Blood vessel parameter measuring method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a blood vessel parameter measuring method and device, equipment and a storage medium. The blood vessel parameter measuring method comprises the steps that a surface temperature signal set of a target area under laser irradiation within a preset time period is obtained; preprocessing the surface temperature signal set to obtain a temperature sequential sequence; the temperature sequential sequence is input into a preset neural network model, at least one blood vessel parameter of the blood vessel temperature, depth and thickness of the target area is obtained, and the preset neural network model is obtained by training an initial neural network model according to sample data collected in a historical measurement scene; the sample data comprises a plurality of groups of temperature sequential sequences obtained by preprocessing an acquired surface temperature signal set in a historical measurement scene. The blood vessel parameters are measured through laser irradiation in combination with collection of the surface temperature signals, the skin or blood vessel tissue does not need to be directly contacted or damaged, and therefore non-invasive blood vessel parameter measurement is achieved.

Description

Technical Field

[0001] The present application relates to the field of signal processing technology, and in particular to a method and device, equipment, and storage medium for measuring blood vessel parameters. Background Art

[0002] In the prior art, the method of obtaining the parameters of blood vessels under the skin is usually to use laser-induced thermal therapy to make the hemoglobin in the blood vessels produce a sharp temperature rise through the selective absorption of laser energy, thereby obtaining the parameters of blood vessels under the skin. However, laser-induced thermal therapy will cause thermal damage to the capillaries in the skin, thereby damaging the skin. In order to solve this problem, a technical solution for obtaining the parameters of blood vessels under the skin without damaging the human body is needed.

[0003] Therefore, how to obtain subcutaneous vascular parameters without damaging the skin or vascular tissue is an urgent problem to be solved. Summary of the invention

[0004] In view of this, the embodiment of the present application provides a method and device for measuring blood vessel parameters, equipment, and storage medium, which can obtain blood vessel parameters under the skin without damaging the skin or blood vessel tissue. The embodiment of the present application provides a method and device for measuring blood vessel parameters, equipment, and storage medium as follows:

[0005] The present application provides a method for measuring blood vessel parameters, including:

[0006] Acquire a surface temperature signal set of the target area under laser irradiation within a preset time period;

[0007] Preprocessing the surface temperature signal set to obtain a temperature time series sequence;

[0008] The temperature time series sequence is input into a preset neural network model to obtain at least one vascular parameter of the target area including vascular temperature, depth and thickness. The preset neural network model is obtained by training an initial neural network model based on sample data collected in historical measurement scenarios. The sample data includes multiple groups of temperature time series sequences obtained by preprocessing a set of surface temperature signals collected in historical measurement scenarios.

[0009] In some embodiments, the temperature time series is input into a preset neural network model to obtain at least one vascular parameter of the target area's vascular temperature, depth, and thickness, including:

[0010] Acquire low-level features and high-level features of the temperature time series, wherein the low-level features include at least one of a temperature peak value, a valley value, and a temperature difference between adjacent acquisition points within a preset time period, and the high-level features include at least one of a temperature distribution, a temperature change, and a heat conduction feature;

[0011] Fusing the low-level features and the high-level features to obtain processed fused features;

[0012] The fusion feature is input into a preset neural network model to obtain at least one vascular parameter of the target area including vascular temperature, depth and thickness.

[0013] In some embodiments, the plurality of sets of surface temperature signal sets include sets of surface temperature signals under different temperature acquisition scenarios, and the sample data collected under the historical measurement scenarios are used to train the initial neural network model, including:

[0014] Acquire a first epidermal temperature, a second epidermal temperature, and a vascular parameter corresponding to the first epidermal temperature of a target area through a blood vessel simulation model, wherein the first epidermal temperature and the second epidermal temperature are used to characterize the skin temperature of the target area;

[0015] Obtaining a training set according to the second skin temperature;

[0016] Obtaining a validation set according to the first epidermal temperature and a blood vessel parameter corresponding to the first epidermal temperature;

[0017] Training the initial neural network model according to the training set to obtain training set vascular parameters;

[0018] Calibrate the initial neural network model according to the validation set and the training set vascular parameters, and obtain a second neural network model when the accuracy of the training set vascular parameters output by the initial neural network model meets a preset threshold;

[0019] Inputting the plurality of sets of surface temperature signal sets into the second neural network model to obtain training results;

[0020] The training result is compared with the preset blood vessel parameters, and when the deviation between the training result and the preset blood vessel parameters is less than a preset threshold, a preset neural network model is obtained.

