A length marking guide wire device for ultrasonic positioning

Abstract

This invention relates to the field of medical device technology and discloses an ultrasound-guided length marking guidewire device, comprising: a guidewire and a control module. The guidewire has a filler layer and a lubricating coating on its exterior. The lubricating coating covers the ends of the guidewire. Length markings are provided on both sides of the filler layer. The length markings are scales / digits / a combination of scales and digits. The control module includes an ultrasound component and an ultrasound module, which are electrically connected. The ultrasound component includes at least three ultrasound sensors evenly arranged in the filler layer. An ultrasound acquisition unit sets the acquisition frequency of the three ultrasound sensors. An ultrasound judgment unit determines the reflected wave signal based on the sensor parameters of the three ultrasound sensors and a tissue propagation model. Based on the comparison between the reflected wave signal and the target echo signal, it determines whether there is a positioning deviation. This invention ensures the stability of distance reading and guidewire positioning.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical devices, in particular to an ultrasonic positioning length identification guide wire device. BACKGROUND

[0002] In the process of endoscopic diagnosis and treatment, guide wire insertion technology is widely used in the diagnosis and treatment of diseases in multiple cavity organs such as digestive tract, respiratory tract and urinary tract. At present, guide wire insertion technology mainly relies on the observation of the guide wire advancing state of the endoscope by the operator, and the approximate position of the guide wire is judged by combining the clinical experience of the operator. Since the endoscope cannot intuitively judge and read the accurate position and advancing distance of the guide wire in the body, the spatial information of the guide wire in the body cannot be obtained, and the operator cannot accurately master whether the guide wire reaches the predetermined lesion area. Moreover, the guide wire lacks effective positioning judgment mechanism in the body, and the three-dimensional spatial position of the guide wire in the body cannot be monitored in real time and accurately, so that the operator cannot comprehensively understand the direction of the guide wire, which not only affects the accuracy of the operation, but also increases the operation time and operation difficulty.

[0003] Therefore, it is necessary to design an ultrasonic positioning length identification guide wire device to solve the problems existing in the prior art. SUMMARY

[0004] In view of this, the present application provides an ultrasonic positioning length identification guide wire device to solve the above problems.

[0005] The present application provides an ultrasonic positioning length identification guide wire device, comprising:

[0006] The guide wire is externally provided with a filling layer and a lubricating coating, the lubricating coating wraps the port of the guide wire, the filling layer and the lubricating coating are fixedly connected, and length identifiers are arranged on both sides of the filling layer. The length identifier is a scale / number / a combination of the scale and the number.

[0007] The material of the guide wire is nickel-titanium alloy / stainless steel;

[0008] The material of the filling layer is polyurethane / polytetrafluoroethylene;

[0009] The control module comprises an ultrasonic assembly and an ultrasonic module, the ultrasonic assembly and the ultrasonic module are electrically connected, the ultrasonic assembly comprises at least three ultrasonic sensors, the three ultrasonic sensors are uniformly arranged on the filling layer, and the ultrasonic module comprises an ultrasonic acquisition unit and an ultrasonic judgment unit.

[0010] The ultrasonic acquisition unit is used to set the acquisition frequency of the three ultrasonic sensors, and determine the target echo signal based on the acquisition frequency;

[0011] The ultrasonic judgment unit determines a reflected wave signal based on sensor parameters of the three ultrasonic sensors and a tissue propagation model, and judges whether there is a positioning deviation based on a comparison result of the reflected wave signal and a target echo signal.

[0012] Further, when setting a collection frequency of the three ultrasonic sensors and determining a target echo signal based on the collection frequency, the method comprises:

[0013] The collection frequency is proportional to the height of the patient to be inserted;

[0014] The ultrasonic collection unit determines three initial echo signals based on the collection frequency, and determines a noise reduction echo signal by performing noise reduction processing on the three initial echo signals, determines a square mean value of the noise reduction echo signal as an echo signal power, and acquires a noise power;

[0015] An echo signal-to-noise ratio is determined based on the echo signal power and the noise power, and a target echo signal is determined based on whether to compensate the noise reduction echo signal based on the echo signal-to-noise ratio.

[0016] Further, when determining a target echo signal based on whether to compensate the noise reduction echo signal based on the echo signal-to-noise ratio, the method comprises:

[0017] The echo signal-to-noise ratio is compared with an echo signal-to-noise ratio threshold value, when the echo signal-to-noise ratio is greater than or equal to the echo signal-to-noise ratio threshold value, it is determined that the noise reduction echo signal is not compensated, and the noise reduction echo signal is determined as the target echo signal;

[0018] When the echo signal-to-noise ratio is less than the echo signal-to-noise ratio threshold value, it is determined that the noise reduction echo signal is compensated, and the compensated noise reduction echo signal is determined as the target echo signal.

