Biological electromagnetic model establishing method for physiological parameter monitoring

Abstract

The invention relates to the technical field of image processing, in particular to a biological electromagnetic model establishing method for physiological parameter monitoring, and a biological electromagnetic model is established according to five steps of data collection, image segmentation, three-dimensional reconstruction, smoothing and electromagnetic characteristic parameter setting. According to the method, the contour connection algorithm is adopted, the two-dimensional mask subjected to image segmentation is subjected to three-dimensional reconstruction, the surface mesh model is established by using real biological data, and the precision degree of the model is improved. A Laplace smoothing algorithm is adopted to smooth the surface of the model, electromagnetic wave reflection on the surface of the model is reduced, and the sensitivity of the model is improved. And a high-order interpolation method is combined to fit a biological tissue dielectric constant curve, so that accurate electromagnetic simulation of experimental organisms is realized. The problem that an existing electromagnetic model is difficult to accurately describe a complex biological tissue structure is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method for establishing a bio-electromagnetic model for monitoring physiological parameters. Background Art

[0002] In biological wearable devices, radio frequency sensors based on radio frequency technology are gradually being widely used in our daily lives due to their advantages such as non-invasiveness, high sensitivity, and miniaturization. In the design process of radio frequency sensors, simulation based on electromagnetic models is crucial. Accurate electromagnetic models can make the simulation closer to the actual situation and enable radio frequency sensors to achieve better working conditions. Previous studies have mostly used simple layered models, cylindrical models, and spherical models. Such regular geometric models are often difficult to accurately describe complex biological tissue structures. At the same time, in the existing bio-electromagnetic models, the parameters of each material are fixed, which makes it difficult to meet the needs of changing parameters in experiments. To this end, we propose a method for establishing a bio-electromagnetic model for physiological parameter monitoring. Summary of the invention

[0003] Based on the technical problems existing in the background technology, the present invention proposes a method for establishing a biological electromagnetic model for physiological parameter monitoring, which improves the accuracy of the model, reduces the reflection of electromagnetic waves on the surface of the model, and improves the sensitivity of the model. Combined with the high-order interpolation method to fit the dielectric constant curve of biological tissue, accurate electromagnetic simulation of experimental organisms is achieved, solving the problem that the existing electromagnetic model is difficult to accurately describe the complex biological tissue structure.

[0004] The present invention provides the following technical solution: a method for establishing a bioelectromagnetic model for monitoring physiological parameters, comprising the following steps:

[0005] S1. Data collection: To improve the accuracy of the biological model, real biological data is used for modeling, MRI cross-sectional images of animals are obtained, and appropriate MRI sequences are selected so that different biological tissues can be clearly distinguished in the MRI cross-sectional images.

[0006] S2. Image segmentation: Considering the complexity of biological tissue structure, automatic segmentation has poor effect. Manual segmentation is used directly to segment the MRI cross-sectional image into a mask containing two-dimensional data based on different tissues.

[0007] S3. 3D reconstruction: After obtaining the mask containing the 2D data, a contour connection algorithm is used to perform 3D reconstruction.

[0008] S4, smoothing: In order to remove noise and irregularities on the mesh surface and make the model surface smoother, the Laplace smoothing algorithm is used to smooth the model surface.

[0009] S5. Electromagnetic parameter setting: Import the constructed surface mesh model into the electromagnetic simulation software and fill it into a solid model, assign the material properties of each tissue to obtain a biological electromagnetic model.

[0010] Preferably, the specific process of performing three-dimensional reconstruction in step S3 is as follows:

[0011] S31. Extracting the contour information of the object from each layer of the two-dimensional mask, and using a contour tracking algorithm to obtain contour points representing the boundary.

[0012] S32. Arrange the contour points in counterclockwise order, and use the Douglas-Peucker algorithm to simplify the contour and remove redundant contour points.

[0013] S33, establishing a corresponding relationship between two adjacent layers of two-dimensional contours, and matching each point on one layer of contours with the nearest point on the other layer of contours.

[0014] S34. According to the established corresponding relationship, corresponding points on two adjacent layers of contours are connected to form a three-dimensional surface patch. Three adjacent points are connected to form a triangular patch. A three-dimensional surface mesh model is formed by combining multiple triangular patches.

