according to the target feature vector. Since the target feature vector integrates the multi-dimensional features of the pulse wave signal and the individual features of the measured body, the accuracy of PWV measurement can be improved. In addition, in the process of determining PWV according to the target feature vector, a pre-trained key feature classification model is first used to obtain a key feature vector related to the pulse wave velocity, and then the PWV is determined based on a fusion feature vector of the target feature vector and the key feature vector. Since the key feature vector is closely related to the pulse wave velocity but cannot be obtained directly from the pulse wave signal, the accuracy of PWV measurement can be further improved by determining PWV through fusing the key feature vectors that are more difficult to obtain.
[0074] The PWV measurement method provided in the embodiment of the present application can be applied to portable electronic devices. For example, the portable electronic device may include a mobile phone, a tablet computer, a laptop computer, and a wearable device. The wearable device may include a wrist-worn wearable device (such as a smart watch or a smart bracelet), a neck-hanging wearable device (such as a neck massager), and an arm-worn wearable device. The embodiment of the present application does not limit the device form of the portable electronic device.
[0075] In some scenarios, the PWV measurement method provided in the embodiment of the present application can be applied to a single portable electronic device. That is, the PWV measurement method provided in the embodiment of the present application may only involve one portable electronic device. Specifically, a pulse wave sensor may be integrated on the portable electronic device, and the portable electronic device may have a PWV calculation function. Exemplarily, the portable electronic device may be a smart watch, a smart bracelet or a mobile phone.
[0076] In this scenario, the portable electronic device can collect the pulse wave signal of the corresponding part of the user through its integrated pulse wave sensor, and can process the pulse wave signal locally to obtain the PWV measurement result.
[0077] For example, taking the portable electronic device as a smart watch, see Figure 4 , is a schematic diagram of a measurement process of a PWV measurement method applied to a single portable electronic device provided in an embodiment of the present application. Figure 4 As shown, when measuring PWV, the smart watch 41 can collect the pulse wave signal 411 of the user's wrist through its integrated pulse wave sensor, and then the smart watch 41 can process the pulse wave signal 411 locally to obtain the measurement result of PWV.
[0078] For another example, taking the portable electronic device as a mobile phone, please refer to Figure 5 , is a schematic diagram of a measurement process of a PWV measurement method applied to a single portable electronic device provided by another embodiment of the present application. Figure 5 As shown, when measuring PWV, the mobile phone 51 can collect the pulse wave signal 511 of the user's chest through its integrated pulse wave sensor, and then the mobile phone 51 can process the pulse wave signal 511 locally to obtain the measurement result of PWV.
[0079] In other scenarios, the PWV measurement method provided in the embodiments of the present application can be applied to multiple portable electronic devices. That is, the PWV measurement method provided in the embodiments of the present application can involve multiple portable electronic devices. Exemplarily, the multiple portable electronic devices may include a first electronic device and a second electronic device. Among them, a pulse wave sensor may be integrated on the first electronic device, and the second electronic device may have a PWV calculation function. The first electronic device may be, for example, a smart watch, a smart bracelet, or a neck massager, and the second electronic device may be, for example, a mobile phone, a tablet computer, or a laptop computer.
[0080] In this scenario, a communication connection can be established between the first electronic device and the second electronic device. The communication connection may include, for example, a wired connection or a wireless connection. The first electronic device may collect the pulse wave signal of the corresponding part of the user through its integrated pulse wave sensor, and may send the collected pulse wave signal to the second electronic device. The second electronic device may process the pulse wave signal from the first electronic device to obtain a measurement result of PWV.
[0081] For example, the first electronic device is a smart watch and the second electronic device is a mobile phone. Figure 6 , is a schematic diagram of a measurement process of a PWV measurement method applied to multiple portable electronic devices provided in an embodiment of the present application. Figure 6 As shown, the smart watch 61 can collect the pulse wave signal 611 of the user's wrist through its integrated pulse wave sensor, and send the pulse wave signal 611 to the mobile phone 62. The mobile phone 52 can process the pulse wave signal from the smart watch 61 to obtain the measurement result of PWV.
[0082] The structure of the portable electronic device involved in the PWV measurement method provided in the embodiment of the present application is described in detail below. Figure 7 , which is a schematic architecture diagram of a hardware system of a portable electronic device provided in an embodiment of the present application.
[0083] like Figure 7As shown, the portable electronic device may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, a camera 193, a display screen 194, and a subscriber identification module (SIM) card interface 195, etc. Among them, the sensor module 180 may include a pressure sensor 180A, a gyroscope sensor 180B, an air pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity light sensor 180G, a fingerprint sensor 180H, a temperature sensor 180I, a touch sensor 180J, a bone conduction sensor 180K, an ambient light sensor 180L, a pulse wave sensor 180M, etc.
[0084] Understandably, Figure 7 The illustrated structure does not constitute a specific limitation on the portable electronic device. In other embodiments, the portable electronic device may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
[0085] For example, when the portable electronic device is a single portable electronic device in some of the above scenarios, the portable electronic device may at least include a processor 110 and a pulse wave sensor 180M, etc.
[0086] For another example, when the portable electronic device is the first electronic device among the multiple portable electronic devices in the above-mentioned other scenarios, the portable electronic device may include at least a pulse wave sensor 180M and a data transmission module. When the portable electronic device is the second electronic device among the multiple portable electronic devices in the above-mentioned other scenarios, the portable electronic device may include at least a processor 110 and a data transmission module. Exemplarily, the data transmission module may include any one or more combinations of the mobile communication module 150, the wireless communication module 160, and the USB interface 130.
[0087] The processor 110 may include one or more processing units, for example, the processor 110 may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.
[0088] The processor 110 may also be provided with a memory for storing instructions and data.
[0089] The wireless communication function of the portable electronic device can be implemented through the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, the modem processor and the baseband processor.
[0090] Antenna 1 and antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in a portable electronic device can be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve the utilization of the antennas. For example, antenna 1 can be reused as a diversity antenna for a wireless local area network. In some other embodiments, the antenna can be used in combination with a tuning switch.
[0091] The mobile communication module 150 can provide solutions for wireless communications including 2G / 3G / 4G / 5G, etc., applied to portable electronic devices. The mobile communication module 150 may include at least one filter, a switch, a power amplifier, a low noise amplifier (LNA), etc. The mobile communication module 150 can receive electromagnetic waves from the antenna 1, and filter, amplify, and process the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation. The mobile communication module 150 can also amplify the signal modulated by the modulation and demodulation processor, and convert it into electromagnetic waves for radiation through the antenna 1. In some embodiments, at least some of the functional modules of the mobile communication module 150 can be set in the processor 110. In some embodiments, at least some of the functional modules of the mobile communication module 150 can be set in the same device as at least some of the modules of the processor 110.
[0092] The modem processor may include a modulator and a demodulator. The modulator is used to modulate the low-frequency baseband signal to be transmitted into a medium-high frequency signal. The demodulator is used to demodulate the received electromagnetic wave signal into a low-frequency baseband signal. In some embodiments, the modem processor may be an independent device. In other embodiments, the modem processor may be independent of the processor 110 and may be arranged in the same device as the mobile communication module 150 or other functional modules.
[0093] The wireless communication module 160 can provide wireless communication solutions for portable electronic devices, including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared (IR), etc. The wireless communication module 160 can be one or more devices integrating at least one communication processing module. The wireless communication module 160 receives electromagnetic waves via the antenna 2, modulates the frequency of the electromagnetic wave signal and filters it, and sends the processed signal to the processor 110. The wireless communication module 160 can also receive the signal to be sent from the processor 110, modulate the frequency of it, amplify it, and convert it into electromagnetic waves for radiation through the antenna 2.