[0021] In some embodiments, the step of acquiring a set of surface temperature signals of a target area under laser irradiation within a preset time period includes:

[0022] Acquire multiple temperature images of the target area under laser irradiation within a preset time;

[0023] Extracting a plurality of temperature signals corresponding to the plurality of temperature images;

[0024] The plurality of temperature signals are sorted according to the time sequence in which the corresponding plurality of temperature images are acquired to obtain the surface temperature signal set.

[0025] In some embodiments, the wavelength of the laser is 532 nm and the measurement range is 0-100 degrees.

[0026] In some embodiments, multiple temperature images of the target area under laser irradiation within a preset time are obtained by an infrared detector.

[0027] In some embodiments, the preset neural network model includes an input layer, a first convolutional layer, a pooling layer, fifteen residual layers, two fully connected layers and an output layer, and the residual layer includes two batch normalization layers, two rectified linear unit activation function layers, two second convolutional layers and a shortcut connection.

[0028] An embodiment of the present application provides a device for measuring blood vessel parameters, comprising:

[0029] An acquisition module, used to acquire a set of surface temperature signals of a target area under laser irradiation within a preset time period;

[0030] A processing module, used for preprocessing the surface temperature signal set to obtain a temperature time series sequence;

[0031] An input module is used to input the temperature time series sequence into a preset neural network model to obtain at least one vascular parameter of the target area, including vascular temperature, depth and thickness. The preset neural network model is obtained by training an initial neural network model based on sample data collected in historical measurement scenarios, and the sample data includes multiple groups of surface temperature signal sets collected in historical measurement scenarios.

[0032] The computer device provided in the embodiment of the present application includes a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the processor executes the program, the method described in the embodiment of the present application is implemented.

[0033] The computer-readable storage medium provided in the embodiment of the present application stores a computer program thereon, and when the computer program is executed by a processor, the method provided in the embodiment of the present application is implemented.

[0034] The embodiment of the present application provides a method and device for measuring blood vessel parameters, equipment, and storage medium, which obtains a set of surface temperature signals of a target area under laser irradiation within a preset time period; pre-processes the set of surface temperature signals to obtain a temperature time series sequence; inputs the temperature time series sequence into a preset neural network model to obtain at least one blood vessel parameter of the target area's blood vessel temperature, depth, and thickness. In this way, the blood vessel parameters are measured by laser irradiation combined with the collection of surface temperature signals, without the need to directly contact or damage the skin or blood vessel tissue, thereby achieving non-invasive blood vessel parameter measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0036] Figure 1 A schematic diagram of the implementation flow of a method for measuring blood vessel parameters provided in an embodiment of the present application;

[0037] Figure 2 A schematic diagram of a specific implementation flow of a method for measuring blood vessel parameters provided in an embodiment of the present application;

[0038] Figure 3 A model training flow chart of a method for measuring blood vessel parameters provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of a specific implementation flow of a method for measuring blood vessel parameters provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of the structure of a device for measuring blood vessel parameters disclosed in an embodiment of the present application;

[0041] Figure 6 A schematic diagram of the structure of an electronic device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the specific technical solution of the present application will be further described in detail below in conjunction with the drawings in the embodiments of the present application. The following embodiments are used to illustrate the present application, but are not used to limit the scope of the present application.

[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0044] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0045] It should be pointed out that the terms "first\second\third" involved in the embodiments of the present application are used to distinguish similar or different objects, and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described here can be implemented in an order other than that illustrated or described here.

[0046] Figure 1 The following is a schematic diagram of a method for measuring blood vessel parameters provided in an embodiment of the present application. Figure 1 As shown, the method may include the following steps 101 to 103:

[0047] Step 101: Acquire a set of surface temperature signals of a target area under laser irradiation within a preset time period.

[0048] In the embodiment of the present application, by selecting an appropriate laser type (such as a near-infrared laser) and laser irradiation intensity, it is ensured that no damage is caused to the skin or tissue. The time, power and frequency of laser irradiation can be optimized according to the specific conditions of the target area to ensure that changes in the surface temperature of the blood vessels can be stimulated.

[0049] Use a high-precision infrared thermal imager or infrared sensor. The equipment needs to have a high time resolution (at least able to record temperature data hundreds of times per second) to ensure that the temperature change process can be accurately captured.

[0050] In order to improve the accuracy of the measurement, multiple points in the target area can be collected simultaneously to form a spatial temperature signal set. This can be achieved by setting multiple detection points or using a thermal imager with a wider field of view.

[0051] During the acquisition process, the environmental stability (such as temperature, humidity, airflow, etc.) should be maintained to avoid interference of environmental factors on temperature measurement. In addition, the synchronization of laser irradiation and temperature data acquisition can be ensured through software control or manual intervention.

[0052] A dataset containing time and corresponding surface temperature in the form of a time series data matrix, where each row represents a time point and each column represents a different measurement point or channel.

