System and method for realizing vagina tactile super-resolution contact force measurement

Abstract

The invention discloses a system and a method for realizing vagina tactile super-resolution contact force measurement, relates to the technical field of use of medical instruments, constructs a high-precision measurement method and a spatial distribution system for contact force in a vagina, and solves the problems of high time delay, low precision, large crosstalk, low spatial resolution and the like of a traditional array type pressure vagina probe. Therefore, the contact perception capability and the spatial resolution of vagina biological force are improved, and the physical integration and wiring complexity is reduced; according to the invention, the biological force applied by the vagina wall to the probe can be accurately measured in a mode of directly measuring the contact force, so that a vagina biomechanical map is drawn; meanwhile, a low-density sensing system and a super-resolution pressure space prediction method are designed, so that the arrangement complexity and the process cost of a high-density sensing unit are reduced, and the vagina tactile sensing resolution is effectively improved through automatic calibration and prediction means.

Description

Technical Field

[0001] The present invention relates to the technical field of medical device use, and in particular to a system and method for realizing vaginal tactile super-resolution contact force measurement. Background Art

[0002] The contact force measurement of pelvic floor and vaginal tissue is usually divided into direct measurement and indirect measurement.

[0003] Direct measurement methods such as vaginal pressure measurement usually use specific instruments such as vaginal pressure balloons or vaginal dilators to directly feel the tension and touch of the vagina. Vaginal pressure balloons can quantitatively collect and analyze the touch of vaginal tissue, but this method is often less accurate and less sensitive; vaginal dilators usually evaluate the touch of pelvic floor tissue through subjective feelings, and the evaluation results vary from person to person, lacking a scientific and objective evaluation system.

[0004] Indirect measurement methods include vaginal impedance measurement and shear wave tactile measurement. The vaginal impedance measurement method indirectly reflects the moisture, fat content, collagen and other information of the vagina through the impedance of the vagina at different current frequencies, and then indirectly reflects the tactile sensation of the vaginal tissue. This method is affected by many interference factors such as vaginal temperature and electrode placement, and has poor stability and inconsistent results from multiple measurements. The shear wave tactile measurement method generates transverse shear waves by focusing the acoustic radiation force of the probe, and uses fast plane wave excitation to track and detect the displacement and shear wave velocity of the tissue during shear wave propagation. The propagation velocity of the shear wave is directly related to the tactile sensation and force tactile sensation of the tissue. In tissues with greater tactile sensation, the propagation velocity of the shear wave is faster, while in tissues with less tactile sensation, the propagation velocity is slower. By measuring the propagation velocity of the shear wave, the tactile sensation of the tissue can be indirectly evaluated. The measurement results of this method are greatly affected by the operator's experience and technical level, and the equipment purchase and maintenance costs are high.

[0005] Therefore, the existing technology has the following problems: the schemes for directly or indirectly measuring the tactile sensation of the pelvic floor and vaginal tissue are not accurate, have poor sensitivity and resolution, and suffer from severe crosstalk and distortion; the measurement method is easily interfered by external factors, overly relies on the operator's operating experience, and the measurement results are unstable; the measurement method is not direct and objective enough, and often uses indirect measurement to map the biomechanical tactile characteristics of the vagina, resulting in large errors; the equipment is expensive, and the purchase and maintenance costs are high. Summary of the invention

[0006] The object of the present invention is to provide a system and method for realizing vaginal tactile super-resolution contact force measurement, so as to solve the problems raised in the prior art.

[0007] To achieve the above object, the present invention provides the following technical solutions. A method for realizing vaginal tactile super-resolution contact force measurement, the method comprising:

[0008] Step S1: Construct a collection array and deploy the collection array on a calibration table, perform a contact force test on the collection array through the calibration table, and record the contact force information and position information during the contact force test;

[0009] Step S2: Calibrate the contact force information to obtain an average tactile vector and a linear slope;

[0010] Step S3: Construct a training data set according to the average tactile vector, the linear slope and the position information, and input the training data set into a prediction model for training;

[0011] Step S4: Collect the sites in the vagina that need to be measured for contact force through the collection array, input the sites into the trained model for prediction, and transmit the prediction results to an image display module.

[0012] Further, in step S1, the collection array is a 2-row N-column tactile array composed of 2×N tactile units, and each of the tactile units includes a sensor.