[0019] Further, when compensating the noise reduction echo signal, the method comprises:

[0020] An echo Doppler frequency of the noise reduction echo signal is acquired, and the echo Doppler frequency is compared with a Doppler database;

[0021] The Doppler database comprises a plurality of historical echo Doppler frequencies and a plurality of historical phase compensation values, and each historical echo Doppler frequency corresponds to a historical phase compensation value;

[0022] A frequency similarity of the echo Doppler frequency and each historical echo Doppler frequency is acquired, and a phase of the echo Doppler frequency is compensated based on the frequency similarity.

[0023] Further, when compensating the phase of the echo Doppler frequency based on the frequency similarity, the method comprises:

[0024] When there is a historical echo Doppler frequency with a frequency similarity greater than a frequency similarity threshold in the Doppler database, if the historical echo Doppler frequency is unique, the historical phase compensation value corresponding to the historical echo Doppler frequency is used to compensate the phase of the echo Doppler frequency, and if the historical echo Doppler frequency is not unique, the mean value of the historical phase compensation values corresponding to each historical echo Doppler frequency is used to compensate the phase of the echo Doppler frequency.

[0025] When there is no historical echo Doppler frequency with a frequency similarity greater than a frequency similarity threshold in the Doppler database, the phase of the echo Doppler frequency is compensated based on a clustering algorithm.

[0026] Further, when the phase of the echo Doppler frequency is compensated based on the clustering algorithm, it includes:

[0027] The echo Doppler frequency and the Doppler database are taken as a clustering set to be clustered, the historical phase compensation value corresponding to each historical echo Doppler frequency in the clustering set to be clustered is extracted, the expected cluster number k is determined as 2, the parameters of the Gaussian distribution are initialized, and the responsibility value of the probability that each data in the clustering set to be clustered belongs to each Gaussian distribution is determined;

[0028] Based on the responsibility value, a clustering set corresponding to the echo Doppler frequency is determined, and the mean value of the historical phase compensation values in the clustering set is used to compensate the phase of the echo Doppler frequency.

[0029] Further, when the reflected wave signal is determined based on the sensor parameters of the three ultrasonic sensors and the tissue propagation model, it includes:

[0030] The ultrasonic judgment unit constructs the tissue parameters of the patient to be inserted, the tissue parameters of the age stage other than the patient to be inserted, and the ultrasonic data set into a model sample set, and divides the model sample set into a training set and a test set;

[0031] The grid search is used to find the establishment parameters and construct a generative adversarial network model, the training set is used to train the generative adversarial network model, and the test set is used to test the trained generative adversarial network model to determine the tissue propagation model.

[0032] Further, when the training set is used to train the generative adversarial network model, and the test set is used to test the trained generative adversarial network model to determine the tissue propagation model, it includes:

[0033] If the iteration step ratio of the current trained generative adversarial network model is 1:1, the training is stopped, and the current trained generative adversarial network model is determined as the organization propagation model;

[0034] If the iteration step ratio of the current trained generative adversarial network model is not 1:1, the adjustment direction is determined based on the relationship between D and G, and the training is continued until the iteration step ratio of the trained generative adversarial network model is 1:1.

[0035] The sensor parameters are substituted into the organization propagation model to determine the reflected wave signal.

[0036] Further, when determining the adjustment direction based on the relationship between D and G, it includes:

[0037] If the discrimination ability of D is greater than the generation ability of G, a regularization term is added to D.

[0038] If the discrimination ability of D is less than the generation ability of G, the depth of the convolution layer / full connection layer of D is increased.

[0039] Further, when determining whether there is a positioning deviation based on the comparison result of the reflected wave signal and the target echo signal, it includes:

[0040] If the reflected wave signal is consistent with the target echo signal, it is determined that there is no positioning deviation.

[0041] If the reflected wave signal is inconsistent with the target echo signal, it is determined that there is a positioning deviation, and manual verification is prompted.

[0042] Compared with the prior art, the beneficial effects of the present application are that the scales, numbers or length marks combined with both on both sides of the guide wire provide intuitive length quantification reference, realize visualization of the length of the guide wire, provide length dimension data support for the precision of surgical operation, the filler layer selects polyurethane or polytetrafluoroethylene, which has good biocompatibility and structural stability, on the one hand, it can protect the guide wire of internal nickel-titanium alloy or stainless steel material, on the other hand, it provides a stable installation carrier for the arrangement of the ultrasonic sensor, the lubricating coating is wrapped around the port of the guide wire, reduces the friction resistance of the guide wire when advancing in the body cavity, respiratory tract and other body cavities, improves the smoothness of the guide wire placement, reduces the risk of damage to the body tissue, at least three ultrasonic sensors are evenly arranged in the filler layer, and cooperate with the ultrasonic acquisition unit and the ultrasonic judgment unit of the ultrasonic module to construct a spatial positioning system, which can judge in real time and accurately whether the guide wire has a positioning deviation. Realize real-time monitoring of the spatial position of the guide wire in the body, so that the operator can fully master the guide wire direction, and avoid operation deviation caused by position judgment error. BRIEF DESCRIPTION OF DRAWINGS