[0094] In some embodiments, antenna 1 of the portable electronic device is coupled to mobile communication module 150, and antenna 2 is coupled to wireless communication module 160, so that the portable electronic device can communicate with the network and other devices through wireless communication technology.
[0095] The internal memory 121 may be used to store computer executable program codes, which may include instructions. The processor 110 executes various functional applications and data processing of the electronic device 100 by running the instructions stored in the internal memory 121 and / or the instructions stored in the memory provided in the processor.
[0096] The pulse wave sensor 180M can be used to collect pulse wave signals. The portable electronic device can process the pulse wave signals through the processor 110 to obtain information such as the heart rate or pulse velocity of the measured object (eg, human body).
[0097] Exemplarily, the pulse wave sensor 180M may be a PPG sensor; the pulse wave signal may be a PPG signal. A PPG sensor is an optical sensor for measuring changes in blood volume. Changes in blood volume may be used to reflect information such as the heart rate or pulse wave velocity of the subject. Exemplarily, the PPG sensor may include a light emitter (e.g., a light emitting diode) and a light receiver (e.g., a photosensitive diode). The PPG sensor may emit light of a specific wavelength (e.g., infrared light) to the skin surface of the subject through a light emitter, and may receive reflected light reflected back through the skin tissue through a light receiver, and may convert the reflected light into a corresponding digital signal, and obtain blood volume changes based on the digital signal.
[0098] The software system of the portable electronic device with the above hardware system can adopt a layered architecture, an event-driven architecture, a micro-core architecture, a micro-service architecture, or a cloud architecture, etc. The embodiment of the present application takes the Android system with a layered architecture as an example to illustrate the software system of the portable electronic device.
[0099] See also Figure 8 , which is a schematic architecture diagram of a software system of a portable electronic device provided in an embodiment of the present application.
[0100] like Figure 8 As shown in Figure 1, the layered architecture can divide the software into several layers, each layer has a clear role and division of labor, and the layers communicate with each other through software interfaces. For example, the layered architecture can divide the Android system into four layers, from top to bottom: the application (APP) layer, the application framework layer, the Android runtime (Android runtime) and system library, and the kernel layer.
[0101] The application layer may include a series of application packages, such as vascular health research, WLAN, Bluetooth, sports health and other applications.
[0102] The vascular health research APP can be used to measure the pulse wave velocity of the subject, and can also evaluate the degree of arteriosclerosis (or arterial elasticity) of the subject based on the pulse wave velocity of the subject. Exemplarily, the vascular health research APP can at least include a feature extraction module, a first prediction module, a second prediction module, and a fusion module.
[0103] Among them, the feature extraction module can be used to extract the multi-dimensional features of the pulse wave signal, and obtain the target feature vector according to the multi-dimensional features of the pulse wave signal and the individual features of the subject.
[0104] The first prediction module can be used to input the target feature vector into a pre-trained key feature classification model for processing to obtain a key feature vector related to the pulse wave velocity; fuse the target feature vector and the key feature vector to obtain a fused feature vector; input the fused feature vector into a pre-trained first pulse wave velocity prediction model for processing to obtain a first predicted pulse wave velocity of the subject.
[0105] The second prediction module can be used to input the target feature vector into a pre-trained second pulse wave velocity prediction model for processing to obtain a second predicted pulse wave velocity of the subject.
[0106] The fusion module may be used to determine the actual pulse wave velocity of the subject according to the first predicted pulse wave velocity and the second predicted pulse wave velocity.
[0107] It should be noted that the specific implementation process of the functions of each module in the Vascular Health Research APP can refer to the relevant description in the subsequent method embodiments, which will not be described in detail here.
[0108] Sports health APP can be used to record and analyze users' sports and health data, and can also be used to connect or manage sports health equipment.
[0109] The application framework layer provides an application programming interface (API) and a programming framework for applications in the application layer. The application framework layer may include some predefined functions.
[0110] Exemplarily, the application framework layer may include a window manager, a content provider, a view system, a phone manager, a resource manager, a notification manager, and the like.
[0111] Android Runtime includes core libraries and virtual machines. Android runtime is responsible for scheduling and management of the Android system.
[0112] The core library consists of two parts: one part is the function that needs to be called by the Java language, and the other part is the Android core library.
[0113] The application layer and the application framework layer run in a virtual machine. The virtual machine executes the Java files of the application layer and the application framework layer as binary files. The virtual machine is used to perform functions such as object life cycle management, stack management, thread management, security and exception management, and garbage collection.
[0114] The system library can include multiple functional modules, such as surface manager, 3D graphics processing library (such as OpenGL ES), 2D graphics engine (such as SGL), media library, etc.
[0115] The kernel layer is the layer between hardware and software. The kernel layer contains at least display driver, camera driver, audio driver, and sensor driver.
[0116] It should be noted that Figure 8 Only modules related to the embodiments of the present application are shown. In other embodiments, each layer may also include any other possible modules, and each module may also include one or more sub-modules, which are not limited in the present application.
[0117] The pulse wave velocity measurement method provided in the embodiment of the present application is described in detail below with reference to the accompanying drawings.
[0118] See also Fig. 9 , is a schematic flow chart of a pulse wave velocity measurement method provided in an embodiment of the present application. The pulse wave velocity measurement method in this embodiment can be applied to a single portable electronic device. The portable electronic device can be integrated with a pulse wave sensor, and the portable electronic device can have a PWV calculation function. Fig. 9 As shown, the pulse wave velocity measurement method may include S901 to S907, which are described in detail as follows:
[0119] S901, the portable electronic device collects a pulse wave signal at a target measurement site on the measured body.
[0120] The object to be tested may be any animal body with vital signs, such as a human body.
[0121] The number of target measurement sites is 1. That is, the portable electronic device collects pulse wave signals for only a single target measurement site on the measured object each time, and obtains a pulse wave signal of the target measurement site. Exemplarily, the target measurement site may be any position on the measured object where an artery is distributed. For example, taking the measured object as a human body as an example, since the human body has a radial artery distributed at the wrist, a carotid artery distributed at the neck, a brachial artery distributed at the upper arm, and a femoral artery distributed at the thigh, the target measurement site may be any position of the human body such as the wrist, neck, upper arm, and thigh.
[0122] It is understandable that the device form of the portable electronic device may be different depending on the target measurement site. For example, when the target measurement site is the wrist, the portable electronic device may be a wrist-worn wearable device. Wrist-worn wearable devices may include, for example, smart watches or smart bracelets. When the target measurement site is the neck, the portable electronic device may be a neck-hanging wearable device. Neck-hanging wearable devices may include, for example, neck massagers. When the target measurement site is the upper arm, the portable electronic device may be an arm-worn wearable device. When the target measurement site is the thigh, the portable electronic device may be a leg-worn wearable device.
[0123] In a specific embodiment, the portable electronic device can collect the pulse wave signal of the target measurement site on the measured body through the pulse wave sensor configured therein. Exemplarily, the pulse wave sensor can be a PPG sensor, and the pulse wave signal can be a PPG signal. For example, taking the portable electronic device as a smart watch, the measured body as a human body, and the target measurement site as a human wrist as an example, the smart watch can collect the PPG signal of the human wrist through the PPG sensor configured therein, and the collected PPG signal of the human wrist is the pulse wave signal of the human wrist.
[0124] In a specific embodiment, a vascular health research APP may be installed on the portable electronic device, and a start measurement control may be configured in the vascular health research APP. Based on this, the portable electronic device may collect the pulse wave signal of the target measurement site on the measured body when detecting that the start measurement control in the vascular health research APP is triggered.