[0053] Step 102: pre-process the surface temperature signal set to obtain a temperature time series.

[0054] In the embodiment of the present application, since the temperature signal acquisition process may be subject to various interferences (such as equipment noise, changes in the external environment), a filtering algorithm can be used to process the original signal. Common filtering methods include: 1. Kalman filtering: suitable for dynamic noise suppression, which can effectively remove random noise in the signal. 2. Wavelet transform: It can separate high-frequency noise and low-frequency signals in the signal and is suitable for complex temperature signals.

[0055] In order to make the data more consistent with the input requirements of the neural network, the signal can be smoothed (such as sliding average) and normalized (such as standardized to the range of [0,1]) to ensure that the temperature values ​​at each time point are processed on the same scale.

[0056] If data is lost or damaged during the signal acquisition process, an interpolation algorithm (such as linear interpolation, spline interpolation, etc.) is needed to fill in the missing data points.

[0057] Step 103: input the temperature time series sequence into a preset neural network model to obtain at least one vascular parameter of the target area's vascular temperature, depth, and thickness.

[0058] In the embodiment of the present application, a residual network (ResNet) is used to process the temperature time series. The ResNet structure has multiple "residual blocks", each of which contains a jump connection. The "residual blocks" can effectively alleviate the gradient vanishing or gradient exploding problems in deep neural networks.

[0059] The input layer is used to receive a temperature time series sequence, which may be temperature change data within a time window of fixed length (eg, a temperature signal within 10 seconds).

[0060] A combination of multi-layer convolutional layers and fully connected layers is adopted. The convolutional layers are used to extract features from time series data, and the fully connected layers are used to integrate and classify the extracted features.

[0061] The output layer is a multidimensional regression problem that predicts parameters such as the temperature, depth, and thickness of the blood vessels. The specific output dimensions are set according to actual needs (for example, outputting the temperature value, depth value, and thickness value of the blood vessels).

[0062] During the training process, the input data is fed into the neural network in batches, and forward and backward propagation are performed. The weights of the network are adjusted by the gradient descent algorithm. Cross-validation or leave-one-out method is used to verify the generalization ability of the model.

[0063] The trained neural network model can accept the temperature time series and output the predicted parameters such as blood vessel temperature, depth, thickness, etc.

[0064] Based on the relationship between the temperature signal and the temperature response of the blood vessels, the neural network can estimate the temperature of the blood vessels. Usually, there is a certain spatial and temporal delay between the blood vessel temperature and the surface temperature.

[0065] By modeling the decay pattern of the temperature signal in time and the distribution pattern in space, the neural network can infer the depth and thickness of the blood vessels. This can rely on the spatial diffusion characteristics of the temperature signal. Deeper blood vessels will cause the temperature signal to be delayed or decay more slowly.

[0066] The embodiment of the present application can accurately obtain vascular parameters such as vascular temperature, depth and thickness of the target area by inputting the surface temperature signal into the neural network model without directly contacting or destroying the skin or vascular tissue, thereby realizing non-invasive vascular parameter measurement.

[0067] In the above Figure 1 Based on the above, the present application also provides a schematic diagram of a specific implementation flow of a method for measuring vascular parameters. Figure 2 As shown, the temperature time series sequence is input into a preset neural network model to obtain at least one vascular parameter of the vascular temperature, depth and thickness of the target area, including the following steps 201 to 203:

[0068] Step 201, obtaining low-level features and high-level features of a temperature time series.

[0069] Specifically, we first extract basic and intuitive temperature features from the temperature time series, including: temperature peak, which is the maximum value in the temperature signal within a preset time period; temperature valley, which is the minimum value in the temperature signal within a preset time period; and temperature difference, which is the temperature difference between adjacent acquisition points within a preset time period.

[0070] Through more complex signal processing methods, higher-level features are extracted from the temperature time series. These include: temperature distribution, the distribution of temperature over time, which reflects the thermal characteristics of blood vessels. Temperature change, analyzing the trend of temperature changes over time, revealing the thermal conduction characteristics of blood vessels. Thermal conduction characteristics, through thermal conduction models or thermal diffusion models, extract the pattern of temperature changes and their conduction characteristics in different materials (such as blood vessels, surrounding tissues).

[0071] Step 202: Fusing the low-level features and the high-level features to obtain processed fused features.

[0072] The extracted low-level features (such as temperature peaks, temperature valleys, temperature differences, etc.) are fused with high-level features (such as temperature distribution, temperature changes, heat conduction features, etc.). The following methods can be used, and this application does not limit them here. 1. Directly splice low-level features and high-level features in the feature dimension to form a richer feature vector. 2. According to the importance of different features, give each feature different weights, and fuse them after weighting. 3. Use a mapping method (such as principal component analysis PCA, etc.) to map low-level and high-level features into a unified feature space, and then perform subsequent processing.