[0043] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a better understanding of the preferred embodiments, and are not intended to limit the scope of the application, and, wherein:

[0044] Figure 1 A structural schematic diagram of an ultrasonic positioning length identification guide wire device provided by an embodiment of the present application;

[0045] Figure 2 A cross-sectional view of the middle part of an ultrasonic positioning length identification guide wire device provided by an embodiment of the present application;

[0046] Figure 3 A cross-sectional view at the guide wire port provided by an embodiment of the present application;

[0047] Figure 4 A structural schematic diagram of a control module provided by an embodiment of the present application;

[0048] Wherein: 1, guide wire; 2, filling layer; 3, lubricating coating; 4, length identification. DETAILED DESCRIPTION

[0049] Exemplary embodiments of the present disclosure will be described in detail with reference to the drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood, and the scope of the present disclosure can be accurately conveyed to those skilled in the art. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0050] Reference Figures 1-4As shown, in some embodiments of the present application, an ultrasonic positioning length identification guide wire device comprises: a guide wire 1, an external filling layer 2 and a lubricating coating 3 are arranged on the guide wire 1, the lubricating coating 3 wraps the port of the guide wire 1, the filling layer 2 and the lubricating coating 3 are fixedly connected, length marks 4 are arranged on both sides of the filling layer 2, the length marks 4 are scales / numbers / combination of scales and numbers, the material of the guide wire 1 is nickel-titanium alloy / stainless steel, the material of the filling layer 2 is polyurethane / polytetrafluoroethylene, a control module, the control module comprises an ultrasonic assembly and an ultrasonic module, the ultrasonic assembly and the ultrasonic module are electrically connected, the ultrasonic assembly comprises at least three ultrasonic sensors, the three ultrasonic sensors are uniformly arranged on the filling layer 2, the ultrasonic module comprises an ultrasonic acquisition unit and an ultrasonic judgment unit, the ultrasonic acquisition unit is used for setting the acquisition frequency of the three ultrasonic sensors, and determines a target echo signal based on the acquisition frequency, the ultrasonic judgment unit determines a reflected wave signal based on the sensor parameters of the three ultrasonic sensors and a tissue propagation model, and judges whether there is a positioning deviation based on the comparison result of the reflected wave signal and the target echo signal.

[0051] Specifically, the material of the guide wire 1 is selected from nickel-titanium alloy or stainless steel. Such materials have excellent flexibility and strength. The flexibility enables the guide wire 1 to conform to the natural curvature of the digestive tract, respiratory tract, urinary tract and other cavity organs, smoothly extend into the body, and the strength ensures that the guide wire 1 is not easily broken during the advancing process, laying a foundation for subsequent diagnosis and treatment operations. The filling layer 2 is made of polyurethane or polytetrafluoroethylene and wrapped outside the guide wire 1. On the one hand, it protects the guide wire 1 inside, reducing damage caused by direct friction between the guide wire 1 and the cavity tissue in the body. On the other hand, the filling layer 2 provides an installation carrier for the length marker 4 and the ultrasonic sensor, enabling the length marker 4 to be stably distributed on both sides of the filling layer 2, and allowing the ultrasonic sensor to be evenly arranged on the surface of the filling layer 2, ensuring the stability of the position of each component and laying a foundation for the accuracy of subsequent length reading and ultrasonic positioning. The lubricating coating 3 is wrapped around the port of the guide wire 1 and is fixedly connected with the filling layer 2. Its function is to reduce the friction between the port of the guide wire 1 and the cavity tissue (such as mucosa, pipe wall, etc.) in the body, making the guide wire 1 more smooth during insertion, reducing the stimulation and damage to the cavity tissue, and facilitating the operator to push, adjust and operate the guide wire 1. The length marker 4 is arranged on both sides of the filling layer 2, which can be in the form of a scale, a number or a combination of a scale and a number. During operation, the operator can intuitively read the length of the guide wire 1 entering the body through the length marker 4, thereby roughly judging the position of the guide wire 1 in the body and assisting in determining whether the guide wire 1 approaches the predetermined lesion area, making up for the deficiency of the traditional endoscope that cannot directly read the advancing distance of the guide wire 1. The control module includes an ultrasonic assembly and an ultrasonic module. The ultrasonic assembly includes at least three ultrasonic sensors evenly arranged on the filling layer 2. The multiple ultrasonic sensors emit and receive target echo signals from different angles, thereby comprehensively capturing the in-vivo tissue information around the guide wire 1 and providing multi-dimensional data support for the judgment of the three-dimensional spatial position of the guide wire 1. The ultrasonic acquisition unit is used to set the acquisition frequency of the three ultrasonic sensors and determine the target echo signal based on the frequency. Setting a unified acquisition frequency makes the work of the ultrasonic sensors consistent and comparable. The ultrasonic judgment unit determines the reflected wave signal based on the sensor parameters (reflecting the performance characteristics of the sensor itself) of the three ultrasonic sensors and the tissue propagation model (summarizing the propagation law of ultrasonic waves in different tissues in the body). By comparing the reflected wave signal with the target echo signal, the real-time monitoring of the spatial position of the guide wire 1 in the body is realized, and the operator can clearly grasp the direction of the guide wire 1, avoiding the guide wire 1 from touching normal tissues or deviating from the lesion area, which not only solves the problem that the advancing distance of the traditional guide wire 1 cannot be quantified, but also makes up for the defect that the spatial direction cannot be monitored. The operator obtains basic advancing information through the length marker 4, and then masters the three-dimensional position and deviation of the guide wire 1 in real time through ultrasonic positioning, thereby comprehensively improving the accuracy, safety and efficiency of the guide wire 1 insertion operation.

