An absolute wide range blood flow measurement method, apparatus and device

Abstract

The application relates to the biomedical field and provides an absolute wide-range blood flow measurement method, device and equipment.The method comprises the following steps: collecting speckle image data containing blood flow information through a laser speckle imaging device; pre-processing the speckle image data to extract the time intensity fluctuation frequency characteristics of the speckle; inputting the pre-processed speckle image data into a trained laser speckle contrast imaging model based on deep learning, and predicting the absolute blood flow velocity based on the time intensity fluctuation frequency characteristics of the speckle; and the trained laser speckle contrast imaging model based on deep learning is obtained according to a training set, and the training set is the pre-processed speckle image data.The application solves the problem that the blood flow measurement range is limited in the prior art, and realizes wide-range absolute blood flow velocity measurement.

Description

Technical Field

[0001] This invention relates to the field of biomedical technology, and in particular to a method, apparatus and device for measuring blood flow over an absolute wide range. Background Technology

[0002] Currently, commonly used techniques for blood flow measurement mainly include Transit Time Flow Measurement (TTFM) and Laser Speckle Contrast Imaging (LSCI). TTFM, as the standard method for evaluating cardiac blood flow, measures blood flow by measuring the difference in ultrasound propagation time. However, in complex surgeries such as coronary artery bypass grafting (CABG), TTFM can only perform point-to-point measurements on the dissected graft, making it difficult to comprehensively capture the spatiotemporal evolution of blood flow in the region distal to cardiac stenosis.

[0003] On the other hand, LSCI technology infers blood flow characteristics by analyzing the dynamic changes in laser speckle caused by erythrocyte movement. This technology boasts advantages such as high spatiotemporal resolution, ease of operation, and low cost, and is therefore widely used in blood flow monitoring of superficial tissues (such as the brain, skin, and retina). However, LSCI technology primarily relies on relative blood flow index (rBFI) or speckle contrast (K) as surrogate indicators of blood perfusion. These indicators lack the ability to directly measure absolute blood flow velocity and perform poorly in monitoring high-velocity flows (such as myocardial coronary blood flow). Furthermore, these indicators are affected by various parameters (such as the proportion of static scattering in the sample, the configuration of the experimental equipment, and the scattering characteristics of the medium), making it difficult to obtain consistent absolute blood flow velocities across different samples. Summary of the Invention

[0004] This invention provides an absolute wide-range blood flow measurement method, apparatus, and equipment, which solves the problem of limited blood flow measurement range in the prior art and realizes wide-range absolute blood flow velocity measurement.

[0005] This invention provides an absolute wide-range blood flow measurement method, comprising the following steps:

[0006] The speckle image data containing blood flow information is acquired using a laser speckle imaging device;

[0007] The speckle image data is preprocessed to extract the temporal intensity fluctuation frequency characteristics of the speckle.

[0008] The preprocessed speckle image data is input into a trained deep learning-based laser speckle contrast imaging model, and the absolute blood flow velocity is predicted based on the temporal intensity fluctuation frequency characteristics of the speckle.

[0009] The trained deep learning-based laser speckle contrast imaging model is obtained by training a training set, which is preprocessed speckle image data.

[0010] According to the present invention, an absolute wide-range blood flow measurement method is provided, wherein the laser speckle imaging device includes a near-infrared laser, a camera, a filter, a polarizer, a sleeve lens, and a microscope objective; the camera is used to acquire speckle images of the sample to be tested; the filter, polarizer, sleeve lens, and microscope objective are arranged coaxially in sequence; the filter is used to filter the laser, the polarizer is used to polarize the laser, the sleeve lens and microscope objective form a microscope system for magnifying the speckle image of the sample to be tested, and the near-infrared laser is used to irradiate the sample to be tested with near-infrared laser light, adjusting the light intensity so that the average gray value of the speckle image is 105.

[0011] According to the present invention, an absolute wide-range blood flow measurement method is provided, wherein the speckle image data is preprocessed, the speckle image video is divided into multiple segments containing consecutive frames, and the pixel values ​​of the segments are adjusted to preset values ​​as input data for a deep learning-based laser speckle contrast imaging model.