[0013] Further, in step S1, the contact force information includes tactile data and pressure data. Among them, the tactile data includes the orientation of each pressure applied to the acquisition system during the contact force test, and the pressure data includes the pressure value of each pressure applied to the acquisition system during the contact force test; the position information includes the coordinates of the sensor corresponding to the change in the pressure value during the contact force test on the collection array.

[0014] Further, in step S1, the contact force information and the position information are recorded in the form of a matrix, and the form of the matrix is:

[0015] PosPress = {(Pos i (X, Y), Press i )};

[0016] Among them, PosPress represents the matrix for recording the position information and the contact force information, Pos i (X, Y) represents the characteristic parameter for recording the i-th position information collected by the collection array, and Press i represents the characteristic parameter for recording the i-th contact force information collected by the collection array, 0 < i ≤ 2N×Coe, Coe represents the calibration coefficient, and N represents the number of columns of the collection array.

[0017] Furthermore, in step S2, the contact force information collected by the collection array is calibrated using a calibration platform to obtain an average tactile vector and a linear slope, wherein the step of obtaining the average tactile vector and the linear slope includes:

[0018] Step S2-1: Name the sensor in each tactile unit and the pressure value collected by the sensor in the following manner:

[0019] Sensor a×b ;

[0020] P a×b ;

[0021] Wherein, a and b represent the horizontal and vertical serial numbers of the sensor in the tactile array, respectively, Sensor represents the sensor, and P represents the pressure value collected by the sensor;

[0022] Step S2-2: define the four sensors contained in the four adjacent tactile units in the tactile array as a sensing unit group, and the sensing unit group is expressed as follows:

[0023]

[0024] Among them, Group k represents the kth pressure sensing unit group;

[0025] Step S2-3: Calibrate the contact force information by Gaussian fitting to obtain an average tactile vector. The average tactile vector is obtained as follows:

[0026]

[0027] Among them, (μ xi ,μ yi ) is the mean vector of the contact force information collected by the four sensors around the i-th tactile unit at (x, y), m i is the amplitude value corresponding to the i-th contact force information, σ i is the fitting variance corresponding to the i-th contact force information, (x, y) represents the coordinates of the sensor corresponding to the contact force information, and p i (x, y) represents the average tactile vector of the i-th tactile unit at (x, y);

[0028] Step S2-4: Obtaining the linear slope corresponding to the contact force information by Gaussian fitting;

[0029]

[0030] Among them, v i represents the amplitude coefficient corresponding to the i-th contact force information, ki (x, y) represents the linear slope corresponding to the contact force information collected by the sensor of the i-th tactile unit at (x, y);

[0031] The above steps place the acquisition array on a calibration table for calibration. The calibration includes but is not limited to the calibration of contact force information, position information and temperature compensation coefficient. The larger the calibration coefficient, the more calibration data, and the more accurate the calibration result. Calibration using two-dimensional Gaussian function fitting can reduce the complexity of calibration and improve efficiency. In each pressure sensing unit group, the values ​​of pressure and position information at each pressure sensing unit are related to the tactile information of the four fixed pressure sensing units around it. The tactile information includes tactile data, pressure data, position information and slope information of tactile pressure changes. This method can complete the data when there is less training data, reducing the amount of calculation and manual operation.

[0032] Furthermore, the training data set described in step S3 includes a feature matrix and a target vector, wherein the feature matrix is ​​a two-column and n-row matrix consisting of the average tactile vector and the linear slope, and the target vector is composed of n rows of the position information.

[0033] Furthermore, in step S4, the KNN model is used to predict the site in the vagina where the contact force measurement is required, and the pressure value at the site is calculated by the KNN model, and the calculation method is:

[0034] F target =min(d(X i , X j )) / k i (x est ,y est );

[0035] Among them, min(d(x i ,x j )) represents the minimum distance between the site (xi,yi) and the site (xj,yj) with the largest pressure value among the four surrounding sites, F target Indicates the pressure value, k i (x est ,y est ) indicates that the i-th tactile unit is at (x est ,y est ) corresponding to the linear slope of the contact force information collected by the sensor at ;

[0036] The KNN algorithm in the above steps can usually use multiple distance metrics. The present invention uses the Euclidean distance metric. The parameter K in the KNN algorithm is set to 1, and the data can be directly classified into a certain position.