[0052] In some embodiments of the present application, when the collection frequency of the three ultrasonic sensors is set and the target echo signal is determined based on the collection frequency, it includes: the collection frequency is proportional to the height of the patient to be inserted, the ultrasonic collection unit determines three initial echo signals based on the collection frequency, and determines a denoising echo signal by denoising processing on the three initial echo signals, determines the square mean value of the denoising echo signal as the echo signal power, acquires the noise power, determines the echo signal-to-noise ratio based on the echo signal power and the noise power, and determines whether to compensate the denoising echo signal based on the echo signal-to-noise ratio to determine the target echo signal.

[0053] In some embodiments of the present application, when the echo signal-to-noise ratio is determined to determine whether to compensate the denoising echo signal to determine the target echo signal, it includes: comparing the echo signal-to-noise ratio with the echo signal-to-noise ratio threshold, when the echo signal-to-noise ratio is greater than or equal to the echo signal-to-noise ratio threshold, it is determined that the denoising echo signal is not compensated, and the denoising echo signal is determined as the target echo signal, when the echo signal-to-noise ratio is less than the echo signal-to-noise ratio threshold, it is determined that the denoising echo signal is compensated, and the compensated denoising echo signal is determined as the target echo signal.

[0054] Specifically, when the collection frequency of the three ultrasonic sensors is set and the target echo signal is determined, first, the collection frequency is set according to the height of the patient to be inserted. The depth and path length of the cavity organs in the body (such as the digestive tract, respiratory tract, etc.) of patients of different heights are different, and the appropriate frequency is different when the ultrasonic wave propagates in the tissue path of different lengths. For taller people, the tissue path may be longer, and a higher collection frequency needs to be adapted to ensure signal propagation efficiency. For shorter people, the opposite is true. Therefore, the collection frequency is proportional to the height to adapt to the differences of different individuals. The ultrasonic collection unit acquires three initial echo signals through the collection frequency and denoises the initial echo signals to filter out possible interference noise and determine a relatively pure denoising echo signal. The square mean value of the denoising echo signal is calculated to determine the echo signal power, and the noise power is determined by collecting the signal in the period when the ultrasonic sensor does not receive the initial echo signal. Specifically, the idle time before the ultrasonic sensor transmits the ultrasonic wave or the signal period when the guide wire 1 is in the uniform tissue area without obvious reflection can be selected, and the signal in this period is collected. At this time, the signal is mainly composed of various interference noises, and the noise power is obtained in this way. The echo signal-to-noise ratio is determined by the following formula:

[0055]

[0056] The SNR represents an echo signal-to-noise ratio, P represents an echo signal power, P0 represents a noise power, and the signal power is calculated based on a square mean value of the noise-reduced echo signal. The echo signal-to-noise ratio is compared with a preset echo signal-to-noise ratio threshold value. If the echo signal-to-noise ratio is greater than or equal to the echo signal-to-noise ratio threshold value, it indicates that the signal quality of the noise-reduced echo signal after noise reduction meets the standard, and the noise-reduced echo signal is directly determined as the target echo signal. If the echo signal-to-noise ratio is less than the echo signal-to-noise ratio threshold value, it indicates that the signal quality of the noise-reduced echo signal after noise reduction does not meet the standard, and the noise-reduced echo signal needs to be compensated to ensure the reliability of the target echo signal and provide an accurate benchmark for the subsequent ultrasonic judgment unit to compare the reflection signal with the target echo signal to judge the positioning deviation.