[0125] For example, taking the portable electronic device as a smart watch as an example, see Fig.10 , is a schematic diagram of a user interface involved in the process of measuring PWV through a smart watch provided in an embodiment of the present application. Fig.10 As shown in FIG. 1 , when a user needs to measure PWV, the user can wear the smart watch on the wrist and can Fig.10 The first operation may be inputted by clicking the icon 101 of the Vascular Health Research APP on the smartwatch shown in (a) of FIG. 1 . The first operation may be, for example, a click operation. The smartwatch may respond to the first operation by displaying a Fig.10 The main interface of the vascular health research APP shown in (b) of FIG. 10 is a main interface of the vascular health research APP. A start measurement control 1021 may be configured in the main interface of the vascular health research APP. The user may input a second operation for the start measurement control 1021. The second operation may be, for example, a click operation. In response to the second operation, the smart watch may call the pulse wave sensor to collect the pulse wave signal at the user's wrist, and may display the pulse wave signal as shown in FIG. Fig.10 The PWV measurement interface shown in (c) of the figure. After the smart watch completes the PWV measurement, it can display the following Fig.10 The PWV result display interface shown in (d)
[0126] In an optional embodiment, the portable electronic device can periodically collect the pulse wave signal of the target measurement site on the measured object based on a preset collection period. The preset collection period can be set according to actual needs. For example, the preset collection period can be 1 minute. That is, the portable electronic device can collect the pulse wave signal of the target measurement site every 1 minute, thereby realizing continuous monitoring of the PWV of the measured object.
[0127] S902, the portable electronic device extracts features of the pulse wave signal from multiple dimensions to obtain multi-dimensional features of the pulse wave signal.
[0128] Exemplarily, the multiple dimensions may include time domain dimension, frequency domain dimension, time-frequency domain dimension, etc. Correspondingly, the multi-dimensional features of the pulse wave signal may include time domain features, frequency domain features, and time-frequency domain features.
[0129] Based on this, S902 may specifically include the following: Fig.11 The S9021~S9023 shown are described in detail as follows:
[0130] S9021, the portable electronic device extracts features of the pulse wave signal from a time domain dimension to obtain time domain features of the pulse wave signal.
[0131] Since the pulse wave signal collected by the pulse wave sensor is a time domain signal, after the portable electronic device collects the pulse wave signal at the target measurement site, the time domain characteristics of the pulse wave signal can be directly extracted from the collected pulse wave signal.
[0132] Exemplarily, the time domain characteristics of the pulse wave signal may include physiological related characteristics, waveform characteristics, and heart rate related characteristics of the pulse wave signal.
[0133] Among them, the physiological related characteristics of the pulse wave signal may include the main wave height of the pulse wave signal, the main wave peak time, the heartbeat cycle, the normalized peak time, the ejection time, the reflection index, the vascular sclerosis index and the sub-endocardial viability ratio (SEVR) etc.
[0134] See also Fig.12 , is a waveform diagram of a pulse wave signal in a heartbeat cycle provided by an embodiment of the present application. Fig.12 As shown, in one heartbeat cycle, the main wave, the descending middle wave and the heavy wave will appear in sequence in the pulse wave signal.
[0135] The main wave is generated when the left ventricle of the heart contracts and blood is pumped into the aorta. Specifically, when blood is pumped into the aorta, the internal pressure of the artery rises rapidly, causing a steeper peak in the waveform of the pulse wave signal, which is usually called the main wave. The main wave can be used to reflect information such as the elasticity of the artery and blood pressure.
[0136] The descending isthmus is caused by the brief reverse flow of blood when the aortic valve closes. Specifically, at the end of cardiac contraction, the aortic valve closes to prevent blood from flowing back into the left ventricle. At this time, the internal pressure of the arterial blood vessels drops briefly, causing a small depression in the waveform of the pulse wave signal, which is usually called the descending isthmus. The descending isthmus can be used to reflect information such as the function of the heart valve and the elasticity of the arterial blood vessels.
[0137] Dicrotic waves are caused by blood rebounding in the arteries after the aorta is closed. This rebound phenomenon will cause the waveform of the pulse wave signal to produce a second peak, which is usually gentle and is called a dicrotic wave. The presence and morphology of dicrotic waves can provide information about peripheral vascular resistance and blood reflexes.
[0138] Specifically, the main wave height can be used to represent the peak height (ie, amplitude) of the main wave in the pulse wave signal. Fig.12 The h1 in the figure is determined as the main wave height.
[0139] The main wave peak time can be used to indicate the time required from the start of left ventricular contraction to the peak of the main wave. Fig.12 The t1 in the figure is determined as the main wave peak duration.
[0140] The heartbeat cycle can be used to indicate the duration of a complete heartbeat of the subject. Fig.12 The T in is determined as the heartbeat period.
[0141] The normalized time to peak can be used to represent the time required for the waveform of the pulse wave signal to rise by a unit of main wave height. For example, the portable electronic device can determine the ratio t1 / h1 of the main wave peak time t1 to the main wave height h1 as the normalized time to peak.
[0142] Ejection time can be used to indicate the time required for the left ventricle of the heart to perform a contraction and ejection action. Fig.12 The t2 in the equation is determined as the ejection time.
[0143] The reflection index can be used to indicate the intensity of the dicrotic wave relative to the main wave of the pulse wave signal. Exemplarily, the portable electronic device can determine the ratio of the peak height h3 of the dicrotic wave to the peak height h1 of the main wave as the reflection index.
[0144] The vascular sclerosis index can be used to indicate the degree of vascular sclerosis of the subject. Exemplarily, the portable electronic device can determine the ratio H / t4 of the subject's height H to the target duration t4 as the vascular sclerosis index. The subject's height H can be pre-input into the portable electronic device by the user. The target duration t4 can be used to indicate the time interval between the peak of the main wave and the peak of the dicrotic wave.
[0145] SEVR can be used to assess the blood supply to the myocardium below the endocardium. Fig.12 The ratio A2 / A1 of the waveform area A2 on the right side of the middle gorge to the waveform area A1 on the left side of the middle gorge is determined as SEVR.
[0146] The waveform characteristics of the pulse wave signal may include one or more preset percentage heights of the main wave height of the pulse wave signal, the start time and end time corresponding to each preset percentage height, the waveform width corresponding to each preset percentage height, and the K spectrum value of the pulse wave signal.
[0147] The number and value of the preset percentage heights can be set according to actual needs. For example, the preset percentage heights can include 10% height, 30% height, 50% height, 70% height, and 90% height, etc. The embodiment of the present application can improve the accuracy of PWV measurement by selecting multiple preset percentage heights of the main wave height to participate in the prediction of PWV.
[0148] For example, see Fig.13 , is a schematic diagram of the waveform characteristics of a pulse wave signal in a heartbeat cycle provided in an embodiment of the present application. Assuming that the main wave height of the pulse wave signal is h1, the preset percentage heights may be h1×10%, h1×30%, h1×50%, h1×70%, and h1×90%, etc. It should be noted that the intervals between each two adjacent preset percentage heights may be equal or unequal, and the embodiment of the present application does not specifically limit this.
[0149] The start time corresponding to the preset percentage height may refer to the earliest time point among the multiple time points corresponding to the preset percentage height. The end time corresponding to the preset percentage height may refer to the latest time point among the multiple time points corresponding to the preset percentage height. The waveform width corresponding to the preset percentage height may refer to the time interval between the start time and the end time corresponding to the preset percentage height. Fig.13For example, the start time and end time corresponding to 50% of the main wave height may be t5 and t6 respectively, and the waveform width corresponding to 50% of the main wave height may be t6-t5.