[0037] Furthermore, in order to better implement the above method, a system for realizing vaginal tactile super-resolution contact force measurement is also provided, the system comprising a data acquisition module, a calibration module, a model prediction module, and an image display module;

[0038] A data acquisition module, used to record the position information and contact force information during the contact force test of the acquisition array before the acquisition array leaves the factory, and to collect the positions in the vagina where the contact force measurement is required after the acquisition array leaves the factory;

[0039] A calibration module, used to calibrate the contact force information collected by the acquisition module to obtain an average tactile vector and a linear slope;

[0040] A model prediction module predicts the site in the vagina collected by the collection module where contact force measurement is required, and calculates the pressure value of the site through the model;

[0041] The image display module is used to display the prediction results of the model prediction module in real time.

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention constructs a high-precision measurement method and spatial distribution system for intra-vaginal contact force, aiming to solve the problems of high delay, low precision, large crosstalk, and low spatial resolution of traditional array-type pressure vaginal probes, thereby improving the contact perception ability and spatial resolution of vaginal bioforce, and reducing the complexity of physical integration and wiring, providing more basis for the diagnosis of pelvic floor diseases; the invention can accurately measure the bioforce exerted by the vaginal wall on the probe by directly measuring the contact force, and then draw a vaginal biomechanical map; at the same time, a relatively low-density sensing system is designed, with the help of a super-resolution pressure space prediction method, which not only reduces the arrangement complexity and process cost of high-density sensing units, but also achieves an effective improvement in the resolution of vaginal tactile perception through automatic calibration and prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A method flow chart of a system and method for realizing vaginal tactile super-resolution contact force measurement according to the present invention;

[0044] Figure 2 A schematic diagram of the system structure of a system and method for realizing vaginal tactile super-resolution contact force measurement according to the present invention;

[0045] Figure 3 A schematic diagram of the acquisition array structure of a system and method for realizing vaginal tactile super-resolution contact force measurement according to the present invention;

[0046] Figure 4 A schematic diagram of a linear slope corresponding to contact force information of a system and method for realizing vaginal tactile super-resolution contact force measurement according to the present invention;

[0047] Figure 5 Schematic diagram of the prediction result of a system and method for realizing vaginal tactile super-resolution contact force measurement according to the present invention. Specific implementation manners

[0048] Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0049] Embodiment: As Figure 1-Figure 5 shown, the present invention provides a technical solution, a method for realizing vaginal tactile super-resolution contact force measurement, the method includes:

[0050] Step S1: Construct an acquisition array and arrange the acquisition array on a calibration table, perform a contact force test on the acquisition array through the calibration table, and record the contact force information and position information during the contact force test;

[0051] Among them, referring to Figure 3 the structural schematic diagram of the acquisition array, it can be known that the acquisition array is a 2-row and N-column tactile array composed of 2×N tactile units, and each of the tactile units includes a sensor;

[0052] Among them, the contact force information in step S1 includes tactile data and pressure data. Among them, the tactile data includes the azimuth of each pressure applied to the acquisition system during the contact force test, and the pressure data includes the pressure value of each pressure applied to the acquisition system during the contact force test; the position information includes the coordinates corresponding to the sensor with a changing pressure value during the contact force test on the acquisition array;

[0053] Among them, in step S1, the contact force information and position information are recorded in the form of a matrix, and the form of the matrix is:

[0054] PosPress = {(Pos i (X, Y), Press i )};

[0055] Among them, PosPress represents the matrix for recording position information and contact force information, Pos i (X,Y) represents the characteristic parameter for recording the i-th position information collected by the acquisition array, and Press i represents the characteristic parameter for recording the i-th contact force information collected by the acquisition array, 0 < i ≤ 2N×Coe, Coe represents the calibration coefficient, and N represents the number of columns of the acquisition array;

[0056] Step S2: obtaining an average tactile vector and a linear slope by calibrating the contact force information;

[0057] In step S2, the contact force information collected by the collection array is calibrated using a calibration platform to obtain an average tactile vector and a linear slope, wherein the step of obtaining the average tactile vector and the linear slope includes:

[0058] Step S2-1: Name the sensor in each tactile unit and the pressure value collected by the sensor in the following manner:

[0059] Sensor a×b ;

[0060] P a×b ;