[0057] In some embodiments of the present application, when the noise-reduced echo signal is compensated, the echo Doppler frequency of the noise-reduced echo signal is obtained, and the echo Doppler frequency is compared with a Doppler database. The Doppler database includes a plurality of historical echo Doppler frequencies and a plurality of historical phase compensation values, and each historical echo Doppler frequency corresponds to a historical phase compensation value. The frequency similarity of the echo Doppler frequency and each historical echo Doppler frequency is obtained, and the phase of the echo Doppler frequency is compensated based on the frequency similarity.

[0058] In some embodiments of the present application, when the phase of the echo Doppler frequency is compensated based on the frequency similarity, when there is a historical echo Doppler frequency with a frequency similarity greater than a frequency similarity threshold value in the Doppler database, if the historical echo Doppler frequency is unique, the historical phase compensation value corresponding to the historical echo Doppler frequency is used to compensate the phase of the echo Doppler frequency. If the historical echo Doppler frequency is not unique, the mean value of the historical phase compensation values corresponding to each historical echo Doppler frequency is used to compensate the phase of the echo Doppler frequency. When there is no historical echo Doppler frequency with a frequency similarity greater than a frequency similarity threshold value in the Doppler database, the phase of the echo Doppler frequency is compensated based on a clustering algorithm.

[0059] Specifically, although the noise-reduced echo signal removes part of the noise, the phase of the echo Doppler frequency may be shifted due to the influence of the relative motion of the guide wire 1 advancing, in-vivo tissue peristalsis, etc., thereby causing distortion of the signal quality. The echo Doppler frequency reflects the signal characteristic changes generated by the relative motion of the guide wire 1 and the surrounding tissue. By comparing the echo Doppler frequency with the Doppler database, the Doppler database stores a plurality of historical echo Doppler frequencies and corresponding historical phase compensation values, and each historical frequency has a unique phase compensation record. Through frequency similarity comparison, similar scene compensation experience can be quickly matched to ensure the pertinence and consistency of compensation. The frequency similarity can be determined according to the cosine similarity and the Euclidean distance. In the embodiment, the frequency similarity threshold is preferably 0.85. When there is a historical echo Doppler frequency with a frequency similarity exceeding the frequency similarity threshold in the Doppler database, if there is only one historical echo Doppler frequency meeting the condition, the corresponding historical phase compensation value of the historical echo Doppler frequency can be directly used to compensate the phase of the current echo Doppler frequency. If there are multiple historical echo Doppler frequencies meeting the condition, the mean value of the historical phase compensation values corresponding to these historical echo Doppler frequencies is taken as the compensation basis. When there is no historical frequency with a frequency similarity exceeding the frequency similarity threshold in the Doppler database, the echo Doppler frequency is classified with the data with similar characteristics in the Doppler database through a clustering algorithm, and the compensation value is generated based on the common characteristics of the similar data to compensate the phase. By matching the data of the Doppler database and compensating the phase, the historical compensation scheme can be quickly reused, the compensation efficiency and accuracy are improved, the signal distortion caused by the phase shift is reduced, and the positioning accuracy of the guide wire 1 and the safety of the surgical operation are further improved.

[0060] In some embodiments of the present application, when the phase of the echo Doppler frequency is compensated based on the clustering algorithm, the echo Doppler frequency and the Doppler database are taken as the clustering set to be clustered, the historical phase compensation value corresponding to each historical echo Doppler frequency in the clustering set to be clustered is extracted, the expected cluster number k is determined as 2, the parameters of the Gaussian distribution are initialized, the responsibility value of the probability that each data in the clustering set to be clustered belongs to each Gaussian distribution is determined, the clustering set corresponding to the echo Doppler frequency is determined based on the responsibility value, and the mean value of the historical phase compensation values in the clustering set is used to compensate the phase of the echo Doppler frequency.

[0061] Specifically, by analyzing the Doppler database through a clustering algorithm, when there is a lack of directly similar historical data, the closest cluster set to the current operation condition can still be found based on the data feature correlation, improving the accuracy of compensating the phase of the echo Doppler frequency, thereby maintaining the stability and efficiency of the positioning process, improving the overall automation level and reliability, compensating the phase of the echo Doppler frequency with the mean value of the historical phase compensation values in the cluster set, balancing the individual differences between the data, reducing the influence of single data deviation, and ensuring the reliability of the compensation result.

[0062] In some embodiments of the present application, when determining the reflected wave signal based on the sensor parameters of the three ultrasonic sensors and the tissue propagation model, the ultrasonic judgment unit constructs the tissue parameters of the patient to be inserted, the tissue parameters of the non-patient age stage to be inserted, and the ultrasonic data set into a model sample set, divides the model sample set into a training set and a test set, finds the establishment parameters based on grid search, and constructs a generative adversarial network model, trains the generative adversarial network model based on the training set, and tests the trained generative adversarial network model based on the test set to determine the tissue propagation model.