[0073] Step 203: Input the fusion features into a preset neural network model to obtain at least one vascular parameter of the target area including vascular temperature, depth and thickness.

[0074] The fused feature vector is input into a preset neural network model, which is trained based on sample data from historical measurement scenarios.

[0075] This application outputs the vascular parameters of the target area based on the input fusion features through the residual network. The obtained vascular parameters include vascular temperature: the temperature level of the vascular area, which reflects the thermal characteristics of the blood vessel. Vascular depth: the position of the blood vessel from the surface, which is related to the attenuation of the temperature signal. Vascular thickness: the thickness of the vascular wall, which can be inferred from the thermal conductivity characteristics.

[0076] By fusing low-level and high-level features, the example of this application can more comprehensively and accurately reflect the temperature characteristics of the target area, enhance the expressiveness and adaptability of the neural network model, and thus improve the prediction accuracy and robustness of vascular parameters (such as vascular temperature, depth, and thickness).

[0077] The present application also provides a model training flow chart of a method for measuring vascular parameters. Figure 3 As shown, the plurality of sets of surface temperature signal sets include sets of surface temperature signal sets under different temperature collection scenarios, and the sample data collected under the historical measurement scenario are used to train the initial neural network model, including the following steps 301 to 306:

[0078] Step 301 : obtaining a first epidermal temperature, a second epidermal temperature, and a vascular parameter corresponding to the first epidermal temperature of a target area through a blood vessel simulation model, wherein the first epidermal temperature and the second epidermal temperature are both used to characterize the skin temperature of the target area.

[0079] Specifically, the blood vessel simulation model is used to simulate the relationship between the skin surface temperature and the blood vessel parameters. Through the simulation model, two sets of data can be obtained: a first epidermal temperature and the blood vessel parameters corresponding to the first epidermal temperature, and a second epidermal temperature.

[0080] Step 302 , obtaining a training set according to the second epidermal temperature and obtaining a validation set according to the first epidermal temperature and the blood vessel parameters corresponding to the first epidermal temperature.

[0081] Specifically, a training set is constructed based on the second epidermal temperature data, and this part of data is used to train the neural network model so that it can predict the blood vessel parameters related to the first epidermal temperature.

[0082] According to the first epidermal temperature and its corresponding vascular parameters, a validation set is constructed. The validation set is used to evaluate the generalization ability of the model in prediction and ensure that the model can effectively learn the relationship between vascular parameters and temperature signals.

[0083] Step 303, processing the initial neural network model according to the training set to obtain the training set blood vessel parameters.

[0084] The second epidermal temperature is input through the training set. The output of the model is the vascular parameters, that is, the vascular information related to the surface temperature, such as the temperature, depth, thickness, etc. of the blood vessels.

[0085] The model is trained through a neural network using the training set data. During training, an optimization algorithm (such as Adam) is used to minimize the loss function (such as mean square error) so that the network is as close to the true value as possible when predicting vascular parameters. During the training process, the weights and biases of the network are continuously adjusted to optimize the performance of the model.

[0086] Step 304, the initial neural network model is processed for accuracy according to the vascular parameters of the validation set and the training set, and a second neural network model is obtained when the accuracy meets a preset threshold.

[0087] The accuracy of the model is evaluated through the validation set. The performance of the model can be evaluated by calculating the loss function (such as mean square error MSE) and the evaluation index (such as R 2 ) to quantify. If the prediction results of the model meet the preset accuracy threshold (for example, the error on the training set and the validation set is lower than a certain value), the model is considered to have achieved the desired effect.

[0088] When the accuracy of the model meets a preset threshold, a second neural network model is obtained.

[0089] Step 305: input multiple groups of surface temperature signal sets into the second neural network model to obtain training results.

[0090] Multiple sets of surface temperature signal sets are input as input data into the second neural network model. The surface temperature signals under different scenarios will be processed by the network to predict vascular parameters.

[0091] The training results (i.e., the vascular parameters predicted by the model) are compared with the preset vascular parameters. At this time, the preset vascular parameters are target outputs set according to actual conditions or standard data.

[0092] The actual performance of the model is evaluated by comparing the model's prediction results with the preset results.

[0093] If the deviation between the training results and the preset vascular parameters is less than a preset threshold (such as the error is lower than a certain set tolerance), it is considered that the neural network model has achieved the expected performance and can be used in practical applications.

[0094] For example, when the deviation is less than 5%, it can be considered that the prediction effect of the model has reached the standard, the model has passed the verification, and meets the requirements of practical application.