[0061] Wherein, a and b represent the horizontal and vertical serial numbers of the sensor in the tactile array, respectively, Sensor represents the sensor, and P represents the pressure value collected by the sensor;

[0062] Step S2-2: define the four sensors contained in the four adjacent tactile units in the tactile array as a sensing unit group, and the sensing unit group is expressed as follows:

[0063]

[0064] Among them, Group k represents the kth pressure sensing unit group;

[0065] Step S2-3: Calibrate the contact force information by Gaussian fitting to obtain an average tactile vector. The average tactile vector is obtained as follows:

[0066]

[0067] Among them, (μ xi ,μ yi ) is the mean vector of the contact force information collected by the four sensors around the i-th tactile unit at (x, y), m i is the amplitude value corresponding to the i-th contact force information, σ i is the fitting variance corresponding to the i-th contact force information, (x, y) represents the coordinates of the sensor corresponding to the contact force information, and p i (x, y) represents the average tactile vector of the i-th tactile unit at (x, y);

[0068] Step S2-4: Obtaining the linear slope corresponding to the contact force information by Gaussian fitting;

[0069]

[0070] Among them, v i represents the amplitude coefficient corresponding to the i-th contact force information, k i (x, y) represents the linear slope corresponding to the contact force information collected by the sensor of the i-th tactile unit at (x, y);

[0071] See also Figure 4 , in an embodiment of the present invention, the tactile data and the pressure data included in the contact force information are fitted by a two-dimensional Gaussian function to obtain a linear slope between the tactile data and the pressure data;

[0072] Step S3: constructing a training data set according to the average tactile vector and the linear slope and the position information, and inputting the training data set into a prediction model for training;

[0073] The training data set in step S3 includes a feature matrix and a target vector, wherein the feature matrix is ​​a two-column n-row matrix consisting of the average tactile vector and the linear slope, and the target vector is composed of n rows of the position information;

[0074] Step S4: collecting the sites in the vagina that need to be measured for contact force through the collection array, inputting the sites into the trained model for prediction, and transmitting the prediction results to the image display module;

[0075] In step S4, the KNN model is used to predict the site in the vagina where the contact force measurement is required, and the pressure value at the site is calculated by the KNN model. The calculation method is:

[0076] F target =min(d(X i , X j )) / k i (x est ,y est );

[0077] Among them, min(d(x i ,x j )) represents the minimum distance between the site (xi,yi) and the site (xj,yj) with the largest pressure value among the four surrounding sites, F target Indicates the pressure value, k i (x est ,y est ) indicates that the i-th tactile unit is at (x est ,y est ) corresponding to the linear slope of the contact force information collected by the sensor at ;

[0078] Among them, in order to better implement the above method, a system for realizing vaginal tactile super-resolution contact force measurement is also provided, the system includes a data acquisition module, a calibration module, a model prediction module, and an image display module;

[0079] A data acquisition module, used to record the position information and contact force information during the contact force test of the acquisition array before the acquisition array leaves the factory, and to collect the positions in the vagina where the contact force measurement is required after the acquisition array leaves the factory;

[0080] A calibration module, used to calibrate the contact force information collected by the acquisition module to obtain an average tactile vector and a linear slope;

[0081] A model prediction module predicts the site in the vagina collected by the collection module where contact force measurement is required, and calculates the pressure value of the site through the model;

[0082] The image display module is used to display the prediction results of the model prediction module in real time.

[0083] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for realizing vaginal tactile super-resolution contact force measurement, characterized in that: The method comprises: Step S1: constructing a collection array and placing the collection array on a verification table, performing a contact force test on the collection array through the verification table, and recording contact force information and position information during the contact force test; Step S2: obtaining an average tactile vector and a linear slope by calibrating the contact force information; Step S3: constructing a training data set according to the average tactile vector and the linear slope and the position information, and inputting the training data set into a prediction model for training; Step S4: The locations in the vagina that require contact force measurement are collected through the collection array, the locations are input into the trained model for prediction, and the prediction results are transmitted to the image display module.

2. The method for realizing vaginal tactile super-resolution contact force measurement according to claim 1, characterized in that: In step S1, the acquisition array is a tactile array of 2 rows and N columns consisting of 2×N tactile units, and each of the tactile units includes a sensor.

