A comprehensive evaluation method for muscle fiber differentiation and orientation effect of biological muscle actuator

Abstract

The present invention discloses a comprehensive evaluation method for the differentiation and orientation effect of muscle fibers of a biological muscle driver, including: S1, establishing a muscle fiber differentiation effect evaluation algorithm, the muscle fiber differentiation effect evaluation algorithm includes establishing a differentiation evaluation algorithm under two-dimensional culture conditions and a differentiation evaluation algorithm under three-dimensional culture conditions based on the number, length and width of muscle fibers; S2, establishing a muscle fiber orientation effect evaluation algorithm; S3, establishing an evaluation model and training based on the muscle fiber differentiation effect evaluation algorithm described in step S1 and the muscle fiber orientation effect evaluation algorithm described in step S2; S4, inputting any photographed picture into the evaluation model to obtain quantitative muscle fiber differentiation effect and orientation effect. The present invention constructs a method for judging muscle fiber differentiation in two different dimensions of a biological muscle driver and performs classification evaluation, as well as a general orientation judgment method, which has strong pertinence and can effectively help researchers conduct specific analysis and judgment according to different situations.

Description

Technical Field

[0001] The present invention relates to the technical field of biological muscle fiber differentiation, and more specifically, to a method for comprehensively evaluating the muscle fiber differentiation and orientation effect of a biological muscle driver. Background Art

[0002] With the development of the times and the progress of science and technology, various types of micro-bionic robots with a size of millimeters or less have entered the field of vision of researchers. Micro-robots can perform various difficult tasks in narrow and complex environments, and have great application prospects in battlefield reconnaissance, medical surgery, tissue regeneration, etc. However, they are still a long way from the driving performance of real organisms, and the driver plays a decisive role in this process. Existing micro-robots are mainly driven by traditional driving devices or intelligent materials such as motors, dielectric elastomers, and shape memory alloys. Each type of drive has its own advantages. For example, the motor has high manufacturing precision and mature manufacturing technology; the dielectric elastomer has high flexibility and large strain; the shape memory alloy has high power density, etc. However, their overall performance is still a long way from the real organism. Therefore, the biological muscle actuator constructed by real biological muscle cells and bioactive materials (such as hydrogels such as fibrinogen, matrix glue, and methacryloyl gelatin (GelMA) in natural polymer materials) has received widespread attention from researchers. It has the advantages of high biocompatibility, high energy conversion efficiency, and high flexibility.

[0003] The muscle tissue of real organisms is composed of several nanoscale muscle fibers arranged in parallel. Therefore, the biological muscle actuator simulates the structure of real biological muscle tissue, which is prepared by real biological muscle cells and bioactive materials according to the designed shape and formed under certain external excitation and culture conditions. Muscle fibers are the basic units for real biological muscle tissue to generate driving force, so the differentiation and orientation effects of muscle fibers are the key factors that determine the motor performance of biological muscle actuators. Researchers use mechanical stimulation, electrical stimulation and other methods to promote the differentiation and orientation of muscle fibers. However, there is still a lack of comprehensive evaluation algorithms for the differentiation and orientation effects of muscle fibers for quantitative judgment and comparison. Based on the quantitative evaluation algorithm, researchers can obtain appropriate external excitation parameters (such as the optimal voltage amplitude and pulse under electrical stimulation). In addition, researchers can also establish theoretical mathematical models with different external excitation parameters to provide references for the driving performance of different types of biological muscle actuators. To this end, the work that needs to be done is: based on the microscopic photos taken, a comprehensive evaluation algorithm for the differentiation and orientation effects of muscle fibers under various dimensional conditions can be established, and the two can be quantitatively compared at the same time, so as to quantitatively compare the differentiation and orientation performance of muscle fibers. Summary of the invention

[0004] The purpose of the present invention is to provide a method for comprehensively evaluating the differentiation and orientation effects of muscle fibers of a biological muscle actuator, so as to overcome the defects existing in the prior art.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A method for comprehensively evaluating the differentiation and orientation effects of muscle fibers of a biological muscle actuator comprises the following steps:

[0007] S1. Establishing a muscle fiber differentiation effect evaluation algorithm, wherein the muscle fiber differentiation effect evaluation algorithm includes establishing a differentiation evaluation algorithm under two-dimensional conditions and a differentiation evaluation algorithm under three-dimensional conditions based on the number, length and width of muscle fibers;

[0008] S2. Establish an algorithm for evaluating the effect of muscle fiber orientation;

[0009] S3, establishing and training an evaluation model based on the muscle fiber differentiation effect evaluation algorithm described in step S1 and the muscle fiber orientation effect evaluation algorithm described in step S2;

[0010] S4. Input any captured images into the evaluation model to obtain quantitative muscle fiber differentiation effect and orientation effect.