[0012] According to the present invention, an absolute wide-range blood flow measurement method is provided, wherein the trained deep learning-based laser speckle contrast imaging model includes: multiple convolutional and pooling modules connected in sequence for feature extraction from input data; a fully connected layer connected to the last convolutional and pooling module for outputting the probability distribution of blood flow categories; the deep learning-based laser speckle contrast imaging model uses ReLU as the activation function and Adam as the optimizer to accelerate the convergence speed of the model.

[0013] According to the present invention, an absolute wide-range blood flow measurement method is provided, wherein the prediction of absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle image specifically includes: extracting spatiotemporal features from the speckle image layer by layer through multiple convolution and pooling modules; after feature extraction, passing the feature map to a fully connected layer, integrating and transforming the extracted features through the fully connected layer, and finally outputting a probability distribution of the corresponding blood flow velocity category; and determining the predicted absolute blood flow velocity value by searching a predefined mapping relationship or probability threshold according to the output probability distribution.

[0014] According to the present invention, an absolute wide-range blood flow measurement method is provided, which trains a deep learning-based laser speckle contrast imaging model based on a training set. Specifically, the method includes: inputting preprocessed training set data into the deep learning-based laser speckle contrast imaging model; adjusting the model parameters through a backpropagation algorithm to minimize the loss function of the deep learning-based laser speckle contrast imaging model on the training set; repeating the training process until the model converges or reaches a predetermined number of training rounds, and outputting the trained deep learning-based laser speckle contrast imaging model.

[0015] The present invention also provides an absolute wide-range blood flow measurement device, comprising the following modules:

[0016] The acquisition module is used to acquire speckle image data containing blood flow information through a laser speckle imaging device;

[0017] The preprocessing module is used to preprocess the speckle image data to extract the temporal intensity fluctuation frequency characteristics of the speckle.

[0018] The prediction module is used to input the preprocessed speckle image data into a trained deep learning-based laser speckle contrast imaging model, and predict the absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle.

[0019] The trained deep learning-based laser speckle contrast imaging model is obtained by training a training set, which is preprocessed speckle image data.

[0020] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the absolute wide-range blood flow measurement method as described above.

[0021] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the absolute wide-range blood flow measurement method as described above.

[0022] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the absolute wide-range blood flow measurement method as described above.

[0023] The present invention provides an absolute wide-range blood flow measurement method, device and equipment, which has the following beneficial effects: by combining laser speckle imaging and deep learning technology, blood flow information is captured by a laser speckle imaging device, and after preprocessing, the absolute blood flow velocity is predicted by a trained deep learning model, thereby realizing rapid and accurate measurement of blood flow velocity. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0025] Figure 1 This is a flowchart illustrating the absolute wide-range blood flow measurement method provided by the present invention.

[0026] Figure 2 This is a schematic diagram of the structure of the laser speckle imaging device provided by the present invention.

[0027] Figure 3 This is the DL-LSCI model structure provided by the present invention.

[0028] Figure 4 This is a schematic diagram of the prediction results of DL-LSCI in body membrane experiments provided by the present invention.

[0029] Figure 5 This is a schematic diagram illustrating the prediction results of DL-LSCI in animal experiments provided by this invention.

[0030] Figure 6 This is a schematic diagram of the structure of the absolute wide-range blood flow measurement device provided by the present invention.

[0031] Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention.

[0032] Figure label:

[0033] 1. Near-infrared laser; 2. Camera; 3. Filter; 4. Polarizer; 5. Sleeve lens; 6. Microscope objective; 7. Sample to be tested. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0035] Currently, the commonly used techniques for blood flow measurement are Transit-Time Flow Measurement (TTFM) and Laser Speckle Contrast Imaging (LSCI). TTFM is the standard method for evaluating cardiac blood flow, measuring blood flow through the difference in ultrasound propagation time. However, in coronary artery bypass grafting (CABG), TTFM can only perform point-to-point measurements on the dissected graft, making it difficult to capture the spatiotemporal evolution of blood flow in the distal region of cardiac stenosis, which is crucial for improving surgical quality control. LSCI-based blood flow measurement infers blood flow characteristics by analyzing the dynamic changes in laser speckle caused by red blood cell movement. Due to its advantages of high spatiotemporal resolution, simple operation, and low cost, it is widely used for monitoring blood flow in superficial tissues (such as the brain, skin, and retina). This method mainly relies on the relative blood flow index (rBFI) or speckle contrast (K) as substitute indicators for blood perfusion, lacking the ability to measure absolute blood flow velocity and having difficulty monitoring high-velocity blood flow velocities in the myocardium and coronary arteries. Furthermore, these indicators are affected by parameters... β , ρ and n The effects of static scattering on blood flow velocity are related to the proportion of static scattering in the sample, the configuration of the experimental equipment, and the scattering characteristics of the medium, making it difficult to obtain consistent absolute blood flow velocities across different samples. In recent years, to improve the accuracy of blood flow measurement and obtain absolute blood flow velocities in LSCI, many researchers have attempted to combine deep learning techniques to improve existing methods. Hao et al. used a 3D convolutional neural network to achieve quantitative predictions in the range of 0.08–10.74 mm / s, but the estimation range was limited and performance was poor under static scattering conditions. Chen et al. developed a deep learning method based on UNet and ResNet to reconstruct the 3D blood flow structure in thick tissues, enabling depth-dependent flow estimation and reducing bias caused by multiple scattering. However, while these techniques significantly improve the performance of LSCI, they still cannot provide true absolute blood flow velocity measurements in standardized and clinical applications.

[0036] To address the aforementioned issues, this invention proposes a deep learning-based laser speckle contrast imaging model (DL-LSCI). This model predicts absolute blood flow velocity by analyzing the spatiotemporal frequency characteristics of speckle patterns in different blood flows, aiming to obtain a wide range of absolute blood flow velocities. The invention mainly comprises two parts: a laser speckle imaging device for data acquisition and a DL-LSCI model structure for predicting absolute blood flow velocities. This method is applicable not only to ideal conditions without static scattering but also to static scattering conditions introduced by scatterers or biological blood vessels, accurately predicting actual blood flow velocities up to 308 mm / s. Experimental verification shows that this scheme exhibits excellent accuracy and robustness in in vivo experiments, providing consistent quantitative results for hemodynamic studies under various experimental conditions.

[0037] The following is combined with Figures 1-7 The embodiments of the present invention are described in detail.

[0038] Figure 1 This is a flowchart illustrating the absolute wide-range blood flow measurement method provided by the present invention, as shown below. Figure 1 As shown, the method includes the following steps:

[0039] S110. Acquire speckle image data containing blood flow information using a laser speckle imaging device.

[0040] According to the present invention, an absolute wide-range blood flow measurement method includes a laser speckle imaging device comprising a near-infrared laser, a camera, a filter, a polarizer, a sleeve lens, and a microscope objective. The camera is used to acquire speckle images of the sample to be tested. The filter, polarizer, sleeve lens, and microscope objective are arranged coaxially in sequence. The filter is used to filter the laser, the polarizer is used to polarize the laser, and the sleeve lens and microscope objective form a microscope system for magnifying the speckle image of the sample to be tested. The near-infrared laser is used to irradiate the sample to be tested with near-infrared laser light, and the light intensity is adjusted so that the average gray value of the speckle image is 105.

[0041] Specifically, such as Figure 2 The diagram shows a laser speckle imaging device, which includes a near-infrared laser 1, a camera 2, a filter 3, a polarizer 4, a sleeve lens 5, a microscope objective 6, and a sample to be tested 7.

[0042] Near-infrared laser 1 serves as a light source, generating a stable near-infrared laser beam to provide illumination for imaging.

[0043] Camera 2 is positioned in the transmission path of the laser beam to capture the laser speckle image scattered back from the sample. Camera 2 is arranged at an angle to the laser beam to ensure that the speckle pattern can be clearly received.

[0044] Filter 3 is located in front of the camera and is used to filter out unwanted spectral components, allowing only near-infrared light of specific wavelengths to pass through, thereby enhancing image quality.

[0045] The polarizer is located in the optical path in front of the camera to adjust the polarization state of light, reduce interference from reflected and scattered light, and improve the signal-to-noise ratio of the image.