[0095] Step 306, comparing the training result with the preset blood vessel parameters, and obtaining a preset neural network model when the deviation between the training result and the preset blood vessel parameters is less than a preset threshold.

[0096] After the model is tested and evaluated and meets the above deviation requirements, the second neural network model can be used as the final model for actual tasks. The model can continue to be used for prediction of new data and further optimization.

[0097] This application example can effectively improve the accuracy, efficiency and automation of vascular parameter prediction through optimized data collection, neural network training and verification mechanisms, and has good adaptability and generalization capabilities.

[0098] In the above Figure 1 Based on the above, the present application also provides a schematic diagram of a specific implementation flow of a method for measuring vascular parameters. Figure 4 As shown, obtaining a surface temperature signal set of a target area under laser irradiation within a preset time period includes the following steps 401 to 403:

[0099] Step 401, acquiring a plurality of temperature images of a target area under laser irradiation within a preset time.

[0100] During a preset period of time, a laser device is used to irradiate the target area. The surface temperature of the target area changes through thermal effects. Usually infrared lasers or other types of lasers are used. The surface temperature changes through the interaction between the laser and the target surface (such as absorption, reflection, transmission, etc.).

[0101] Under laser irradiation, multiple temperature images of the target area within a preset time period are obtained through an infrared thermal imaging camera or other temperature sensing equipment. The temperature image records the temperature distribution of the target area at different time points. Each temperature image reflects the temperature field of the target area under laser irradiation at a certain moment.

[0102] Step 402: extract multiple temperature signals from multiple temperature images.

[0103] Each temperature image is processed to extract the surface temperature signal of the target area. These temperature signals include the temperature value of each pixel, or can be aggregated according to the image area to obtain the average temperature of an area.

[0104] If the average temperature signal of a region is to be extracted, the target region may be divided into a plurality of small regions (eg, grid division), and then temperature data may be extracted from each small region.

[0105] Step 403: sort the multiple temperature signals according to a preset time sequence to obtain a surface temperature signal set.

[0106] All temperature signals acquired from different time points are sorted in a preset time sequence to form a temperature time series. Each temperature signal corresponds to a time point, and the temperature signals in the sequence should be arranged in the order of acquisition time so that the time characteristics of temperature changes can be reflected in subsequent analysis.

[0107] Finally, through the above steps, a surface temperature signal set is obtained, which is used as input data for subsequent analysis and model processing (such as training and reasoning in neural network models). These temperature signal sets are used to reflect the thermal response changes of the target area under laser irradiation and provide data support for the measurement of vascular parameters.

[0108] This application example can continuously monitor the temperature changes of the target area and obtain high-precision, real-time temperature data by acquiring multiple temperature images of the target area under laser irradiation. Each temperature image represents the temperature distribution at a specific time point, which can fully reflect the changes in the thermal effect of laser irradiation on the target area. Multiple temperature signals are sorted in chronological order to obtain a complete set of time series temperature signals. In this way, the temperature data is not just a snapshot at a single moment, but can fully reflect the dynamic process of the temperature change of the target area over time, which is helpful for in-depth analysis of the thermal response characteristics of the target area.

[0109] As an example, the wavelength of the laser is 532 nm and the measuring range is 0-100 degrees.

[0110] The laser wavelength is 532nm, which is a typical wavelength of green laser, commonly found in laser diodes or lasers. The 532nm wavelength laser light source has good beam quality, high brightness, and can effectively interact with many surface materials. The wavelength selection of the laser has a direct impact on the generation of thermal effects, the focusing performance of the beam, and the interaction mode with the target surface.

[0111] In practical applications, 532nm wavelength lasers can be effectively focused on the target area to produce uniform heat distribution, which is suitable for scenarios such as material surface heating and laser heat treatment.

[0112] The temperature measurement device or system used in the method of the embodiment of the present application can accurately measure the temperature range of the target area from 0°C to 100°C. In order to meet this requirement, it is necessary to select a thermal imaging device or infrared thermometer suitable for the low to medium temperature range. For example, an infrared thermal imager with a measurement range of 0-100°C can accurately capture the temperature changes of the target surface within this temperature range.

[0113] This application example uses a 532nm wavelength laser to provide high-precision temperature measurement, reduce external interference, ensure the accuracy of temperature data, is suitable for a temperature range of 0-100 degrees, meets the measurement needs of medium and low temperature ranges, and has broad application prospects.

[0114] As an example, multiple temperature images of the target area under laser irradiation within a preset time are obtained by an infrared detector.