[0011] Furthermore, in step S1, establishing a differentiation evaluation algorithm under two-dimensional conditions based on the number, length and width of muscle fibers includes:

[0012] S11. Derive the dimensionless coefficients Q and P:

[0013]

[0014]

[0015]

[0016]

[0017] In the formula, S i are the area of ​​each muscle fiber in a single image, L i is the length of each muscle fiber in a single image, W i is the width of each muscle fiber in a single image, S all is the total area of ​​a single image, n is the total number of muscle fibers in a single image, and m is the number of muscle fiber images at different positions of a single sample, m ≥ 3;

[0018] Using the single muscle fiber area S i Divide by its length L i Get the average width Make it a replacement for W i :

[0019]

[0020] S12, Q j With P j is the dimensionless coefficient of the jth image, Q j The range of Q is [0, 1]. Under ideal conditions, Q = 1, P j The range of P is (0, 1). The closer P is to 0, the closer the fiber shape is to the real biological muscle fiber shape. The formula of the differentiation evaluation algorithm under two-dimensional conditions is:

[0021] W=Q i+P j

[0022]

[0023] In the formula, i, j are unit vectors, |W| is the modulus of W, and under ideal conditions, the closer |W| is to 1, the better the differentiation effect;

[0024] S13. A weight relationship formula is established by combining the driving force and driving frequency of the muscle driver under different muscle fiber differentiation conditions to determine the influence of the two values ​​of dimensionless coefficients Q and P on the muscle fiber differentiation effect and the muscle driver driving effect.

[0025] Furthermore, the step S1 of establishing a differentiation evaluation algorithm under three-dimensional conditions specifically includes:

[0026] S14, when the muscle cells differentiate into three-dimensional muscle fibers inside the material, staining the proteins of the muscle fiber differentiation effect by immunofluorescence staining method;

[0027] S15. Establish dimensionless coefficients PT and QT, where PT is for the area of ​​the fluorescence region and QT is for the fluorescence intensity:

[0028]

[0029]

[0030] Where m is the number of immunofluorescence images of a single sample at different locations, m≥3, FA j , FA0, FL j , FL0 are the fluorescence area of ​​the jth three-dimensional immunofluorescence staining image, the fluorescence area of ​​the muscle cell two-dimensional culture immunofluorescence staining, the fluorescence intensity of the jth three-dimensional immunofluorescence staining image, and the fluorescence intensity of the muscle cell two-dimensional culture immunofluorescence staining, respectively;

[0031] S16. The formula of the differentiation evaluation algorithm under three-dimensional conditions is introduced as follows:

[0032] WT=QTi+PTj

[0033]

[0034] In the formula, i, j are unit vectors, |WT| is the modulus of WT, and under ideal conditions, the closer |WT| is to 1, the better the differentiation effect;

[0035] S17. A weight relationship formula is established by combining the driving force and driving frequency of the muscle driver under different muscle fiber differentiation conditions to determine the influence of the two values ​​of dimensionless coefficients PT and QT on the muscle fiber differentiation effect and the muscle driver driving effect.

[0036] Furthermore, the step S2 is specifically as follows:

[0037] Taking the external excitation along a certain direction in a single muscle fiber micrograph as the reference, the angle between different muscle fibers and the reference direction is θ i The dimensionless coefficient DO is used to judge the effect of muscle fiber orientation:

[0038]

[0039]

[0040] In the formula, is the mean of the angles formed between the muscle fibers in a single image and the reference, SD(θ) is the standard deviation of the angle, i, j are unit vectors, |DO| is the modulus of DO, The logic is that under ideal conditions, all muscle fibers should have consistent directionality, so the standard deviation should be much smaller than the mean. By converting it into a dimensionless coefficient through the denominator, it is transformed from a physical meaning to a pure mathematical formula. The logic is that it is 0° under ideal conditions, and to convert it into a dimensionless coefficient, divide it by 180° (π angle). and Ideally, |DO| should be equal to 0. Therefore, the closer |DO| is to 0, the better the orientation effect.