[0046] The sleeve lens 5 and the microscope objective 6 are located between the camera 2 and the sample 7 under test. Together, they form a microscope system used to magnify the speckle image, enabling the camera to capture finer speckle structures. The focal length and magnification of the microscope system are adjusted according to specific application requirements.

[0047] The sample to be tested, 7, is located at the bottom of the device and is the target object of laser speckle imaging, such as blood vessels in biological tissue.

[0048] The aforementioned components work together, with the laser beam generated by the near-infrared laser 1 directly illuminating the sample 7 under test. The laser light scattered back from the sample is amplified by the sleeve lens 5 and the microscope objective 6, then passes sequentially through the polarizer 4 and the filter 3 before finally entering the camera 2 for imaging. Each optical element (filter 3, polarizer 4, sleeve lens 5, microscope objective 6) is arranged sequentially along the transmission path of the laser beam, ensuring an unobstructed optical path while each performs its specific optical function. The camera 2 is located on one side of the entire device, at a certain angle to the laser beam, to capture the scattered laser speckle image. This layout and design enable the laser speckle imaging device to effectively capture and analyze blood flow information in the sample, providing high-quality image data for subsequent data processing and absolute blood flow velocity measurement.

[0049] In one embodiment of the present invention, the exposure time of the laser speckle imaging device is first standardized to 80°C. μ m, sampling frequency of 100 Hz, blood flow data from 0-462 mm / s were collected by laser speckle imaging device. Each set of data consists of 200 consecutive speckle patterns. The intensity fluctuation of speckle patterns varies over time at different flow velocities. The lower the blood flow velocity, the more intense the intensity fluctuation.

[0050] S120. Preprocess the speckle image data to extract the temporal intensity fluctuation frequency characteristics of the speckle.

[0051] According to the present invention, an absolute wide-range blood flow measurement method preprocesses speckle image data, divides the speckle image video into multiple segments containing consecutive frames, and adjusts the pixel values ​​of the segments to preset values ​​as input data for a deep learning-based laser speckle contrast imaging model.

[0052] Specifically, to ensure the extraction of high signal-to-noise ratio blood flow information from the speckle pattern, the average grayscale value of the speckle image was stabilized at 105 by adjusting the laser intensity throughout the experiment, keeping most pixel values ​​within the 8-bit dynamic range. Therefore, this invention can utilize the temporal intensity fluctuation frequency characteristics of the speckle as an important basis for estimating blood flow in the DL-LSCI model. Furthermore, to prevent overfitting of the deep learning model, this invention expands the dataset using data augmentation methods, segmenting the speckle image video into multiple segments, each containing 25 consecutive frames with a step size of 1. For example, frames 1-25 are the first segment, frames 2-26 are the second segment, and so on until the last frame. This yields 176 segments per flow velocity, totaling 3520 segments, which are then divided into training and test sets in a 7:3 ratio. Each segment contains similar speckle variation information in the spatiotemporal dimension, while also exhibiting slight differences. These videos are then resized to 112×112 pixels and used as input to the DL-LSCI model. The above methods ensure the sufficiency of data and the effectiveness of feature learning during model training, thereby improving the accuracy of absolute blood flow velocity estimation.

[0053] S130. Input the preprocessed speckle image data into the trained deep learning-based laser speckle contrast imaging model, and predict the absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle. The trained deep learning-based laser speckle contrast imaging model is obtained by training the preprocessed speckle image data.

[0054] According to the absolute wide-range blood flow measurement method provided by the present invention, the trained deep learning-based laser speckle contrast imaging model includes: multiple convolutional and pooling modules connected in sequence for feature extraction of input data; a fully connected layer connected to the last convolutional and pooling module for outputting the probability distribution of blood flow categories; the deep learning-based laser speckle contrast imaging model uses ReLU as the activation function and Adam as the optimizer to accelerate the convergence speed of the model.

[0055] According to the present invention, an absolute wide-range blood flow measurement method is provided, which trains a deep learning-based laser speckle contrast imaging model based on a training set. Specifically, the method includes: inputting preprocessed training set data into the deep learning-based laser speckle contrast imaging model; adjusting the model parameters through a backpropagation algorithm to minimize the loss function of the deep learning-based laser speckle contrast imaging model on the training set; repeating the training process until the model converges or reaches a predetermined number of training rounds, and outputting the trained deep learning-based laser speckle contrast imaging model.