[0115] Specifically, you first need to select a suitable infrared detector, and the operating wavelength range of the selected detector should match the radiation characteristics of the laser source. According to the wavelength of the laser (such as infrared laser), you can choose an infrared detector, such as a thermal imager or a specialized infrared detector array. These detectors can sense the temperature changes in the target area after laser irradiation and obtain temperature images at different time points.

[0116] The conditions of laser irradiation include laser power, duration, irradiation position and irradiation angle. These parameters determine the temperature change of the target area. The laser irradiation process may be carried out within a preset time period, and the infrared detector needs to continuously record the temperature change of the target area during this period.

[0117] During the laser irradiation process, the infrared detector will capture the temperature images of the target area at different time points. The temperature image can be generated by an infrared camera, and each pixel on the image represents the temperature of the area. In order to obtain multiple temperature images, the detector may take pictures at a certain time interval. For example, an image is taken every 1 second to record the temperature distribution changes of the target area under laser irradiation.

[0118] The infrared detector provided in the example of this application can measure the surface temperature of an object without contact and with high accuracy. Through infrared thermal imaging technology, the temperature distribution image of the target area under laser irradiation can be obtained, providing detailed temperature data for subsequent analysis. It ensures the accurate capture of the temperature change of the target area under laser irradiation, thereby improving the accuracy of the entire detection process.

[0119] As an example, the preset neural network model includes an input layer, a first convolutional layer, a pooling layer, fifteen residual layers, two fully connected layers, and an output layer. The residual layer includes two batch normalization layers, two rectified linear unit activation function layers, two second convolutional layers, and a shortcut connection.

[0120] Specifically, the input layer is responsible for receiving external input data. In this application example, the input data is a preprocessed temperature time series sequence. The original input data is passed to the subsequent layers in the neural network.

[0121] The first convolutional layer extracts local features through convolution operations. The main function of this layer is to identify basic local features (such as the change pattern of the temperature signal) from the input data. The convolution operation is performed on the input through the convolution kernel.

[0122] The pooling layer is used to reduce the spatial size of the feature map (such as dimensionality reduction), thereby reducing the amount of calculation, extracting more representative features, and avoiding overfitting. Through methods such as maximum pooling or average pooling, the feature map output by the convolutional layer is reduced to reduce the dimension of the data.

[0123] The fifteen residual layers introduce skip connections to alleviate the vanishing gradient problem during deep network training and enable the model to learn more complex features. The core idea of ​​residual learning is to add the input signal to the output signal through short-circuit connections so that information can flow through multiple layers more efficiently. Each residual layer contains multiple convolution operations and activation functions, which can help the model better capture and retain important features in the input signal.

[0124] Each residual layer contains the following sublayers:

[0125] 1. The two second convolutional layers are used to further extract deep features, usually used to process more complex features or perform more refined processing on the output of the previous layer.

[0126] 2. Two batch normalization layers are used to standardize the output of each layer to reduce internal covariate shift and improve training efficiency.

[0127] 3. The two rectified linear unit activation function layers increase the nonlinearity through the activation function, thereby improving the expressiveness of the model.

[0128] 4. Shortcut connections are used to skip the calculation of certain layers and pass the input directly to the next layer, which helps to reduce the difficulty of training deep neural networks.

[0129] The two fully connected layers integrate all the local features extracted previously and output the final prediction results. Each neuron in each layer is connected to all neurons in the previous layer.

[0130] Through multiple fully connected layers, the model is able to make final predictions (such as the temperature, depth, or thickness of a blood vessel) based on the extracted features.

[0131] The output layer is responsible for giving the final result of the model and outputting the vascular parameters (such as temperature, depth, thickness) of the target area. Based on the features extracted by the previous layer, the predicted values ​​of the vascular parameters are output.

[0132] The examples in this application can significantly improve the performance of neural networks in complex tasks by adopting designs such as multi-layer convolution, residual learning, batch normalization, and rectified linear unit activation functions. The depth and accuracy of feature extraction and learning are improved. The gradient problem during deep network training is reduced, and the stability and efficiency of training are enhanced. The nonlinear expression ability and computational efficiency of the network are improved. The generalization ability of the model is strengthened, enabling it to achieve better performance on more diverse tasks and data.

[0133] Based on the foregoing embodiments, an embodiment of the present application provides a device for measuring vascular parameters, which includes the modules included and the units included in the modules, and can be implemented by a processor; of course, it can also be implemented by a specific logic circuit; in the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP) or a field programmable gate array (FPGA), etc.