[0041] Furthermore, in step S3, the evaluation model is trained using methods not limited to machine learning or deep learning.

[0042] Compared with the prior art, the advantages of the present invention are: the present invention constructs a method for judging muscle fiber differentiation in two different dimensions (two-dimensional culture and three-dimensional culture) of biological muscle actuators and performs classification evaluation, as well as a general directional judgment method, which is highly targeted and can effectively help researchers conduct specific analysis and judgment according to different situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is a flow chart of the comprehensive evaluation method of muscle fiber differentiation and orientation effect of the biological muscle driver of the present invention.

[0045] Figure 2 Figure a is a schematic diagram of a method for evaluating the effect of two-dimensional differentiation of muscle fibers, and figure b is a schematic diagram of a method for evaluating the effect of directional muscle fibers. DETAILED DESCRIPTION

[0046] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings so that the advantages and features of the present invention can be more easily understood by those skilled in the art, thereby making a clearer and more definite definition of the protection scope of the present invention.

[0047] See also Figure 1 As shown, this embodiment discloses a method for comprehensively evaluating the differentiation and orientation effects of muscle fibers of a biological muscle actuator, comprising the following steps:

[0048] Step S1, establishing a muscle fiber differentiation effect evaluation algorithm, wherein the muscle fiber differentiation effect evaluation algorithm includes establishing a two-dimensional differentiation evaluation algorithm and a three-dimensional differentiation evaluation algorithm based on the number, length and width of muscle fibers.

[0049] Step S2: establishing an algorithm for evaluating the effect of muscle fiber orientation.

[0050] Step S3: establishing and training an evaluation model based on the muscle fiber differentiation effect evaluation algorithm described in step S1 and the muscle fiber orientation effect evaluation algorithm described in step S2.

[0051] Step S4: input any captured images into the evaluation model to obtain quantitative muscle fiber differentiation effect and orientation effect.

[0052] Specifically, the two-dimensional differentiation evaluation algorithm is established based on the number, length and width of muscle fibers in step S1, including:

[0053] Step S11: According to the shape and arrangement characteristics of muscle fibers inside muscle tissue, it can be known that a good differentiation effect needs to meet two conditions: slenderness and a large number. Taking the cultivation of C2C12 cells (skeletal muscle cells) on the surface of GelMA material and the differentiation of muscle fibers as an example, a two-dimensional differentiation evaluation algorithm is established based on the number, length and width of muscle fibers, and the dimensionless coefficients Q and P are derived:

[0054]

[0055]

[0056]

[0057]

[0058] In the formula, S i are the area of ​​each muscle fiber in a single image, L i is the length of each muscle fiber in a single image, W i is the width of each muscle fiber in a single image, S all is the total area of ​​a single image, n is the total number of muscle fibers in a single image, and m is the number of muscle fiber images at different positions of a single sample, m ≥ 3;

[0059] Using the single muscle fiber area S i Divide by its length L i Get the average width Make it a replacement for W i :

[0060]

[0061] Step S12, Q j With P j is the dimensionless coefficient of the jth image, Q j The range of Q is [0, 1]. Under ideal conditions, Q = 1, P j The range of P is (0, 1). The closer P is to 0, the closer the fiber shape is to the real biological muscle fiber shape. Therefore, the formula for the comprehensive evaluation algorithm for the differentiation effect of two-dimensional cultured muscle fibers is derived as follows:

[0062] W=Q i+P j (5)

[0063]

[0064] Where i, j are unit vectors, |W| is the modulus of W. Under ideal conditions, the closer |W| is to 1 (i.e., the closer Q, P are to 1), the better the differentiation effect.