[0063] Specifically, the tissue parameters of the patient to be inserted include the tissue density, tissue elasticity (which determines the ultrasonic reflection intensity), and thickness (such as the thickness of the esophageal wall and tracheal wall) of the cavity organs (such as the esophagus / intestine of the digestive tract, the trachea / bronchus of the respiratory tract, and the ureter / bladder of the urinary tract), and the tissue parameters of the non-patient age stage to be inserted are consistent with the type and dimension of the tissue parameters of the patient to be inserted. The core is to provide a reference for the characteristics of tissues of different ages to reduce the interference of individual differences on the model, thereby balancing individual specificity and group commonality. The ultrasonic data set includes ultrasonic transmission intensity, propagation time (corresponding to propagation path length) of ultrasonic waves in different tissues (such as normal mucosa, muscle layer, and adipose tissue), and energy attenuation degree (directly related to tissue density and thickness) during propagation. Integrating these three types of data into a model sample set ensures the generalization ability of the model. The model sample set is divided into a training set and a test set, usually with a division ratio of 4:1. The training set is used for model learning, and the test set is used for verifying the model effect. Grid search finds the optimal establishment parameters (i.e., configuration information of the model) in the parameter space. Based on these establishment parameters, a generative adversarial network model is constructed. The generative adversarial network model is good at capturing the mapping relationship between complex data and is suitable for simulating the dynamic process of the interaction between ultrasonic waves and different tissues. The training set is used to train the generative adversarial network model to learn the propagation law of ultrasonic waves under different tissue parameters. Then, the test set is used to test the output effect of the trained model. The tissue propagation model is finally determined through verification, further improving the positioning accuracy of the guide wire 1.

[0064] In some embodiments of the present application, when the generative adversarial network model is trained based on the training set and the trained generative adversarial network model is tested based on the test set to determine the tissue propagation model, if the iteration step ratio of the current trained generative adversarial network model is 1:1, the training is stopped, and the current trained generative adversarial network model is determined as the tissue propagation model; if the iteration step ratio of the current trained generative adversarial network model is not 1:1, the adjustment direction is determined based on the relationship between D and G, and the training is continued until the iteration step ratio of the trained generative adversarial network model is 1:1, and the sensor parameters are substituted into the tissue propagation model to determine the reflected wave signal.

[0065] In some embodiments of the present application, when the adjustment direction is determined based on the relationship between D and G, if the discrimination ability of D is greater than the generation ability of G, a regularization term is added to D; if the discrimination ability of D is less than the generation ability of G, the depth of the convolution layer / full connection layer of D is increased.

[0066] Specifically, when the generative adversarial network model is trained based on the training set, the generative adversarial network model includes a generator (G) and a discriminator (D), which learn the propagation rule of ultrasonic waves in tissue through adversarial game. In the training process, the iteration step ratio reflects the overall performance level of the model. If the current iteration step ratio is 1:1, the simulated data generated by G can approach the ultrasonic wave propagation characteristics in the real tissue, and D can objectively judge the authenticity of the data. The model can accurately capture the tissue propagation rule and has reached the expected level of training. At this time, the training can be stopped, and the current trained generative adversarial network model is determined as the tissue propagation model. If the iteration step ratio is not 1:1, it means that the abilities of the two are not balanced. If D is too strong, it can easily identify the simulated data generated by G, which leads to G unable to effectively learn the real propagation rule. If G is too strong, the simulated data generated by G can "fool" D, so that the model learns a rule deviating from the actual rule. Therefore, the relationship between the abilities of D and G needs to be adjusted. When adjusting, if the discrimination ability (the ability to distinguish between real data and simulated data generated by G) of D is stronger than the generation ability (the ability to generate simulated data close to real ultrasonic wave propagation data) of G, a regularization term is added to D to limit its overfitting to the training data to weaken the discrimination ability. If the discrimination ability of D is weaker than the generation ability of G, the depth of the convolution layer or the full connection layer of D is increased to enhance its ability to extract and distinguish data features. The training and adjustment are repeated until the iteration step ratio can reach 1:1, avoiding the risk of model failure due to the over-strength of one side, improving the stability and efficiency of model training, so as to determine the reflected wave signal according to the tissue propagation model, ensuring the reliability of the guide wire 1 positioning.

[0067] In some embodiments of the present application, when judging whether there is a positioning deviation based on the comparison result of the reflected wave signal and the target echo signal, if the reflected wave signal is consistent with the target echo signal, it is determined that there is no positioning deviation, and if the reflected wave signal is inconsistent with the target echo signal, it is determined that there is a positioning deviation, and a manual verification is prompted.