[0134] Figure 5 A schematic diagram of the structure of a device for measuring blood vessel parameters provided in an embodiment of the present application is shown in the figure. As shown in the figure, the device 500 includes an acquisition module 501, a processing module 502 and an input module 503, wherein:

[0135] An acquisition module 501 is used to acquire a set of surface temperature signals of a target area under laser irradiation within a preset time period;

[0136] The processing module 502 is used to pre-process the surface temperature signal set to obtain a temperature time series sequence;

[0137] The input module 503 is used to input the temperature time series sequence into a preset neural network model to obtain at least one vascular parameter of the target area including the vascular temperature, depth and thickness. The preset neural network model is obtained by training an initial neural network model based on sample data collected in historical measurement scenarios. The sample data includes multiple groups of temperature time series sequences obtained by preprocessing a set of surface temperature signals collected in historical measurement scenarios.

[0138] In some embodiments, the acquisition module 501 is further specifically used to acquire low-level features and high-level features of the temperature time series, wherein the low-level features include at least one of a temperature peak value, a valley value, and a temperature difference between adjacent acquisition points within a preset time period, and the high-level features include at least one of a temperature distribution, a temperature change, and a heat conduction feature;

[0139] The processing module 502 is further specifically used to fuse the low-level features and the high-level features to obtain the processed fused features;

[0140] The input module 503 is further specifically used to input the fusion feature into a preset neural network model to obtain at least one vascular parameter of the target area including the vascular temperature, depth and thickness.

[0141] In some embodiments, the acquisition module 501 is further specifically used to acquire a first epidermal temperature, a second epidermal temperature, and a vascular parameter corresponding to the first epidermal temperature of the target area through a vascular simulation model, wherein the first epidermal temperature and the second epidermal temperature are used to characterize the skin temperature of the target area;

[0142] The acquisition module 501 is further specifically used to obtain a training set according to the second skin temperature;

[0143] The acquisition module 501 is further specifically configured to obtain a verification set according to the first epidermal temperature and the blood vessel parameters corresponding to the first epidermal temperature;

[0144] The processing module 502 is further specifically used to train the initial neural network model according to the training set to obtain the training set blood vessel parameters;

[0145] The acquisition module 501 is further specifically used to calibrate the initial neural network model according to the vascular parameters of the validation set and the training set, and obtain the second neural network model when the accuracy of the vascular parameters of the training set output by the initial neural network model meets a preset threshold;

[0146] The input module 503 is further specifically used to input multiple sets of surface temperature signal sets into the second neural network model to obtain training results;

[0147] The acquisition module 501 is further specifically used to compare the training result with the preset blood vessel parameters, and obtain the preset neural network model when the deviation between the training result and the preset blood vessel parameters is less than a preset threshold.

[0148] In some embodiments, the acquisition module 501 is further specifically used to acquire multiple temperature images of the target area under laser irradiation within a preset time;

[0149] The acquisition module 501 is also specifically used to extract multiple temperature signals corresponding to the multiple temperature images;

[0150] The processing module 502 is further specifically configured to sort the multiple temperature signals according to the time sequence in which the corresponding multiple temperature images are acquired, so as to obtain a surface temperature signal set.

[0151] The description of the above device embodiment is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiment of the present application, please refer to the description of the method embodiment of the present application for understanding.

[0152] It should be noted that in the embodiments of this application Figure 5 The division of modules in the device for measuring vascular parameters shown is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional unit in each embodiment of the present application may be integrated into a processing unit, or may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of a software functional unit. It may also be implemented in the form of a combination of software and hardware.

[0153] It should be noted that in the embodiment of the present application, if the above method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the embodiment of the present application can be essentially or partly embodied in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including several instructions to enable an electronic device to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk. In this way, the embodiment of the present application is not limited to any specific combination of hardware and software.

[0154] The embodiment of the present application provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 6As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0155] An embodiment of the present application provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the method provided in the above embodiment are implemented.

[0156] An embodiment of the present application provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the steps of the method provided in the above method embodiment.

[0157] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0158] In one embodiment, a blood vessel parameter measuring device provided by the present application can be implemented in the form of a computer program. The computer program can be used in Figure 6 The computer device shown in the figure can be run. The memory of the computer device can store various program modules constituting the above-mentioned device. The computer program composed of various program modules enables the processor to execute the steps of the method of each embodiment of the present application described in this specification.

[0159] It should be noted here that the description of the above storage medium and device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects as the method embodiments. For technical details not disclosed in the storage medium, storage medium and device embodiments of this application, please refer to the description of the method embodiments of this application for understanding.

[0160] It should be understood that "one embodiment" or "an embodiment" or "some embodiments" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in one embodiment" or "in some embodiments" appearing throughout the specification may not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and 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 embodiment of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments. The above description of each embodiment tends to emphasize the differences between the various embodiments, and the same or similar aspects can be referenced to each other. For the sake of brevity, this article will not repeat them.