[0065] Step S13, combining the experimental data such as the driving force and driving frequency of the muscle driver under different muscle fiber differentiation conditions with formulas 5 and 6, so as to establish a weight relationship formula to determine the influence of the two values ​​of the dimensionless coefficients Q and P on the driving effect, that is, to determine which of the two values ​​Q and P has a greater influence on the muscle fiber differentiation effect and the muscle driver driving effect, and what proportion they account for.

[0066] Specifically, the establishment of the three-dimensional culture evaluation algorithm in step S1 specifically includes:

[0067] Step S14, when the muscle cells differentiate into three-dimensional muscle fibers inside the material, the muscle fibers are distributed in different planes, which makes it difficult to use the above method to judge the differentiation effect of the captured images. Therefore, it is necessary to use immunofluorescence staining to stain proteins (such as actin, myosin heavy chain, etc.) that can judge the effect of muscle fiber differentiation, and judge based on the area and fluorescence intensity of the fluorescent area. The muscle cells are cultured in two dimensions on the culture dish under the same conditions as the three-dimensional culture, and their immunofluorescence staining is used as the control group.

[0068] Step S15, establish dimensionless coefficients PT and QT, PT refers to the area of ​​the fluorescence region, and QT refers to the fluorescence intensity:

[0069]

[0070]

[0071] Where m is the number of immunofluorescence images of a single sample at different locations, m≥3, FA j , FA0, FL j , FL0 are the fluorescent area of ​​the j-th three-dimensional immunofluorescence staining image, the fluorescent area of ​​the muscle cell two-dimensional culture immunofluorescence staining, the fluorescence intensity of the j-th three-dimensional immunofluorescence staining image, and the fluorescence intensity of the muscle cell two-dimensional culture immunofluorescence staining respectively; the ranges of the two dimensionless coefficients QT and PT are both [0, +∞). Under ideal conditions, the larger the value, the better the differentiation effect.

[0072] Step S16: The formula for introducing the three-dimensional culture evaluation algorithm is:

[0073] WT=QTi+PTj

[0074]

[0075] In the formula, i, j are unit vectors, |WT| is the modulus of WT, and under ideal conditions, the closer |WT| is to 1, the better the differentiation effect;

[0076] Step S17, combining the experimental data of the driving force and driving frequency of the muscle driver under different muscle fiber differentiation conditions with formulas 9 and 10, so as to establish a weight relationship formula to determine the influence of the two values ​​of dimensionless coefficients PT and QT on the driving effect, that is, to determine which of the two values ​​QT and PT has a greater influence on the muscle fiber differentiation effect and the muscle driver driving effect, and what proportion they account for.

[0077] Based on the above evaluation algorithm, this embodiment can comprehensively evaluate the differentiation effect of muscle fibers of the muscle driver in two dimensions. Specifically, at least three images are measured for each sample (ie, m≥3 in the above formula), and at least three samples are tested to reduce random errors.

[0078] Specifically, the step S2 is as follows:

[0079] Considering that muscle fibers grow in a directional manner according to the designed direction under external excitation conditions, the angle between different muscle fibers and the reference direction is θ, taking the external excitation along a certain direction in a single muscle fiber micrograph as the reference. i (The angles between the muscle fibers and the reference are all considered acute angles). The dimensionless coefficient DO is used to determine the effect of muscle fiber orientation:

[0080]

[0081]

[0082] In the formula, is the mean of the angles formed between the muscle fibers in a single image and the reference, SD(θ) is the standard deviation of the angle, i, j are unit vectors, |DO| is the modulus of DO, The logic is that under ideal conditions, all muscle fibers should have consistent directionality, so the standard deviation should be much smaller than the mean. By converting it into a dimensionless coefficient through the denominator, it is transformed from a physical meaning to a pure mathematical formula. The logic is that it is 0° under ideal conditions, and to convert it into a dimensionless coefficient, divide it by 180° (π angle). and Ideally, |DO| should be equal to 0. Therefore, the closer |DO| is to 0, the better the orientation effect.