[0015] Preferably, the specific process of smoothing in step S4 is as follows:

[0016] First, determine the adjacent vertex information of each vertex in the mesh model. In the triangular face mesh, each vertex is adjacent to other vertices of the triangle that shares the vertex. Then calculate the Laplace coordinates according to the following formula:

[0017] .

[0018] In the formula, for any vertex on the model surface Calculate its Laplace coordinates ,That is the set of adjacent vertices, is the number of adjacent vertices, Vertex The adjacent vertices of .

[0019] After obtaining the Laplace coordinates, calculate the new vertex coordinates according to the following formula:

[0020] .

[0021] in, is any vertex on the model surface, is the Laplace coordinate of the vertex, is the new vertex position, is the relaxation factor, ranging from 0 to 1, used to control the degree of smoothing; The larger the value, the greater the distance the vertex moves toward the average position of the adjacent vertices, and the more obvious the smoothing effect; The smaller the value, the less obvious the smoothing effect is, but it can better maintain the original shape of the model. Perform the above position update operation on all vertices in the mesh model to complete the smoothing operation on the model.

[0022] Preferably, in the electromagnetic parameter setting of step S5, the high-order interpolation method is combined to fit the dielectric constant data of biological tissues to achieve accurate electromagnetic simulation of experimental organisms. The data published by relevant research institutes are used, and the high-order interpolation method is then used to obtain the dielectric constant data of materials at the frequency points required for simulation.

[0023] The present invention provides a method for establishing a biological electromagnetic model for physiological parameter monitoring. By adopting a contour connection algorithm, a two-dimensional mask that has undergone image segmentation is reconstructed in three dimensions, and a surface mesh model is established using real biological data, thereby improving the accuracy of the model. The Laplace smoothing algorithm is used to smooth the surface of the model, reduce the reflection of electromagnetic waves on the surface of the model, and improve the sensitivity of the model. The high-order interpolation method is combined to fit the dielectric constant curve of biological tissues to achieve accurate electromagnetic simulation of experimental organisms.

[0024] This method can quickly establish an accurate electromagnetic model based on MRI data, and can change the model parameters based on experimental needs, which is convenient for adjusting parameters such as body fat percentage and heart rate in experiments, and improves the work efficiency and accuracy in the fields of RF sensor design and SAR measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The present invention provides a flow chart of the method for establishing the bio-electromagnetic model.

[0026] Figure 2 This is a fitting curve diagram of the real part of the blood dielectric constant obtained by electromagnetic simulation of the present invention.

[0027] Figure 3 This is a fitting curve diagram of the imaginary part of the blood dielectric constant obtained by electromagnetic simulation of the present invention.

[0028] Figure 4 This is a fitting curve diagram of the loss tangent of the blood dielectric constant obtained by electromagnetic simulation of the present invention.

[0029] Figure 5 This is a schematic diagram of the application of the electromagnetic model established in the present invention in the radio frequency sensor physiological parameter monitoring system.

[0030] Figure 6 It is a schematic diagram of collecting MRI cross-sectional images in an embodiment of the present invention.

[0031] Figure 7 FIG. 4 is a schematic diagram of using the Laplace algorithm to perform surface smoothing in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] like Figure 1 As shown, the present invention provides a technical solution: a method for establishing a bioelectromagnetic model for physiological parameter monitoring, comprising the following steps:

[0034] S1. Data collection: To improve the accuracy of the biological model, real biological data is used for modeling, and MRI cross-sectional images of animals are obtained. Appropriate MRI sequences can be selected so that different biological tissues can be clearly distinguished in the MRI cross-sectional images, such as T1 sequence and T2 sequence. MRI is magnetic resonance imaging.

[0035] S2. Image segmentation: Considering the complexity of biological tissue structure, the effect of automatic segmentation is poor, so manual segmentation is used directly to segment the MRI cross-sectional image into a mask containing two-dimensional data based on different tissues. In addition, the type of biological tissue of the target biological model needs to be determined. The limbs of the experimental dog used for animal experiments are divided into six types of biological tissues, namely bone marrow, bone, muscle, fat, blood vessels, and skin.

[0036] S3, 3D reconstruction: After obtaining the mask containing 2D data, the contour connection algorithm is used for 3D reconstruction; the contour connection algorithm is an effective method for 3D reconstruction based on 2D mask data. Its core idea is to extract the contour information of the object from a series of 2D mask images, and then establish a correspondence between adjacent 2D contours, and construct the facets of the 3D surface by connecting the points on these corresponding contours, and finally combine them into a complete 3D surface mesh model. The algorithm matches and connects contours based on the continuity and similarity of the object contours in adjacent slices.