[0150] The K spectrum value of the pulse wave signal can be used to reflect the overall shape of the pulse wave signal. In addition, the K spectrum value of the pulse wave signal is also related to the elasticity of the arterial blood vessels. Generally, the larger the K spectrum value, the greater the elasticity of the arterial blood vessels.
[0151] Optionally, the portable electronic device can calculate the K spectrum value of the pulse wave signal by the following formula:
[0152] K =( Q m - Q max ) / ( Q max - Q min );
[0153] in, K is the K spectrum value of the pulse wave signal, Q max is the maximum amplitude of the pulse wave signal in one heartbeat cycle, Q min is the minimum amplitude of the pulse wave signal in one heartbeat cycle, Q m It is the average amplitude of the pulse wave signal in one heartbeat cycle.
[0154] For example, Q m It can be calculated by the following formula:
[0155] ;
[0156] in, T is the heartbeat cycle of the pulse wave signal, Q(t) The pulse wave signal in one heartbeat cycle t The amplitude of the moment.
[0157] The heart rate-related features of the pulse wave signal may include heart rate and one or more preset heart rate variability indicators. The heart rate may refer to the number of heartbeats per minute of the subject, which may be used to reflect the speed of the heartbeat of the subject. The preset heart rate variability indicator may be used to evaluate the heart rate variability (HRV) of the subject.
[0158] Exemplarily, the preset heart rate variability index may include the standard deviation of adjacent normal heartbeat intervals (i.e., (normal-normal, NN) intervals) (standard deviation of NN intervals, SDNN) and the root mean square of adjacent NN interval differences (root mean square of successive differences, RMSSD), etc. Among them, the NN interval may refer to the time interval between the peaks of every two main waves in the pulse wave signal.
[0159] Both SDNN and RMSSD are time domain indicators used to evaluate the heart rate variability of the subject. Among them, SDNN can be used to reflect the overall heart rate variability of the subject. RMSSD can be used to reflect the heart rate variability of the subject in the high frequency band.
[0160] In a specific embodiment, the portable electronic device can extract multiple NN intervals from the pulse wave signal, and calculate the heart rate, SDNN, and RMSSD based on the multiple NN intervals.
[0161] For example, see Fig.14 , is a schematic diagram of the NN interval of a pulse wave signal provided in an embodiment of the present application. Fig.14 Only two NN intervals are shown as examples, namely NN 1 and NN 2. Among them, NN 1 is the NN interval between the peak of main wave 1 and the peak of main wave 2, NN 2 is the NN interval between the peak of the main wave 2 and the peak of the main wave 3. Of course, in a specific application, the pulse wave signal may also include more or fewer NN intervals.
[0162] Based on this, the portable electronic device can calculate the heart rate of the subject using the following formula:
[0163] ;
[0164] in, HR is the heart rate of the subject, is the average of all NN intervals.
[0165] Portable electronic devices can calculate SDNN using the following formula:
[0166] ;
[0167] in, n is the number of NN intervals in the pulse wave signal, NN i The pulse wave signal i NN interval.
[0168] Portable electronic devices can calculate the RMSSD using the following formula:
[0169] ;
[0170] Among them, Δ NN i For the i The NN interval and the i +1 NN interval difference. Exemplarily, Δ NN 1 is the difference between the first NN interval and the second NN interval, that is, Δ NN 1= NN 2- NN 1, Δ NN 2 is the difference between the second NN interval and the third NN interval, that is, Δ NN 2= NN 3- NN 2, and so on, Δ NN i = NN i+1 - NN i .
[0171] S9022, the portable electronic device extracts features of the pulse wave signal from a frequency domain dimension to obtain frequency domain features of the pulse wave signal.
[0172] Optionally, after the portable electronic device collects the pulse wave signal at the target measurement site, it can perform Fourier transform processing on the collected pulse wave signal to obtain a frequency domain signal of the pulse wave signal, and then extract the frequency domain characteristics of the pulse wave signal from the frequency domain signal of the pulse wave signal.
[0173] Exemplarily, the frequency domain characteristics of the pulse wave signal may include the energy proportion of the pulse wave signal in one or more preset frequency ranges. The energy proportion of the pulse wave signal in a preset frequency range may refer to the ratio of the signal energy of the pulse wave in the preset frequency range to the signal energy of the pulse wave signal in the total frequency range. The signal energy of the pulse wave signal in the total frequency range may refer to the sum of the signal energies of the pulse wave in all preset frequency ranges.
[0174] The preset frequency range can be set according to actual needs, and the present application embodiment does not specifically limit this. For example, the preset frequency range may include [0.5Hz, 0.8Hz], [0.8Hz, 1.3Hz], [1.3Hz, 1.8Hz], and [1.8Hz, 2.5Hz], etc. Assume that the signal energies of the pulse wave signal in [0.5Hz, 0.8Hz], [0.8Hz, 1.3Hz], [1.3Hz, 1.8Hz], and [1.8Hz, 2.5Hz] are respectively E 1. E 2. E 3 and E 4, then the signal energy of the pulse wave signal within the total frequency range can be E 1+ E 2+ E 3+ E 4. Based on this, for example, the energy proportion of the pulse wave signal in [0.5Hz, 0.8Hz] can be E 1 / ( E 1+ E 2+ E 3+ E 4).
[0175] S9023, the portable electronic device extracts features of the pulse wave signal from the time-frequency domain dimension to obtain the time-frequency domain features of the pulse wave signal.
[0176] Optionally, after the portable electronic device collects the pulse wave signal, it can perform wavelet transform processing on the pulse wave signal to obtain detail coefficients of the pulse wave signal at multiple different wavelet decomposition levels, and can determine the time-frequency domain characteristics of the pulse wave signal based on the detail coefficients at multiple different wavelet decomposition levels. Since the wavelet transform technology can provide information about the signal in the time domain dimension and the frequency domain dimension, the wavelet transform technology can be used to obtain the time-frequency domain characteristics of the pulse wave signal.
[0177] In a specific embodiment, the portable electronic device performing wavelet transform processing on the pulse wave signal may include: performing multi-layer wavelet decomposition on the pulse wave signal by recursive decomposition to obtain detail coefficients of the pulse wave signal at each wavelet decomposition level. It should be noted that when the portable electronic device recursively decomposes the pulse wave signal, each layer of decomposition can obtain the approximate coefficient and detail coefficient of the pulse wave signal at the wavelet decomposition level. Among them, the approximate coefficient mainly includes the low-frequency component of the pulse wave signal, which can reflect the long-term change information of the pulse wave signal. The detail coefficient mainly includes the high-frequency component of the pulse wave signal, which can reflect the change characteristics of the pulse wave signal in a short time.
[0178] Exemplarily, the specific process of the portable electronic device recursively decomposing the pulse wave signal may include: decomposing the pulse wave signal to obtain the approximate coefficient and detail coefficient of the pulse wave signal at the first wavelet decomposition level; further decomposing the approximate coefficient of the pulse wave signal at the first wavelet decomposition level to obtain the approximate coefficient and detail coefficient of the pulse wave signal at the second wavelet decomposition level. And so on, until the preset number of decomposition levels is reached. Among them, the preset number of decomposition levels can be set according to actual needs, for example, the preset number of decomposition levels can be 3. Exemplarily, after the pulse wave signal is decomposed by 3 layers of wavelets, it can be expressed as: singal=approximation3+detail3+detail2+detail1. Among them, approximation3 and detail3 are the approximate coefficient and detail coefficient of the pulse wave signal at the third wavelet decomposition level, detail2 is the detail coefficient of the pulse wave signal at the second wavelet decomposition level, and detail1 is the detail coefficient of the pulse wave signal at the first wavelet decomposition level.