3. A method for realizing vaginal tactile super-resolution contact force measurement according to any one of claims 1-2, characterized in that: The contact force information in step S1 includes tactile data and pressure data, wherein the tactile data includes the orientation of each pressure applied to the acquisition system during the contact force test, and the pressure data includes the pressure value of each pressure applied to the acquisition system during the contact force test; the position information includes the coordinates of the sensor corresponding to the pressure value change during the contact force test of the acquisition array.

4. The method for realizing vaginal tactile super-resolution contact force measurement according to claim 1, characterized in that: In step S1, the contact force information and position information are recorded in a matrix format, and the matrix format is: PosPress={(Pos i (X,Y),Press i )}; Among them, PosPress represents a matrix for recording position information and contact force information, and Pos i (X, Y) represents a characteristic parameter for recording the i-th position information collected by the acquisition array, and Press i represents a characteristic parameter for recording the i-th contact force information collected by the acquisition array, where 0 < i ≤ 2N × Coe, Coe represents a calibration coefficient, and N represents the number of columns of the acquisition array.

5. The method for realizing vaginal tactile super-resolution contact force measurement according to claim 1, characterized in that: In step S2, the contact force information collected by the collection array is calibrated using a calibration platform to obtain an average tactile vector and a linear slope, wherein the step of obtaining the average tactile vector and the linear slope includes: Step S2-1: Name the sensor in each tactile unit and the pressure value collected by the sensor in the following manner: Sensor a×b ; P a×b ； Wherein, a and b represent the horizontal and vertical serial numbers of the sensor in the tactile array, respectively, Sensor represents the sensor, and P represents the pressure value collected by the sensor; Step S2-2: define the four sensors contained in the four adjacent tactile units in the tactile array as a sensing unit group, and the sensing unit group is expressed as follows: Among them, Group k represents the kth pressure sensing unit group; Step S2-3: Calibrate the contact force information by Gaussian fitting to obtain an average tactile vector. The average tactile vector is obtained as follows: Among them, (μ xi ,μ yi ) is the mean vector of the contact force information collected by the four sensors around the i-th tactile unit at (x, y), m i is the amplitude value corresponding to the i-th contact force information, σ i is the fitting variance corresponding to the i-th contact force information, (x, y) represents the coordinates of the sensor corresponding to the contact force information, and pi(x, y) represents the average tactile vector of the i-th tactile unit at (x, y); Step S2-4: Obtaining the linear slope corresponding to the contact force information by Gaussian fitting; Among them, v i represents the amplitude coefficient corresponding to the i-th contact force information, k i (x, y) represents the linear slope corresponding to the contact force information collected by the sensor of the i-th tactile unit at (x, y).

6. The method for realizing vaginal tactile super-resolution contact force measurement according to claim 1, characterized in that: The training data set described in step S3 includes a feature matrix and a target vector, wherein the feature matrix is ​​a two-column and n-row matrix consisting of the average tactile vector and the linear slope, and the target vector is composed of n rows of the position information.

7. The method for realizing vaginal tactile super-resolution contact force measurement according to claim 1, characterized in that: In step S4, the KNN model is used to predict the site in the vagina where the contact force measurement is required, and the pressure value at the site is calculated by the KNN model. The calculation method is: F target =min(d(X i ,X j )) / k i (x est ,y est ); Among them, min(d(x i ,x j )) represents the minimum distance between the site (xi,yi) and the site (xj,yj) with the largest pressure value among the four surrounding sites, F target Indicates the pressure value, k i (x est ,y est ) indicates that the i-th tactile unit is at (x est ,y est ) corresponds to the linear slope of the contact force information collected by the sensor at ).

8. A system for realizing vaginal tactile super-resolution contact force measurement, used for executing a method for realizing vaginal tactile super-resolution contact force measurement according to any one of claims 1 to 7, characterized in that: The system includes a data acquisition module, a calibration module, a model prediction module, and an image display module; The data acquisition module is used to record the position information and contact force information during the contact force test of the acquisition array before the acquisition array leaves the factory, and to collect the positions in the vagina where the contact force measurement is required after the acquisition array leaves the factory; The calibration module is used to calibrate the contact force information collected by the collection module to obtain an average tactile vector and a linear slope; The model prediction module predicts the site in the vagina collected by the collection module where contact force measurement is required, and calculates the pressure value of the site through the model; The image display module is used to display the prediction results of the model prediction module in real time.

Images