[0056] According to the present invention, an absolute wide-range blood flow measurement method predicts absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of speckle. Specifically, the method includes: extracting spatiotemporal features from a speckle image layer by layer through multiple convolution and pooling modules; after feature extraction, passing the feature map to a fully connected layer, integrating and transforming the extracted features through the fully connected layer, and finally outputting a probability distribution of the corresponding blood flow velocity category; and determining the predicted absolute blood flow velocity value by searching a predefined mapping relationship or probability threshold based on the output probability distribution.

[0057] Specifically, such as Figure 3 As shown, the DL-LSCI model is constructed next. After preprocessing such as adjustment and augmentation, the data size is 25×112×112, which is then passed through five convolutional and pooling modules. In the first and second convolutional and pooling modules, one convolution and one pooling operation are performed, with a convolution kernel size of 3×3×3 and a stride of 1×1×1. The pooling layer size in the first convolutional and pooling module is 1×2×2, and the pooling layer size in the second convolutional and pooling module is 2×2×2. In the third, fourth, and fifth convolutional and pooling modules, two convolutions and one pooling operation are performed, with the convolution kernel size still being 3×3×3 and the stride 1×1×1. The pooling layer size is 2×2×2, resulting in a final feature map size of 512×1×4×4. Finally, the feature map is passed through two fully connected layers, each with 4096 neurons, and the output is the probability distribution of blood flow categories. In this DL-LSCI model, the activation function is ReLU, the optimizer is Adam to accelerate convergence, and the learning rate is 5×10⁻⁶. -5 The batch size is 32. This invention simplifies the blood flow estimation problem into a classification problem, thereby effectively improving the accuracy of absolute blood flow velocity estimation and the convergence performance of the model.

[0058] To verify the effectiveness of the DL-LSCI model in this invention, body membrane experiments and animal experiments were conducted. In the body membrane experiments, blood flow velocity was controlled by a stepper motor within the range of 0 mm / s to 462 mm / s, and corresponding speckle data were collected. Data were collected under four conditions: blood flow in a transparent microfluidic cavity, blood flow in a microfluidic cavity covered with a 220-mesh scattering sheet, blood flow in simulated blood vessels in the coronary artery and internal mammary artery of pigs. Using the above data as a dataset, the flow velocity was predicted using DL-LSCI, and the results are as follows. Figure 4As shown in the figure, the values ​​represent the prediction probability at different flow velocities. A higher probability value indicates more accurate DL-LSCI prediction. At low flow velocities, the model's prediction accuracy is above 90%. When the blood flow velocity exceeds 100 mm / s, the probability drops to 70%-90%, but it can still correctly predict the flow velocity without affecting overall accuracy. However, when the flow velocity increases to 462 mm / s, the prediction probability drops below 50%, indicating that DL-LSCI cannot effectively estimate a blood flow velocity of 462 mm / s. This is because as the blood flow velocity increases, the prediction probability decreases at exposure times (80 minutes). μ The DL-LSCI model integrates more spatiotemporal information of the speckle pattern, smoothing out dynamic fluctuations and reducing differences between high flow velocities. Therefore, in models with and without a covered medium and with a covered speckle, DL-LSCI can predict blood flow velocities from 0 to 385 mm / s, while for blood samples covered by the other two types of tissue, the range is 0 to 308 mm / s. These differences arise from different static scattering components. Lower static scattering components allow the speckle pattern to contain more blood flow information, i.e., a higher signal-to-noise ratio. Therefore, it can be concluded that the DL-LSCI model can achieve an absolutely wide range of blood flow measurements from 0 mm / s to 308 mm / s, which is significant for significantly improving the accuracy of LSCI theory. In animal experiments, an infarction / recovery model was constructed in the rat carotid artery to obtain different flow velocities. Blood flow was simultaneously recorded using TTFM and the method of this invention, and the results are as follows: Figure 5 As shown. The horizontal axis represents the blood flow velocity measured by TTFM, and the vertical axis represents the blood flow velocity predicted by DL-LSCI. Smaller points represent trained samples, and larger points represent samples outside the dataset to be evaluated. The dashed line represents... y = x The reference line is used to compare the blood flow velocity measured by TTFM with the blood flow velocity predicted by the model. The table on the right shows the probability values ​​of the blood flow prediction; all probability values ​​exceed 92%. Therefore, it can be concluded that the DL-LSCI model can achieve absolute blood flow measurements of 13–105 mm / s in animal experiments.