[0037] The specific 3D reconstruction process is as follows:

[0038] S31. Extracting the contour information of the object from each layer of the two-dimensional mask, and using a contour tracking algorithm to obtain contour points representing the boundary.

[0039] S32. Arrange the contour points in a counterclockwise order and use the Douglas-Peucker algorithm to simplify the contour and remove redundant contour points. The Douglas-Peucker algorithm is an existing algorithm that approximates a curve as a series of points and reduces the number of points. It aims to achieve data compression by reducing the number of points representing the curve while maintaining the overall shape of the curve.

[0040] S33, establishing a corresponding relationship between two adjacent layers of two-dimensional contours, and matching each point on one layer of contours with the nearest point on the other layer of contours.

[0041] S34. According to the established corresponding relationship, corresponding points on two adjacent layers of contours are connected to form a three-dimensional surface patch. Three adjacent points are connected to form a triangular patch. A three-dimensional surface mesh model is formed by combining multiple triangular patches.

[0042] S4, smoothing: In order to remove the noise and irregularities on the mesh surface and make the model surface smoother, the Laplace smoothing algorithm is used to smooth the model surface. The Laplace smoothing algorithm is used to smooth the model surface.

[0043] The Laplace smoothing algorithm is a common algorithm used for surface smoothing of face mesh 3D models. Its core idea is to move each vertex in the mesh model to the average position of its adjacent vertices, so that the model surface becomes smoother. This operation of adjusting the vertex position based on local information is similar to moving the vertex to the "center of gravity" around it, and using the information of adjacent vertices to correct the position of the current vertex to reduce the irregularity and noise of the model surface. The specific smoothing process is as follows:

[0044] First, determine the adjacent vertex information of each vertex in the mesh model. In the triangular face mesh, each vertex is adjacent to other vertices of the triangle that shares the vertex. Then calculate the Laplace coordinates according to the following formula:

[0045] .

[0046] In the formula, for any vertex on the model surface Calculate its Laplace coordinates ,That is the set of adjacent vertices, is the number of adjacent vertices, Vertex The adjacent vertices of .

[0047] After obtaining the Laplace coordinates, calculate the new vertex coordinates according to the following formula:

[0048] .

[0049] in, is any vertex on the model surface, is the Laplace coordinate of the vertex, is the new vertex position, is the relaxation factor, ranging from 0 to 1, used to control the degree of smoothing; The larger the value, the greater the distance the vertex moves toward the average position of the adjacent vertices, and the more obvious the smoothing effect; The smaller the value, the less obvious the smoothing effect is, but it can better maintain the original shape of the model. Perform the above position update operation on all vertices in the mesh model to complete the smoothing operation on the model.

[0050] S5. Electromagnetic parameter setting: Import the constructed surface mesh model into the electromagnetic simulation software, fill it into a solid model, assign values ​​to the material properties of each tissue, and obtain a bio-electromagnetic model. Use the data published by relevant research institutes and adopt high-order interpolation methods to obtain the material dielectric constant data at the frequency points required for simulation. Figure 2-4 As shown, the fitting curves of the real part, imaginary part, and loss tangent of the blood dielectric constant are shown. The horizontal axis is the frequency in GHz, the vertical axis is the parameter amplitude, the blue circle is the discrete frequency point data used for curve fitting, and the red line is the continuous curve obtained by high-order interpolation fitting. The obtained continuous dielectric constant data can be used for electromagnetic simulation under the time domain solver.

[0051] According to the above operation, a biological electromagnetic model can be finally obtained. It is established based on real biological data, has high accuracy, and the material parameters of the model can be freely changed, which can well meet the requirements of biological science research experiments. Figure 6 As shown in FIG. 1 , the MRI cross-sectional images of medical rats are obtained when the bio-electromagnetic model is established using medical rats as an example. Figure 7 Figure 2 is a schematic diagram of surface smoothing using the Laplace algorithm.

[0052] According to the biological model, a high-precision radio frequency sensor is designed to monitor physiological parameters such as animal heart rate and body fat percentage. Based on the non-invasive animal physiological parameter monitoring system designed with radio frequency sensors, relevant wearable devices are manufactured to continuously monitor the heart rate and body fat percentage of animals in the experiment, providing physiological parameter data for animal experiments.