[0068] Specifically, the target echo signal is an actual feedback reflecting the state of the guide wire 1 in real time, and the reflected wave signal is a reference optimized through multiple links. When judging the positioning deviation, the target echo signal is taken as the reference. After noise reduction, compensation and other processing, the signal reflects the actual position of the guide wire 1, and the reflected wave signal output by the tissue propagation model reflects the expected position of the guide wire 1. The comparison of the two can directly relate the expected and actual positions of the guide wire 1. If the characteristics of the two are completely matched, it means that the actual position of the guide wire 1 is consistent with the expected position, and it is determined that there is no positioning deviation. If the characteristics of the two are different, such as phase and amplitude, it means that the actual position of the guide wire 1 deviates from the expected position, and it is determined that there is a positioning deviation. At the same time, the prompting mechanism is triggered to prompt the operator to perform manual verification, so as to realize the rapid judgment of the positioning deviation, form the double protection of comparison judgment and manual verification, and ensure the accuracy and reliability of the positioning and distance reading of the guide wire 1.

[0069] In summary, the beneficial effects of the present application are that the scales, numbers or length marks combined with the two on both sides of the guide wire provide intuitive length quantification reference, realize the visualization of the length of the guide wire, provide length dimension data support for the accuracy of surgical operation, the filler layer is made of polyurethane or polytetrafluoroethylene, which has good biocompatibility and structural stability. On the one hand, it can protect the guide wire made of nickel-titanium alloy or stainless steel inside, and on the other hand, it provides a stable mounting carrier for the arrangement of the ultrasonic sensor. The lubricating coating wrapped around the port of the guide wire reduces the frictional resistance of the guide wire when advancing in the body cavity, respiratory tract and other body cavities, improves the smoothness of the guide wire insertion, reduces the risk of damage to the body tissue, and at least three ultrasonic sensors are evenly arranged in the filler layer and cooperate with the ultrasonic acquisition unit and the ultrasonic judgment unit of the ultrasonic module to construct a spatial positioning system, which can judge whether the guide wire has a positioning deviation in real time and accurately. Real-time monitoring of the spatial position of the guide wire in the body enables the operator to fully grasp the guide wire direction and avoid operation deviation caused by position judgment error.

[0070] Those skilled in the art will appreciate that embodiments of the application can be supplied as a method, a system or a computer program product. Accordingly, the application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the application can be embodied in the form of a computer program product on one or more computer readable storage media (including, without limitation, magnetic disks; optical disks; magneto-optical disks; ROMs; flash memory; etc.) having computer usable program code embodied therein.

[0071] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing the one or more steps of the flowchart block or blocks.

[0072] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing the one or more steps of the flowchart block or blocks.

[0073] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more functions specified in the flowchart block or blocks. Figure 1 means for functionally implementing the one or more steps of the flowchart block or blocks.

[0074] Finally, it should be noted that the above-mentioned embodiments are merely intended for describing and illustrating, not limiting, the technical solutions of the present application, and although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and any modifications or replacements made without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application.