[0179] Optionally, the portable electronic device can use different types of wavelets at different wavelet decomposition levels to perform wavelet decomposition on the pulse wave signal. Exemplarily, different types of wavelets may include Daubechies wavelet, symlets wavelet, coiflets wavelet or Meyr wavelet, etc.
[0180] Based on this, the time-frequency domain characteristics of the pulse wave signal can include the average value, standard deviation, energy and entropy of the detail coefficients of the pulse wave signal at all wavelet decomposition levels. Among them, the average value of the detail coefficients at all wavelet decomposition levels can be used to reflect the overall level of the detail coefficients. The standard deviation of the detail coefficients at all wavelet decomposition levels can be used to reflect the degree of discreteness of the detail coefficients. The energy of the detail coefficients at all wavelet decomposition levels can be used to reflect the intensity of the detail coefficients. The entropy of the detail coefficients at all wavelet decomposition levels can be used to reflect the complexity and uncertainty of the detail coefficients.
[0181] Optionally, the portable electronic device may calculate the energy of detail coefficients at all wavelet decomposition levels by the following formula:
[0182] ;
[0183] in, Energy Efficiency is the energy of detail coefficients at all wavelet decomposition levels, m is the number of decomposition layers corresponding to the wavelet transform processing, that is, the number of detail coefficients obtained after the pulse wave signal is processed by wavelet transform. c i The pulse wave signal is iThe detail coefficients at each wavelet decomposition level.
[0184] Optionally, the portable electronic device may calculate the entropy of detail coefficients at all wavelet decomposition levels by the following formula:
[0185] ;
[0186] in, entropy is the entropy of detail coefficients at all wavelet decomposition levels, p i The pulse wave signal is i The probability distribution of detail coefficients at each wavelet decomposition level.
[0187] Exemplarily, the detail coefficient may include high-frequency components and edge information of the pulse wave signal, etc. The high-frequency component may be used to reflect the peak portion of the pulse wave signal. The edge information may be used to reflect the edge portion of the pulse wave signal that rises or falls sharply, which usually corresponds to the rapid changes in the contraction and relaxation of the heart.
[0188] It should be noted that the embodiment of the present application does not limit the execution order of S9021, S9022 and S9023. That is, the portable electronic device may execute S9021, S9022 and S9023 simultaneously, or may not execute S9021, S9022 and S9023 simultaneously.
[0189] It can be understood that the time domain characteristics, frequency domain characteristics and time-frequency domain characteristics of the pulse wave signal obtained after the portable electronic device processes the pulse wave signal are the multi-dimensional characteristics of the pulse wave signal.
[0190] S903, the portable electronic device obtains a target feature vector according to the multi-dimensional features and the individual features of the object being measured.
[0191] The individual characteristics of the subject may be input into the portable electronic device by a user.
[0192] Exemplarily, the individual characteristics of the subject may include individual attribute characteristics and individual health characteristics of the subject.
[0193] The individual attribute characteristics of the subject may include, for example, the subject's gender, age, height, weight, and body mass index (BMI), etc. The individual health characteristics of the subject may include, for example, the subject's smoking status, drinking status, hypertension status, diabetes status, hypertension medication status, and diabetes medication status, etc.
[0194] The smoking status may include smoking or non-smoking. Specifically, when the smoking status is smoking, the portable electronic device may determine that the smoking status is 1; when the smoking status is non-smoking, the portable electronic device may determine that the smoking status is 0.
[0195] The drinking status may include drinking or not drinking. Specifically, when the drinking status is drinking, the portable electronic device may determine that the drinking status has a value of 1; when the drinking status is not drinking, the portable electronic device may determine that the drinking status has a value of 0.
[0196] The hypertension condition may include having hypertension or not having hypertension. Specifically, when the hypertension condition is having hypertension, the portable electronic device may determine that the value of the hypertension condition is 1; when the hypertension condition is not having hypertension, the portable electronic device may determine that the value of the hypertension condition is 0.
[0197] The diabetes condition may include having diabetes or not having diabetes. Specifically, when the diabetes condition is having diabetes, the portable electronic device may determine that the value of the diabetes condition is 1; when the diabetes condition is not having diabetes, the portable electronic device may determine that the value of the diabetes condition is 0.
[0198] The hypertension medication status may include medication or no medication. Specifically, when the hypertension medication status is medication, the portable electronic device may determine that the value of the hypertension medication status is 1; when the hypertension medication status is no medication, the portable electronic device may determine that the value of the hypertension medication status is 0.
[0199] The diabetes medication status may include medication use or non-medication use. Specifically, when the diabetes medication status is medication use, the portable electronic device may determine that the diabetes medication status has a value of 1; when the diabetes medication status is non-medication use, the portable electronic device may determine that the diabetes medication status has a value of 0.
[0200] In a specific embodiment, the portable electronic device can fuse the multi-dimensional features of the pulse wave signal with the individual features of the subject to obtain a target feature vector. Exemplarily, the fusion method between the multi-dimensional features of the pulse wave signal and the individual features of the subject can be splicing, that is, the portable electronic device can splice the multi-dimensional features of the pulse wave signal with the individual features of the subject to obtain a target feature vector.
[0201] For example, see Fig.15, which is a schematic diagram of a fusion process of a target feature vector provided in an embodiment of the present application. Assuming that the time domain features, frequency domain features, and time-frequency domain features of the pulse wave signal are SY, PY, and SPY, respectively, and the individual attribute features and individual health features of the subject are SX and JK, respectively, the portable electronic device splices the multi-dimensional features of the pulse wave signal with the individual features of the subject to obtain a target feature vector: [SY PY SPY SX JK].
[0202] S904, the portable electronic device inputs the target feature vector into a pre-trained key feature classification model for processing to obtain a key feature vector related to the pulse wave velocity.
[0203] The key feature classification model can be used to predict the actual status of key features related to pulse wave velocity. The actual status of key features related to pulse wave velocity may include, for example, the actual status of systolic pressure, the actual status of diastolic pressure, the actual status of systolic pressure and diastolic pressure, and the actual status of pulse wave velocity.
[0204] Among them, the actual condition of systolic pressure can be used to describe whether the systolic pressure is in a high pressure state. Exemplarily, the value of the actual condition of systolic pressure can be 0 or 1. Specifically, when the systolic pressure is greater than or equal to the first pressure threshold, the key feature classification model can determine that the value of the actual condition of systolic pressure is 1; when the systolic pressure is less than the first pressure threshold, the key feature classification model can determine that the value of the actual condition of systolic pressure is 0. Among them, the first pressure threshold can be set according to actual needs. For example, the first pressure threshold can be 140 millimeters of mercury (mmHg).
[0205] The actual condition of the diastolic pressure can be used to describe whether the diastolic pressure is in a high pressure state. Exemplarily, the value of the actual condition of the diastolic pressure can be 0 or 1. Specifically, when the diastolic pressure is greater than or equal to the second pressure threshold, the key feature classification model can determine that the value of the actual condition of the diastolic pressure is 1; when the diastolic pressure is less than the second pressure threshold, the key feature classification model can determine that the value of the actual condition of the diastolic pressure is 0. The second pressure threshold can be set according to actual needs. For example, the second pressure threshold can be 190 mmHg.