[0083] In comparison, SD(θ) is More importantly, as long as SD(θ) is small enough or even close to 0, the cultured muscle fibers do not grow in the direction of the applied external excitation, but at least the direction is consistent. It is only necessary to change the driving direction when driving the structure. However, if SD(θ) is large, the internal muscle fibers are more likely to become isotropic, and the driving force will be greatly weakened. Therefore, the algorithm can intuitively determine the degree of consistency of the muscle fiber direction. At the same time, when there is no external excitation, only the direction of the muscle fiber direction needs to be considered. The size of the muscle fiber can be used to determine the consistency of the direction.

[0084] For two-dimensional differentiation, each sample needs to be photographed at least 3 times at different positions; for three-dimensional differentiation, considering the increase in dimensions, it is necessary to take more than 3 pictures at different positions in the XY plane at different z-axis values, and the z-axis values ​​should be taken at least at 3 different z-axis positions, with a total of at least 9 pictures (three-dimensional differentiation requires immunofluorescence staining to present the shape of muscle fibers). Regardless of two-dimensional or three-dimensional differentiation, at least three samples need to be tested.

[0085] In this embodiment, the evaluation model is trained by machine learning or deep learning in step S3. The present invention can directly perform rapid processing and analysis on different captured images, and efficiently and accurately complete the judgment of muscle fiber differentiation effect and orientation effect of different samples. At the same time, the empirical formula formed by combining the driving performance and external control parameters can be improved and optimized in continuous training, so as to achieve accurate prediction of driving performance.

[0086] For ease of understanding, Figure 2 Figure a is a schematic diagram of a method for evaluating the effect of two-dimensional differentiation of muscle fibers, and figure b is a schematic diagram of a method for evaluating the effect of directional muscle fibers.

[0087] The present invention comprehensively considers the quantitative comprehensive evaluation algorithm of the muscle fiber differentiation effect and the muscle fiber orientation effect of the constructed biological muscle actuator driving performance, providing a concise and clear reference for researchers in this field to judge the differentiation and orientation effects of muscle fibers under different culture conditions or different external stimulation conditions, instead of qualitative judgment by the naked eye, thereby improving accuracy.

[0088] The present invention comprehensively considers the muscle fiber differentiation and orientation judgment methods when constructing two different dimensions (two-dimensional culture and three-dimensional culture) of biological muscle actuators and conducts classification evaluation. It has strong pertinence and can effectively help researchers conduct specific analysis and judgment according to different situations.

[0089] The two-dimensional differentiation effect evaluation method of the present invention under the two-dimensional differentiation condition can be established based on the structural shape of real biological muscle tissue, and a comprehensive evaluation method based on dimensionless coefficients is constructed based on length, width, number and area, thereby converting engineering problems into mathematical problems.

[0090] The muscle fiber orientation evaluation algorithm of the present invention comprehensively considers the mean and standard deviation of the angle with the reference direction, and clarifies the priority of the factors affecting the muscle fiber orientation effect. At the same time, the algorithm is compatible with two-dimensional and three-dimensional dimensions and has universality.

[0091] The algorithms in the present invention can be trained by machine learning or deep learning, so as to quickly process and analyze the images of different muscle fibers and draw conclusions in a simple and efficient manner, thereby greatly shortening the data processing time and improving the accuracy. At the same time, the muscle fiber differentiation effect and orientation effect under the influence of other external control parameters can be predicted, so as to obtain the optimal parameters with the least experiments, so that different muscle driver structures can obtain the best driving performance.

[0092] The differentiation effect evaluation algorithm and the muscle fiber orientation effect evaluation algorithm in the present invention both establish space vectors, so that researchers can make comprehensive judgments by finding the model, and can also directly and intuitively see which part affects the differentiation or muscle fiber orientation effect through the formula. At the same time, researchers can use this as a basis to combine experimental parameters such as muscle driver driving force and driving frequency with these algorithms to establish weight relationships and empirical formulas, so as to predict the muscle fiber differentiation and orientation effects and driving performance under the conditions of other different external control parameters (such as excitation time, excitation direction, driver structure shape, etc.).

[0093] Although the embodiments of the present invention are described in conjunction with the accompanying drawings, the patent owner may make various variations or modifications within the scope of the appended claims. As long as they do not exceed the protection scope described in the claims of the present invention, they should be within the protection scope of the present invention.