Claims

1. An ultrasonically positioned length-marking guidewire device, characterized by, The utility model relates to a kind of medical device, including: guide wire, the guide wire outside is provided with filling layer and lubricating coating, the lubricating coating is wrapped the port of the guide wire, the filling layer and lubricating coating are fixedly connected, the length mark is set on the two sides of the filling layer, the length mark is scale / number / combination of the scale and number; The material of the guide wire is nickel-titanium alloy / stainless steel; The material of the filling layer is polyurethane / polytetrafluoroethylene; Control module, the control module includes ultrasonic assembly and ultrasonic module, the ultrasonic assembly and ultrasonic module are electrically connected, the ultrasonic assembly includes at least three ultrasonic sensors, three the ultrasonic sensor is uniformly arranged in the filling layer, the ultrasonic module includes ultrasonic acquisition unit and ultrasonic judgment unit; The ultrasonic acquisition unit is used to set the acquisition frequency of three the ultrasonic sensor, and determine target echo signal based on the acquisition frequency; The ultrasonic judgment unit determines reflected wave signal based on the sensor parameter of three the ultrasonic sensor and tissue propagation model, and judges whether there is positioning deviation based on the comparison result of reflected wave signal and target echo signal; When setting the acquisition frequency of three the ultrasonic sensor, and determining target echo signal based on the acquisition frequency, it includes: The acquisition frequency is directly proportional to the height of the patient to be inserted; The ultrasonic acquisition unit determines three initial echo signals based on the acquisition frequency, and determines noise reduction echo signal by noise reduction processing to three the initial echo signal, determines the square mean of the noise reduction echo signal as echo signal power, and obtains noise power; Determine echo signal-to-noise ratio based on the echo signal power and the noise power, and determine target echo signal based on the echo signal-to-noise ratio whether to compensate the noise reduction echo signal; When determining target echo signal based on the echo signal-to-noise ratio whether to compensate the noise reduction echo signal, it includes: Compare the echo signal-to-noise ratio with echo signal-to-noise ratio threshold, when the echo signal-to-noise ratio is greater than or equal to the echo signal-to-noise ratio threshold, it is determined that the noise reduction echo signal is not compensated, and the noise reduction echo signal is determined as target echo signal; When the echo signal-to-noise ratio is less than the echo signal-to-noise ratio threshold, it is determined that the noise reduction echo signal is compensated, and the compensated noise reduction echo signal is determined as target echo signal; When compensating the noise reduction echo signal, it includes: Obtain echo Doppler frequency of the noise reduction echo signal, and compare the echo Doppler frequency with Doppler database; The Doppler database includes a plurality of historical echo Doppler frequencies and a plurality of historical phase compensation values, and each historical echo Doppler frequency corresponds to a historical phase compensation value; Obtain the frequency similarity of the echo Doppler frequency and each historical echo Doppler frequency, and compensate the phase of the echo Doppler frequency based on the frequency similarity; When compensating the phase of the echo Doppler frequency based on the frequency similarity, it includes: ​ When the historical echo Doppler frequency with the frequency similarity greater than the frequency similarity threshold exists in the Doppler database, if the historical echo Doppler frequency is unique, the historical phase compensation value corresponding to the historical echo Doppler frequency is used to compensate the phase of the echo Doppler frequency; if the historical echo Doppler frequency is not unique, the mean value of the historical phase compensation values corresponding to each historical echo Doppler frequency is used to compensate the phase of the echo Doppler frequency; When the historical echo Doppler frequency with the frequency similarity greater than the frequency similarity threshold does not exist in the Doppler database, the phase of the echo Doppler frequency is compensated based on a clustering algorithm.

2. The ultrasonically positioned length-marking guidewire device of claim 1, wherein, When the phase of the echo Doppler frequency is compensated based on the clustering algorithm, the method comprises: The echo Doppler frequency and the Doppler database are taken as a clustering set to be clustered, the historical phase compensation value corresponding to each historical echo Doppler frequency in the clustering set to be clustered is extracted, the expected cluster number k is determined as 2, the parameters of the Gaussian distribution are initialized, and the responsibility value of the probability that each data in the clustering set to be clustered belongs to each Gaussian distribution is determined; The clustering set corresponding to the echo Doppler frequency is determined based on the responsibility value, and the mean value of the historical phase compensation values in the clustering set is used to compensate the phase of the echo Doppler frequency.

3. The ultrasonically positioned length-marking guidewire device of claim 2, wherein, When the reflected wave signal is determined based on the sensor parameters of the three ultrasonic sensors and the tissue propagation model, the method comprises: The ultrasonic judgment unit constructs the tissue parameters of the patient to be inserted, the tissue parameters of the patient not in the age range to be inserted, and the ultrasonic data set into a model sample set, and divides the model sample set into a training set and a test set; The establishment parameters are searched based on the grid search, the generative adversarial network model is constructed, the generative adversarial network model is trained based on the training set, and the trained generative adversarial network model is tested based on the test set to determine the tissue propagation model.

4. The ultrasonically positioned length-marking guidewire device of claim 3, wherein, When the generative adversarial network model is trained based on the training set, and the trained generative adversarial network model is tested based on the test set to determine the tissue propagation model, the method comprises: If the iteration step ratio of the current trained generative adversarial network model is 1:1, the training is stopped, and the current trained generative adversarial network model is determined as the tissue propagation model; If the iteration step ratio of the current trained generative adversarial network model is not 1:1, the adjustment direction is determined based on the relationship between D and G, and the training is continued until the iteration step ratio of the trained generative adversarial network model is 1:1; The sensor parameters are substituted into the tissue propagation model to determine the reflected wave signal.

5. The ultrasonically positioned length marker guidewire device of claim 4, wherein, When the adjustment direction is determined based on the relationship between D and G, the method comprises: If the discrimination ability of D is greater than the generation ability of G, a regularization term is added to D; If the discrimination ability of D is less than the generation ability of G, the depth of the convolution layer / full connection layer of D is increased.

6. The ultrasonically positioned length-marking guidewire device of claim 5, wherein, When it is determined whether there is a positioning deviation based on the comparison result of the reflected wave signal and the target echo signal, the method comprises: If the reflected wave signal is consistent with the target echo signal, it is determined that there is no positioning deviation; If the reflected wave signal is inconsistent with the target echo signal, it is determined that there is a positioning deviation, and manual verification is prompted.

Images