[0059] This invention provides a deep learning-based laser speckle contrast imaging model (DL-LSCI) capable of estimating a wide range of absolute blood flow velocities. DL-LSCI achieves absolute blood flow velocity measurement by combining three-dimensional convolutional layers with laser speckle characteristics, eliminating the parameter dependence of traditional LSCI techniques. The model accurately predicts a wide range of blood flow velocities and demonstrates superior performance in both in vivo membrane and animal experiments. Compared to the clinical gold standard TTFM, the DL-LSCI model improves the accuracy of blood flow measurement in laser speckle imaging equipment, possessing potential applications in demanding surgeries such as coronary artery bypass grafting (CABG), enabling regional blood flow imaging visualization, and promoting the development of surgical quality control and biomedical research.

[0060] The absolute wide-range blood flow measurement device provided by the present invention is described below. The absolute wide-range blood flow measurement device described below and the absolute wide-range blood flow measurement method described above can be referred to in correspondence.

[0061] like Figure 6 The image shows an absolute wide-range blood flow measurement device provided by the present invention, comprising:

[0062] Acquisition module 610 is used to acquire speckle image data containing blood flow information through a laser speckle imaging device;

[0063] The preprocessing module 620 is used to preprocess the speckle image data to extract the temporal intensity fluctuation frequency characteristics of the speckle.

[0064] The prediction module 630 is used to input the preprocessed speckle image data into the trained deep learning-based laser speckle contrast imaging model and predict the absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle.

[0065] The trained deep learning-based laser speckle contrast imaging model is obtained by training the preprocessed speckle image data.

[0066] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include a processor 710, a communications interface 720, a memory 730, and a communication bus 740, wherein the processor 710, communications interface 720, and memory 730 communicate with each other via the communication bus 740. The processor 710 can call logic instructions in the memory 730 to execute an absolute wide-range blood flow measurement method. This method includes: acquiring speckle image data containing blood flow information using a laser speckle imaging device; preprocessing the speckle image data to extract the temporal intensity fluctuation frequency characteristics of the speckle; inputting the preprocessed speckle image data into a trained deep learning-based laser speckle contrast imaging model to predict the absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle; the trained deep learning-based laser speckle contrast imaging model is obtained by training a training set, which is the preprocessed speckle image data.

[0067] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0068] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the absolute wide-range blood flow measurement method provided by the above methods. The method includes: acquiring speckle image data containing blood flow information through a laser speckle imaging device; preprocessing the speckle image data to extract the temporal intensity fluctuation frequency characteristics of the speckle; inputting the preprocessed speckle image data into a trained deep learning-based laser speckle contrast imaging model to predict the absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle; the trained deep learning-based laser speckle contrast imaging model is trained according to a training set, which is the preprocessed speckle image data.

[0069] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the absolute wide-range blood flow measurement method provided by the methods described above. This method includes: acquiring speckle image data containing blood flow information using a laser speckle imaging device; preprocessing the speckle image data to extract the temporal intensity fluctuation frequency characteristics of the speckle; inputting the preprocessed speckle image data into a trained deep learning-based laser speckle contrast imaging model to predict absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle; the trained deep learning-based laser speckle contrast imaging model is obtained by training a training set, where the training set is the preprocessed speckle image data.

[0070] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0071] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of various embodiments or some parts of embodiments.