[0206] The actual conditions of systolic and diastolic pressures can be used to describe whether both systolic and diastolic pressures are in a high pressure state. Exemplarily, the values of the actual conditions of systolic and diastolic pressures can be 0 or 1. Specifically, when the systolic pressure is greater than or equal to the first pressure threshold, and the diastolic pressure is greater than or equal to the second pressure threshold, the key feature classification model can determine that the values of the actual conditions of systolic and diastolic pressures are 1; when the systolic pressure is less than the first pressure threshold, or the diastolic pressure is less than the second pressure threshold, the key feature classification model can determine that the values of the actual conditions of systolic and diastolic pressures are 0.
[0207] The actual condition of the pulse wave velocity can be used to describe whether there is arteriosclerosis. Exemplarily, the value of the actual condition of the pulse wave velocity can be 0 or 1. Specifically, when the pulse wave velocity is greater than or equal to the first velocity threshold, the key feature classification model can determine that the value of the actual condition of the pulse wave velocity is 1; when the pulse wave velocity is less than the first velocity threshold, the key feature classification model can determine that the value of the actual condition of the pulse wave velocity is 0. Among them, the first velocity threshold can be set according to actual needs, for example, the first velocity threshold can be 10 meters / second (m / s).
[0208] Exemplarily, the key feature classification model can be obtained by training the first random forest model using the first sample set. The random forest model is a classifier including multiple decision trees, each of which can include multiple classification nodes. The output of the random forest model can be determined by the mode of the categories output by the multiple decision trees. Exemplarily, the first random forest model can include 50 decision trees, the maximum depth of the first random forest model can be 10, and the first random forest model can include a total of 9276 classification nodes. The first sample set can include a number of first sample data, each of which can include a target feature vector and a key feature vector corresponding to a historical measured object (such as a human body).
[0209] In a specific application, when the first random forest model is trained using the first sample set, for each decision tree in the first random forest model, a number of sample data can be randomly selected from the first sample set as the training data of the decision tree, and the target feature vector in the training data of each decision tree is used as the input of the decision tree, and the key feature vector in the training data of each decision tree is used as the output of the decision tree. Since different decision trees in the first random forest model are trained using different first sample data, different decision trees can classify the target feature vector from different angles, thereby improving the accuracy of the classification result finally output by the first random forest model.
[0210] It should be noted that the first random forest model that has been trained is a pre-trained key feature classification model.
[0211] After the portable electronic device processes the target feature vector using a pre-trained key feature classification model, each element in the obtained key feature vector can be used to represent the value of the actual condition of each key feature. That is, the key feature vector can be used to reflect the actual condition of each key feature related to the degree of arteriosclerosis. Exemplarily, assuming that after the key feature classification model processes the target feature vector, the actual condition value of systolic pressure, the actual condition value of diastolic pressure, the actual condition value of systolic pressure and diastolic pressure, and the actual condition value of pulse wave velocity are 0, 1, 1, and 0, respectively, then the key feature vector output by the key feature classification model can be [0 1 10].
[0212] S905, the portable electronic device fuses the target feature vector and the key feature vector to obtain a fused feature vector.
[0213] Optionally, the fusion method between the target feature vector and the key feature vector may be splicing, that is, the portable electronic device may splice the target feature vector and the key feature vector to obtain a fused feature vector.
[0214] Exemplarily, assuming that the target feature vector is [SY PY SPY SX JK] and the key feature vector is [0 1 1 0], the fused feature vector obtained by concatenating the target feature vector and the key feature vector may be [SY PY SPY SX JK 0 1 1 0].
[0215] S906, the portable electronic device inputs the fused feature vector into a pre-trained first pulse wave velocity prediction model for processing to obtain a first predicted pulse wave velocity of the subject.
[0216] The first pulse wave velocity prediction model can be used to predict the pulse wave velocity of the subject according to the fused feature vector.
[0217] Exemplarily, the first pulse wave velocity prediction model may be obtained by training the second random forest model using the second sample set. For example, the second random forest model may include 100 decision trees, the maximum depth of the second random forest model may be 20, and the second random forest model may include a total of 26812 classification nodes. The second sample set may include a plurality of second sample data, each of which may include a fused feature vector and pulse wave velocity corresponding to a historical measured object (e.g., a human body).
[0218] In a specific application, when the second random forest model is trained using the second sample set, for each decision tree in the second random forest model, a number of sample data can be randomly selected from the second sample set as the training data of the decision tree, and the fused feature vector in the training data of each decision tree is used as the input of the decision tree, and the pulse wave velocity in the training data of each decision tree is used as the output of the decision tree. Since different decision trees in the second random forest model are trained using different second sample data, different decision trees can classify the fused feature vector from different angles, thereby improving the accuracy of the pulse wave velocity finally predicted by the second random forest model.
[0219] It should be noted that the trained second random forest model is the pre-trained first pulse wave velocity prediction model.
[0220] S907: The portable electronic device determines the first predicted pulse wave velocity as the actual pulse wave velocity of the subject.
[0221] As can be seen from the above, the pulse wave velocity measurement method provided in this embodiment only needs to collect the pulse wave signal of a single target measurement site on the measured body, so the pulse wave velocity measurement method has low configuration requirements for portable electronic devices, and can be applied to portable electronic devices that are only equipped with one pulse wave sensor, and has a wide range of applications; and when measuring PWV, the user only needs to wear or place the portable electronic device on a single target measurement site of the measured body to achieve continuous measurement of PWV anytime and anywhere, and the measurement difficulty is relatively low. Since PWV is determined by the target feature vector, and the target feature vector integrates the multi-dimensional characteristics of the pulse wave signal and the individual characteristics of the measured body, it can improve the accuracy of PWV measurement. When determining PWV based on the target feature vector, a pre-trained key feature classification model is first used to obtain a key feature vector related to the pulse wave velocity, and then the PWV is determined based on a fusion feature vector of the target feature vector and the key feature vector. The key feature vector is closely related to the pulse wave velocity but cannot be obtained directly from the pulse wave signal. Therefore, by determining PWV by fusing the key feature vectors that are more difficult to obtain, the accuracy of PWV measurement can be further improved.
[0222] It is understandable that the actual conditions of the key features predicted by the key feature classification model will inevitably have errors, which will affect the accuracy of the final PWV measurement result. Based on this, in order to improve the overall accuracy of PWV measurement, after S906, the pulse wave velocity measurement method may also include the following: Fig.16 S908~S909 shown are described in detail as follows:
[0223] S908, the portable electronic device inputs the target feature vector into a pre-trained second pulse wave velocity prediction model for processing to obtain a second predicted pulse wave velocity of the subject.
[0224] The second pulse wave velocity prediction model can be used to predict the pulse wave velocity of the subject according to the target feature vector.
[0225] Exemplarily, the second pulse wave velocity prediction model may be obtained by training the third random forest model using the third sample set. For example, the third random forest model may include 100 decision trees, the maximum depth of the third random forest model may be 20, and the third random forest model may include 56210 classification nodes in total. The third sample set may include a plurality of third sample data, each of which may include a target feature vector and pulse wave velocity corresponding to a historical measured object (e.g., a human body).
[0226] In a specific application, when the third random forest model is trained using the third sample set, for each decision tree in the third random forest model, a number of sample data can be randomly selected from the third sample set as the training data of the decision tree, and the target feature vector in the training data of each decision tree is used as the input of the decision tree, and the pulse wave velocity in the training data of each decision tree is used as the output of the decision tree. Since different decision trees in the third random forest model are trained using different third sample data, different decision trees can classify the target feature vector from different angles, thereby improving the accuracy of the pulse wave velocity finally predicted by the third random forest model.