[0161] The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there may be three relationships. For example, object A and / or object B can represent three situations: object A exists alone, object A and object B exist at the same time, and object B exists alone.

[0162] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0163] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as: multiple modules or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or modules can be electrical, mechanical or other forms.

[0164] The modules described above as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules; they may be located in one place or distributed on multiple network units; some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0165] In addition, all functional modules in the embodiments of the present application may be integrated into one processing unit, or each module may be a separate unit, or two or more modules may be integrated into one unit; the above-mentioned integrated modules may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0166] A person skilled in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM), magnetic disks or optical disks, etc., various media that can store program codes.

[0167] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can essentially or in other words, the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling an electronic device to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0168] The methods disclosed in several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0169] The features disclosed in several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0170] The features disclosed in several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0171] The above is only an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for measuring vascular parameters, characterized in that: include: Acquire a surface temperature signal set of the target area under laser irradiation within a preset time period; Preprocessing the surface temperature signal set to obtain a temperature time series sequence; The temperature time series sequence is input into a preset neural network model to obtain at least one vascular parameter of the target area including vascular temperature, depth and thickness. The preset neural network model is obtained by training an initial neural network model based on sample data collected in historical measurement scenarios. The sample data includes multiple groups of temperature time series sequences obtained by preprocessing a set of surface temperature signals collected in historical measurement scenarios.

2. The method according to claim 1, characterized in that The step of inputting the temperature time series into a preset neural network model to obtain at least one vascular parameter of the target area's vascular temperature, depth, and thickness includes: Acquire low-level features and high-level features of the temperature time series, wherein the low-level features include at least one of a temperature peak value, a valley value, and a temperature difference between adjacent acquisition points within a preset time period, and the high-level features include at least one of a temperature distribution, a temperature change, and a heat conduction feature; Fusing the low-level features and the high-level features to obtain processed fused features; The fusion feature is input into a preset neural network model to obtain at least one vascular parameter of the target area including vascular temperature, depth and thickness.

3. The method according to claim 1, characterized in that The plurality of sets of surface temperature signal collections include surface temperature signal collections in different temperature collection scenarios; The initial neural network model is trained using sample data collected in historical measurement scenarios, including: Acquire a first epidermal temperature, a second epidermal temperature, and a vascular parameter corresponding to the first epidermal temperature of a target area through a blood vessel simulation model, wherein the first epidermal temperature and the second epidermal temperature are used to characterize the skin temperature of the target area; Obtaining a training set according to the second skin temperature; Obtaining a validation set according to the first epidermal temperature and a blood vessel parameter corresponding to the first epidermal temperature; Training the initial neural network model according to the training set to obtain training set vascular parameters; Calibrate the initial neural network model according to the validation set and the training set vascular parameters, and obtain a second neural network model when the accuracy of the training set vascular parameters output by the initial neural network model meets a preset threshold; Inputting the plurality of sets of surface temperature signal sets into the second neural network model to obtain training results; The training result is compared with the preset blood vessel parameters, and when the deviation between the training result and the preset blood vessel parameters is less than a preset threshold, a preset neural network model is obtained.

4. The method according to claim 1, characterized in that: The step of obtaining a surface temperature signal set of a target area under laser irradiation within a preset time period includes: Acquire multiple temperature images of the target area under laser irradiation within a preset time; Extracting a plurality of temperature signals corresponding to the plurality of temperature images; The plurality of temperature signals are sorted according to the time sequence in which the corresponding plurality of temperature images are acquired to obtain the surface temperature signal set.

5. The method according to claim 4, characterized in that The wavelength of the laser is 532 nm, and the measuring range is 0-100 degrees.

6. The method according to claim 4, characterized in that The multiple temperature images of the target area under laser irradiation within the preset time are obtained by an infrared detector.

7. The method according to claim 3, characterized in that The preset neural network model includes an input layer, a first convolutional layer, a pooling layer, fifteen residual layers, two fully connected layers and an output layer, and the residual layer includes two batch normalization layers, two rectified linear unit activation function layers, two second convolutional layers and a shortcut connection.

8. A device for measuring blood vessel parameters, characterized in that: include: An acquisition module, used to acquire a set of surface temperature signals of a target area under laser irradiation within a preset time period; A processing module, used for preprocessing the surface temperature signal set to obtain a temperature time series sequence; An input module is used to input the temperature time series sequence into a preset neural network model to obtain at least one vascular parameter of the target area, including vascular temperature, depth and thickness. The preset neural network model is obtained by training an initial neural network model based on sample data collected in historical measurement scenarios, and the sample data includes multiple groups of surface temperature signal sets collected in historical measurement scenarios.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, wherein: When the processor executes the program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images