Claims

1. A method for comprehensively evaluating the differentiation and orientation effects of muscle fibers of a biological muscle actuator, characterized in that: The following steps are involved: S1. Establishing a muscle fiber differentiation effect evaluation algorithm, wherein the muscle fiber differentiation effect evaluation algorithm includes establishing a differentiation evaluation algorithm under two-dimensional culture conditions and a differentiation evaluation algorithm under three-dimensional culture conditions based on the number, length and width of muscle fibers; S2. Establish an algorithm for evaluating the effect of muscle fiber orientation; S3, establishing and training an evaluation model based on the muscle fiber differentiation effect evaluation algorithm described in step S1 and the muscle fiber orientation effect evaluation algorithm described in step S2; S4, inputting any captured images into the evaluation model to obtain quantitative muscle fiber differentiation effect and orientation effect; The step S1 in which a differentiation evaluation algorithm is established under two-dimensional conditions based on the number, length and width of muscle fibers comprises: S11. Derive the dimensionless coefficients Q and P: In the formula, S i is the area of ​​each muscle fiber in a single image, L i is the length of each muscle fiber in a single image, W i is the width of each muscle fiber in a single image, S all is the total area of ​​a single image, n is the total number of muscle fibers in a single image, and m is the number of muscle fiber images at different positions of a single sample, m ≥ 3; Using the single muscle fiber area S i Divide by its length L i Get the average width Make it a replacement for W i : S12, Q j With P j is the dimensionless coefficient of the jth image, Q j The range of Q is [0, 1]. Under ideal conditions, Q = 1, P j The range of P is (0, 1). The closer P is to 0, the closer the fiber shape is to the real biological muscle fiber shape. The formula of the differentiation evaluation algorithm under two-dimensional conditions is: W=Qi+Pj In the formula, i, j are unit vectors, W is the modulus of W, and under ideal conditions, the closer W is to 1, the better the differentiation effect; S13, combining the driving force and driving frequency of the muscle driver under different muscle fiber differentiation conditions to establish a weight relationship formula to determine the influence of the two values ​​of dimensionless coefficients Q and P on the muscle fiber differentiation effect and the muscle driver driving effect; The step S1 of establishing a differentiation evaluation algorithm under three-dimensional conditions specifically includes: S14, when the muscle cells differentiate into three-dimensional muscle fibers inside the material, staining the proteins of the muscle fiber differentiation effect by immunofluorescence staining method; S15. Establish dimensionless coefficients PT and QT, where PT is for the area of ​​the fluorescence region and QT is for the fluorescence intensity: Where m is the total number of immunostaining fluorescence images of a single sample at different locations, m≥3, FA j , FA0, FL j , FL0 are the fluorescence area of ​​the jth three-dimensional immunofluorescence staining image, the fluorescence area of ​​the muscle cell two-dimensional culture immunofluorescence staining, the fluorescence intensity of the jth three-dimensional immunofluorescence staining image, and the fluorescence intensity of the muscle cell two-dimensional culture immunofluorescence staining, respectively; S16. The formula of the differentiation evaluation algorithm under three-dimensional conditions is introduced as follows: WT=QTi+PTj In the formula, i, j are unit vectors, WT is the modulus of WT, and under ideal conditions, the closer WT is to 1, the better the differentiation effect; S17, combining the driving force and driving frequency of the muscle driver under different muscle fiber differentiation conditions to establish a weight relationship formula to determine the influence of the two values ​​of dimensionless coefficients PT and QT on the muscle fiber differentiation effect and the muscle driver driving effect; The step S2 is specifically as follows: Taking the external excitation along a certain direction in a single muscle fiber micrograph as the reference, the angle between different muscle fibers and the reference direction is θ i The dimensionless coefficient DO is used to judge the effect of muscle fiber orientation: Where θ is the mean of the angles between the muscle fibers in a single image and the reference, SD(θ) is the standard deviation of the angle, i, j are unit vectors, DO is the modulus of DO, and Under ideal conditions, they should all be equal to 0. Therefore, the closer DO is to 0, the better the directional effect. In step S3, the evaluation model is trained in a manner not limited to machine learning or deep learning and is used for quantitative evaluation of new images.

Images