[0227] It should be noted that the trained third random forest model is the pre-trained second pulse wave velocity prediction model.
[0228] S909: The portable electronic device determines the actual pulse wave velocity of the subject according to the first predicted pulse wave velocity and the second predicted pulse wave velocity.
[0229] In a specific implementation, the portable electronic device is configured to: Fig.17 S9091~S9093 shown in the figure determine the actual pulse wave velocity of the measured body, as detailed as follows:
[0230] S9091, the portable electronic device determines whether the first predicted pulse wave velocity and the second predicted pulse wave velocity are both greater than a first velocity threshold, or are both equal to the first velocity threshold, or are both less than the first velocity threshold.
[0231] Exemplarily, the first speed threshold may be 10 m / s.
[0232] S9092: When the first predicted pulse wave velocity and the second predicted pulse wave velocity are both greater than the first velocity threshold, or both equal to the first velocity threshold, or both less than the first velocity threshold, the portable electronic device determines the first predicted pulse wave velocity as the actual pulse wave velocity.
[0233] S9093: When one of the first predicted pulse wave velocity and the second predicted pulse wave velocity is greater than or equal to the first velocity threshold and the other is less than the first velocity threshold, the portable electronic device determines the second predicted pulse wave velocity as the actual pulse wave velocity.
[0234] It can be seen that in this embodiment, the process of determining the actual pulse wave velocity can be as follows: Fig.18 shown.
[0235] For example, see Fig.19 , is a schematic diagram of a PWV measurement result provided in an embodiment of the present application. Fig.19 The ordinate of (a) is the second predicted pulse wave velocity predicted by the second pulse wave velocity prediction model. Fig.19 The vertical axis of (b) is the first predicted pulse wave velocity predicted by the first pulse wave velocity prediction model. Fig.19 The ordinate of (c) is the actual pulse wave velocity determined based on the first predicted pulse wave velocity and the second predicted pulse wave velocity. Fig.19 The horizontal axis of (a) in Fig.19 The horizontal axis of (b) and Fig.19 The horizontal axis of (c) in the figure is used to represent the actual measured pulse wave velocity. Fig.19 The mean absolute error (MAE) corresponding to (a) is 0.8578. Fig.19 The MAE corresponding to (b) is 0.8406. Fig.19 The MAE corresponding to (c) in the figure is 0.8193. MAE can be used to represent the mean absolute error between the ordinate and the abscissa. The smaller the MAE, the more accurate the predicted result.
[0236] from Fig.19 It can be seen that the accuracy of the first predicted pulse wave velocity predicted by the first pulse wave velocity prediction model is higher than the accuracy of the second predicted pulse wave velocity predicted by the second pulse wave velocity prediction model. The actual pulse wave velocity determined based on the first predicted pulse wave velocity and the second predicted pulse wave velocity has the highest accuracy.
[0237] Taking the portable electronic device as a smart watch as an example, in the embodiment of the present application, after obtaining the actual pulse wave velocity of the subject, the smart watch can display the following information: Fig.10(d) shows the PWV measurement result display interface.
[0238] It can be seen from the above that the pulse wave velocity measurement method provided in this embodiment can reduce the measurement error caused by the error in the prediction of the key eigenvector, thereby maintaining the accuracy of PWV measurement at a high level and improving the stability of PWV measurement, by directly determining the second predicted pulse wave velocity of the subject according to the target eigenvector and then determining the actual pulse wave velocity of the subject by combining the first predicted pulse wave velocity and the second predicted pulse wave velocity.
[0239] See also Fig. 20 , which is a schematic flow chart of a pulse wave velocity measurement method provided in another embodiment of the present application. The pulse wave velocity measurement method in this embodiment can be applied to multiple portable electronic devices. Exemplarily, the multiple portable electronic devices may include a first electronic device and a second electronic device. The pulse wave velocity measurement method in this embodiment is described in detail below, taking the example that the first electronic device is integrated with a pulse wave sensor and the second electronic device has a PWV calculation function. Fig. 20 As shown, the pulse wave velocity measurement method may include S2000 to S2008, which are described in detail as follows:
[0240] S2000: The second electronic device sends a pulse wave velocity measurement instruction to the first electronic device.
[0241] For example, a vascular health research APP may be installed on the second electronic device, and a start measurement control may be configured in the vascular health research APP. Based on this, the second electronic device may send a pulse wave velocity measurement instruction to the first electronic device when detecting that the start measurement control in the vascular health research APP is triggered.
[0242] For example, taking the first electronic device as a smart watch and the second electronic device as a mobile phone as an example, please refer to Fig.21 , is a user interface diagram of a process of measuring PWV through a smart watch and a mobile phone provided in an embodiment of the present application. When a user needs to measure PWV, the user can wear the smart watch on the wrist and can Fig.21 The first operation may be inputted into the icon 2101 of the Vascular Health Research APP on the mobile phone shown in (a). The first operation may be, for example, a click operation. The mobile phone may respond to the first operation by displaying the following Fig.21 The main interface of the vascular health research APP shown in (b) of FIG. 1 may be configured with a start measurement control 2102. The user may input a second operation for the start measurement control 2102. The second operation may be, for example, a click operation. The mobile phone may respond to the second operation by sending a Fig.21The smartwatch in (c) sends a pulse wave velocity measurement instruction.
[0243] S2001: A first electronic device collects a pulse wave signal at a target measurement site on a measured body.
[0244] After receiving the pulse wave velocity measurement instruction from the second electronic device, the first electronic device may collect the pulse wave signal of the target measurement site on the measured body.
[0245] It should be noted that since this step is Fig. 9 S901 in the corresponding embodiment is different only in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2001, please refer to the relevant description in the above S901, which will not be repeated here.
[0246] S2002: The first electronic device sends a pulse wave signal of a target measurement site to the second electronic device.
[0247] After the first electronic device collects the pulse wave signal of the target site of the measured body, the collected pulse wave signal can be sent to the second electronic device.
[0248] S2003, the second electronic device extracts features of the pulse wave signal from multiple dimensions to obtain multi-dimensional features of the pulse wave signal.
[0249] It should be noted that since this step is Fig. 9 S902 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2003, please refer to the relevant description in the above S902, which will not be repeated here.
[0250] S2004, the second electronic device obtains a target feature vector according to the multi-dimensional features and the individual features of the object being measured.
[0251] It should be noted that since this step is Fig. 9 S903 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2004, please refer to the relevant description in the above S903, which will not be repeated here.
[0252] S2005, the second electronic device inputs the target feature vector into a pre-trained key feature classification model for processing to obtain a key feature vector related to the pulse wave velocity.
[0253] It should be noted that since this step is Fig. 9S904 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2004, please refer to the relevant description in the above S904, which will not be repeated here.
[0254] S2006: The second electronic device fuses the target feature vector and the key feature vector to obtain a fused feature vector.
[0255] It should be noted that since this step is Fig. 9 S905 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2006, please refer to the relevant description in the above S905, which will not be repeated here.
[0256] S2007, the second electronic device inputs the fused feature vector into a pre-trained first pulse wave velocity prediction model for processing to obtain a first predicted pulse wave velocity of the subject.
[0257] It should be noted that since this step is Fig. 9 S906 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2007, please refer to the relevant description in the above S906, which will not be repeated here.
[0258] S2008: The second electronic device determines the first predicted pulse wave velocity as the actual pulse wave velocity of the subject.
[0259] It should be noted that since this step is Fig. 9 S907 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2008, please refer to the relevant description in the above S907, which will not be repeated here.