[0053] like Figure 5As shown in the figure, in the RF sensor physiological parameter monitoring system, the RF sensor is responsible for the collection of physiological parameters, the high-pass filter and the low-pass filter are responsible for filtering the low-frequency and high-frequency noise in the signal, the analog-to-digital converter in the microcontroller converts the analog signal into a digital signal, and sends it to the data processing host through the UART or USB interface, the lithium battery is responsible for powering the microcontroller, and the voltage regulator makes the battery output a stable voltage. The collected signal is processed on the host, and the physiological parameters are calculated to realize the monitoring of animal physiological parameters.

[0054] In the present invention, a bio-electromagnetic model is established according to five steps, namely, data collection, image segmentation, three-dimensional reconstruction, smoothing, and electromagnetic characteristic parameter setting. Then, the model is used to design a radio frequency sensor for measuring animal physiological parameters, and a physiological parameter monitoring system is designed based on the sensor to realize continuous monitoring of animal physiological parameters.

[0055] The present invention can be used to design non-invasive sensor devices worn on the surface of animals to monitor physiological parameters such as body fat rate and heart rate of animals. Such biological wearable devices can be used in the fields of animal science research, animal health management, animal husbandry, animal protection and wildlife management.

[0056] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for establishing a bioelectromagnetic model for monitoring physiological parameters, characterized in that: The steps include: S1. Data collection: Use real biological data for modeling, obtain MRI cross-sectional images of animals, and select appropriate MRI sequences so that different biological tissues can be clearly distinguished in MRI cross-sectional images; S2. Image segmentation: Using manual segmentation, the MRI cross-sectional images are segmented into masks containing two-dimensional data based on different tissues; S3, 3D reconstruction: after obtaining the mask containing the 2D data, a contour connection algorithm is used to perform 3D reconstruction; S4, smoothing: use Laplace smoothing algorithm to smooth the model surface; S5. Electromagnetic parameter setting: Import the constructed surface mesh model into the electromagnetic simulation software and fill it into a solid model, assign the material properties of each tissue to obtain a biological electromagnetic model.

2. The method for establishing a bio-electromagnetic model for monitoring physiological parameters according to claim 1, characterized in that: The specific process of performing three-dimensional reconstruction in step S3 is as follows: S31, extracting contour information of the object from each layer of the two-dimensional mask, and using a contour tracking algorithm to obtain contour points representing the boundary; S32, arranging the contour points in a counterclockwise order, and using the Douglas-Peucker algorithm to simplify the contour and remove redundant contour points; S33, establishing a corresponding relationship between two adjacent layers of two-dimensional contours, and matching each point on one layer of contours with the nearest point on the other layer of contours; S34. According to the established corresponding relationship, corresponding points on two adjacent layers of contours are connected to form a three-dimensional surface patch. Three adjacent points are connected to form a triangular patch. A three-dimensional surface mesh model is formed by combining multiple triangular patches.

3. The method for establishing a bio-electromagnetic model for monitoring physiological parameters according to claim 1, characterized in that: The specific process of smoothing in step S4 is as follows: First, determine the adjacent vertex information of each vertex in the mesh model. In the triangular face mesh, each vertex is adjacent to other vertices of the triangle that shares the vertex. Then calculate the Laplace coordinates according to the following formula: ； In the formula, for any vertex on the model surface Calculate its Laplace coordinates ,That is the set of adjacent vertices, is the number of adjacent vertices, Vertex Adjacent vertices; after obtaining the Laplace coordinates, calculate the new vertex coordinates according to the following formula: ； in, is any vertex on the model surface, is the Laplace coordinate of the vertex, is the new vertex position, is the relaxation factor, ranging from 0 to 1, used to control the degree of smoothing; The larger the value, the greater the distance the vertex moves toward the average position of the adjacent vertices, and the more obvious the smoothing effect; The smaller the value, the less obvious the smoothing effect is, but it can better maintain the original shape of the model. Perform the above position update operation on all vertices in the mesh model to complete the smoothing operation on the model.

4. The method for establishing a bio-electromagnetic model for monitoring physiological parameters according to claim 1, characterized in that: In the electromagnetic parameter setting of step S5, a high-order interpolation method is combined to fit the dielectric constant data of biological tissue.

Images