[0072] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for measuring blood flow over an absolutely wide range, characterized in that, include: Speckle image data containing blood flow information was acquired using a laser speckle imaging device; the exposure time of the unified laser speckle imaging device was 80 μm, the sampling frequency was 100 Hz, and blood flow data from 0 to 462 mm / s were acquired using the laser speckle imaging device. The speckle image data is preprocessed to extract the temporal intensity fluctuation frequency characteristics of the speckle. The preprocessed speckle image data is input into a trained deep learning-based laser speckle contrast imaging model, and the absolute blood flow velocity is predicted based on the temporal intensity fluctuation frequency characteristics of the speckle. The trained deep learning-based laser speckle contrast imaging model is obtained by training a training set, which is preprocessed speckle image data. The laser speckle imaging device includes a near-infrared laser, a camera, a filter, a polarizer, a sleeve lens, and a microscope objective. The camera is used to acquire speckle images of the sample under test. The filter, polarizer, sleeve lens, and microscope objective are arranged coaxially in sequence. The filter is used to filter the laser, the polarizer is used to polarize the laser, and the sleeve lens and microscope objective form a microscope system for magnifying the speckle image of the sample under test. The near-infrared laser is used to irradiate the sample under test with near-infrared laser light, and the light intensity is adjusted so that the average gray value of the speckle image is 105.

2. The absolute wide-range blood flow measurement method according to claim 1, characterized in that, The speckle image data is preprocessed by dividing the speckle image video into multiple segments containing consecutive frames, and adjusting the pixel values ​​of the segments to preset values, which are then used as input data for a deep learning-based laser speckle contrast imaging model.

3. The absolute wide-range blood flow measurement method according to claim 1, characterized in that, The trained deep learning-based laser speckle contrast imaging model includes: Multiple convolutional and pooling modules are connected in sequence to extract features from the input data; The fully connected layer, connected to the last convolutional and pooling module, is used to output the probability distribution of blood flow categories; The deep learning-based laser speckle contrast imaging model uses ReLU as the activation function and Adam as the optimizer to accelerate the model's convergence speed.

4. The absolute wide-range blood flow measurement method according to claim 3, characterized in that, The prediction of absolute blood flow velocity based on the time intensity fluctuation frequency characteristics of the speckle pattern specifically includes: Spatiotemporal features in speckle images are extracted layer by layer using multiple convolution and pooling modules; After feature extraction, the feature map is passed to the fully connected layer. The fully connected layer integrates and transforms the extracted features, and finally outputs the probability distribution of the corresponding blood flow velocity category. Based on the output probability distribution, the predicted absolute blood flow velocity value is determined by looking up a predefined mapping relationship or probability threshold.

5. The absolute wide-range blood flow measurement method according to claim 1, characterized in that, Training a deep learning-based laser speckle contrast imaging model based on the training set includes: The preprocessed training set data is input into a deep learning-based laser speckle contrast imaging model; The model parameters are adjusted using the backpropagation algorithm to minimize the loss function of the deep learning-based laser speckle contrast imaging model on the training set. Repeat the training process until the model converges or reaches the predetermined number of training rounds, and output the trained deep learning-based laser speckle contrast imaging model.

6. An absolute wide-range blood flow measurement device, characterized in that, include: The acquisition module is used to acquire speckle image data containing blood flow information through a laser speckle imaging device; wherein, the exposure time of the unified laser speckle imaging device is 80μm, the sampling frequency is 100 Hz, and blood flow data of 0-462 mm / s is acquired through the laser speckle imaging device. The preprocessing module is used to preprocess the speckle image data to extract the temporal intensity fluctuation frequency characteristics of the speckle. The prediction module is used to input the preprocessed speckle image data into a trained deep learning-based laser speckle contrast imaging model, and predict the absolute blood flow velocity based on the temporal intensity fluctuation frequency characteristics of the speckle. The trained deep learning-based laser speckle contrast imaging model is obtained by training a training set, which is preprocessed speckle image data. The laser speckle imaging device includes a near-infrared laser, a camera, a filter, a polarizer, a sleeve lens, and a microscope objective. The camera is used to acquire speckle images of the sample under test. The filter, polarizer, sleeve lens, and microscope objective are arranged coaxially in sequence. The filter is used to filter the laser, the polarizer is used to polarize the laser, and the sleeve lens and microscope objective form a microscope system for magnifying the speckle image of the sample under test. The near-infrared laser is used to irradiate the sample under test with near-infrared laser light, and the light intensity is adjusted so that the average gray value of the speckle image is 105.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the absolute wide-range blood flow measurement method as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the absolute wide-range blood flow measurement method as described in any one of claims 1 to 5.

9. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the absolute wide-range blood flow measurement method as described in any one of claims 1 to 5.

Images