[0260] In yet another embodiment, after S2007, the pulse wave velocity measurement method may further include S2009-S2010, which are described in detail as follows:
[0261] S2009: The second electronic device inputs the target feature vector into a pre-trained second pulse wave velocity prediction model for processing to obtain a second predicted pulse wave velocity of the subject.
[0262] It should be noted that since this step is Fig.16 S908 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2009, please refer to the relevant description in the above S908, which will not be repeated here.
[0263] S2010: The second electronic device determines the actual pulse wave velocity of the subject according to the first predicted pulse wave velocity and the second predicted pulse wave velocity.
[0264] It should be noted that since this step is Fig. 9 S909 in the corresponding embodiment is only different in the execution subject, and other contents are the same. Therefore, for the specific contents involved in S2010, please refer to the relevant description in the above S909, which will not be repeated here.
[0265] After the second electronic device determines the actual pulse wave velocity of the subject, it can determine the degree of arteriosclerosis of the subject according to the actual pulse wave velocity, and can display information such as the actual pulse wave velocity of the subject and the degree of arteriosclerosis. The degree of arteriosclerosis may include normal or abnormal. For example, when the actual pulse wave velocity is less than the first velocity threshold, the second electronic device can determine that the degree of arteriosclerosis of the subject is normal; when the actual pulse wave velocity is greater than or equal to the first velocity threshold, the second electronic device can determine that the degree of arteriosclerosis of the subject is abnormal.
[0266] For example, see Fig. 22 , is a schematic diagram of a display interface of a PWV measurement result provided in an embodiment of the present application. Fig. 22 As shown, after obtaining the actual pulse wave velocity of the subject, the mobile phone can display information such as the pulse wave velocity of the subject and the degree of arteriosclerosis in the display interface of the PWV measurement results.
[0267] Based on the same technical concept, an embodiment of the present application also provides a computer-readable storage medium, which stores a computer-executable program. When the computer-executable program is called by a computer, the computer executes one or more steps in any of the above method embodiments.
[0268] Based on the same technical concept, the embodiment of the present application also provides a chip system, including a processor, the processor is coupled to a memory, and the processor executes a computer executable program stored in the memory to implement one or more steps in any of the above method embodiments. The chip system can be a single chip, or a chip module composed of multiple chips.
[0269] Based on the same technical concept, an embodiment of the present application also provides a computer executable program product. When the computer executable program product runs on an electronic device, the electronic device executes one or more steps in any of the above method embodiments.
[0270] In the above embodiments, the description of each embodiment has its own emphasis. For the parts not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. It should be understood that the size of the sequence number of each step in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0271] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state disk (SSD)), etc.
[0272] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.
[0273] The above is only a specific implementation of the embodiment of the present application, but the protection scope of the embodiment of the present application is not limited thereto, and any changes or replacements within the technical scope disclosed in the embodiment of the present application should be included in the protection scope of the embodiment of the present application. Therefore, the protection scope of the embodiment of the present application should be based on the protection scope of the claims.
Claims
1. A pulse wave velocity measurement method, characterized in that: include: Collecting pulse wave signals at target measurement sites on the measured body; the number of the target measurement sites is 1; Extracting features of the pulse wave signal from multiple dimensions to obtain multi-dimensional features of the pulse wave signal; Obtaining a target feature vector according to the multi-dimensional features and the individual features of the subject; The target feature vector is input into a pre-trained key feature classification model for processing to obtain a key feature vector related to pulse wave velocity; the key feature vector is used to reflect the actual status of each key feature related to the degree of arteriosclerosis; each element in the key feature vector is used to represent the value of the actual status of each key feature; Fusing the target feature vector and the key feature vector to obtain a fused feature vector; Inputting the fused feature vector into a pre-trained first pulse wave velocity prediction model for processing to obtain a first predicted pulse wave velocity of the subject; Inputting the target feature vector into a pre-trained second pulse wave velocity prediction model for processing to obtain a second predicted pulse wave velocity of the subject; When the first predicted pulse wave velocity and the second predicted pulse wave velocity are both greater than a first velocity threshold, or both equal to the first velocity threshold, or both less than the first velocity threshold, determining the first predicted pulse wave velocity as an actual pulse wave velocity; When one of the first predicted pulse wave velocity and the second predicted pulse wave velocity is greater than or equal to the first velocity threshold and the other is less than the first velocity threshold, the second predicted pulse wave velocity is determined as the actual pulse wave velocity.
2. The pulse wave velocity measurement method according to claim 1, characterized in that: The multi-dimensional features include time domain features, frequency domain features, and time-frequency domain features; correspondingly, the pulse wave signal is subjected to feature extraction from multiple dimensions to obtain the multi-dimensional features of the pulse wave signal, including: Extracting features of the pulse wave signal from a time domain dimension to obtain time domain features of the pulse wave signal; Extracting features of the pulse wave signal from a frequency domain dimension to obtain frequency domain features of the pulse wave signal; Feature extraction is performed on the pulse wave signal from the time-frequency domain dimension to obtain the time-frequency domain features of the pulse wave signal.
3. The pulse wave velocity measurement method according to claim 2, characterized in that: The pulse wave signal is a time domain signal; correspondingly, feature extraction is performed on the pulse wave signal from the time domain dimension to obtain the time domain feature of the pulse wave signal, including: Extracting the physiological related features, waveform features and heart rate related features of the pulse wave signal from the pulse wave signal; wherein the physiological related features include the main wave height, main wave peak time, heart cycle, normalized peak time, ejection time, reflection index, vascular sclerosis index and subendocardial myocardial vitality rate of the pulse wave; the waveform features include one or more preset percentage heights of the main wave height of the pulse wave signal, the start time and end time corresponding to each preset percentage height, the waveform width corresponding to each preset percentage height and the K spectrum value of the pulse wave signal; the heart rate related features include the heart rate of the subject and one or more preset heart rate variability indicators; the K spectrum value is used to reflect the overall shape of the pulse wave signal.
4. The pulse wave velocity measurement method according to claim 2, characterized in that: Extracting features of the pulse wave signal from a frequency domain dimension to obtain frequency domain features of the pulse wave signal includes: Performing Fourier transform processing on the pulse wave signal to obtain a frequency domain signal of the pulse wave signal; The frequency domain features of the pulse wave signal are extracted from the frequency domain signal; wherein the frequency domain features include the energy proportion of the pulse wave signal within one or more preset frequency ranges.
5. The pulse wave velocity measurement method according to claim 2, characterized in that: Extracting features of the pulse wave signal from the time-frequency domain dimension to obtain the time-frequency domain features of the pulse wave signal includes: Performing wavelet transform processing on the pulse wave signal to obtain detail coefficients of the pulse wave signal at multiple different wavelet decomposition levels; The time-frequency domain characteristics of the pulse wave signal are determined according to the detail coefficients at multiple different wavelet decomposition levels; wherein the time-frequency domain characteristics include the average value, standard deviation, energy and entropy of the detail coefficients of the pulse wave signal at all the wavelet decomposition levels.
6. A portable electronic device, characterized in that: include: one or more processors, and memory; The memory is coupled to the one or more processors, and the memory is used to store computer program codes, wherein the computer program codes include computer instructions, and the one or more processors call the computer instructions to enable the electronic device to perform the method as claimed in any one of claims 1 to 5.
7. A computer-readable storage medium, characterized in that: The computer-readable storage medium comprises instructions, and when the instructions are executed on an electronic device, the electronic device executes the method as claimed in any one of claims 1 to 5.
8. A chip system, characterized in that: The chip system is applied to an electronic device, and the chip system includes one or more processors, and the one or more processors are used to call computer instructions so that the electronic device executes the method as described in any one of claims 1 to 5.
