Ultrasound recognition of markers in biological tissue

Abstract

The application discloses an ultrasonic recognition method of a biomarker in a biological tissue, and comprises the following steps: step one: obtaining a characteristic kernel of the biomarker; 11) collecting ultrasonic signals of the biomarker in static water to obtain a sample signal; 12) performing convolution operation on the sample signal and the sample signal itself to obtain the characteristic kernel; step two: obtaining a search signal stream of the biological tissue; 21) collecting ultrasonic signals of the biological tissue to obtain an original signal; 22) performing convolution operation on the original signal and the sample signal to obtain the search signal stream; step three: recognizing ultrasonic scanning information of the biomarker; 31) respectively calculating the first-order derivatives of the characteristic kernel and the search signal stream; 32) searching for a data segment with the same sign as the first-order derivative of the characteristic kernel in the first-order derivative of the search signal stream; and step four: if the biomarker exists, marking the biomarker signal segment in the original signal to obtain ultrasonic scanning information corresponding to the biomarker.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of ultrasonic detection, and particularly relates to an ultrasonic identification method for a marker in a biological tissue. BACKGROUND

[0002] In the current medical treatment of tumors and other diseases, a marker with a diameter of about 0.2 mm needs to be implanted into a tumor tissue for facilitating later examination. The marker is generally made of metal or composite material, and the main ways to identify the marker currently include the following two ways: the first way is to identify the position of the marker through X-ray; and the second way is to scan the marker by using a B-ultrasound instrument.

[0003] When the position of the marker is identified through X-ray, the position of the marker is identified in an X-ray picture by taking an X-ray chest film, and the disadvantage is that the depth position information of the marker is not available, the X-ray radiation is harmful to the human body in multiple photographing, and the ray equipment is large in size and is not conducive to examination in surgery or examination at any time by a doctor.

[0004] When the marker is scanned by using the B-ultrasound instrument, the B-ultrasound imaging method is based on the ultrasonic equivalent size of the reflection amplitude of a target object to process imaging, but since the ultrasonic reflection equivalent of the marker is very small, the marker and the human tissue are difficult to be distinguished in most cases, resulting in loss of the marker due to unidentifiability. In addition, the existing B-ultrasound processing flow of the ultrasonic signal is as follows: original wideband signal→filtering of a certain form of frequency spectrum filter conforming to Fourier spectrum theory→image processing→final display of a narrowband frequency imaging or a composite image of multiple narrowband frequency imagings. In the signal processing process, the signal processing system actively filters out the frequency information other than the specific frequency (human tissue frequency), and for the marker made of metal or composite material, the ultrasonic information thereof is actually lost. SUMMARY

[0005] Therefore, the purpose of the present application is to provide an ultrasonic identification method for a marker in a biological tissue, which can accurately identify the marker by using an ultrasonic scanning method.

[0006] To achieve the above purpose, the present application provides the following technical scheme.

[0007] An ultrasonic identification method for a marker in a biological tissue, comprising the following steps.

[0008] Step 1: obtaining a feature convolution kernel of the marker

[0009] 11) collecting an ultrasonic wave signal of the marker in still water to obtain a specimen signal;

[0010] 12) performing convolution operation on the specimen signal and itself to obtain a feature convolution kernel of the marker;

[0011] Step two: obtaining the retrieval signal stream of the biological tissue

[0012] 21) Collecting the ultrasonic signal of the biological tissue to obtain the original signal;

[0013] 22) After convolution operation of the original signal and the specimen signal, assigning zero value to the number less than the set threshold value to obtain the retrieval signal stream;

[0014] Step three: identifying the ultrasonic scanning information of the marker

[0015] 31) Calculating the first derivative of the feature convolution kernel and the retrieval signal stream respectively;

[0016] 32) Point by point searching in the first derivative of the retrieval signal stream whether there is a data segment consistent with the sign of the first derivative of the feature convolution kernel: if yes, the marker signal segment is obtained, and step four is executed; if not, it indicates that the marker does not exist in the biological tissue;

[0017] Step four: labeling the marker signal segment in the original signal to obtain the ultrasonic scanning information corresponding to the marker.

[0018] Further, in step 11), the ultrasonic signals of the marker in multiple different directions in still water are collected to obtain the specimen signals of the marker in multiple different directions.

[0019] Further, in step 12), the ultrasonic signals of the marker in each direction in still water are respectively convolved with themselves to obtain the feature convolution kernels of the marker in different directions, and all the feature convolution kernels constitute the feature convolution kernel set.

[0020] Further, in step 22), after the convolution operation of the original signal and the specimen signals of the marker in different directions respectively, the number less than the set threshold value is assigned to zero value to obtain the retrieval signal stream corresponding to the different directions of the marker.

[0021] Further, in step 31), the first derivative of all feature convolution kernels and all retrieval signal streams is calculated.

[0022] Further, in step 32), point by point searching in the first derivative of the retrieval signal stream corresponding to the ultrasonic scanning of the marker in different directions whether there is a data segment consistent with the sign of the first derivative of any feature convolution kernel in the feature convolution kernel set: if yes, the marker signal segment is obtained, and step four is executed; if not, it indicates that the marker does not exist in the biological tissue.

[0023] The beneficial effects of the present application are:

[0024] The ultrasonic recognition method of the marker in the biological tissue of the present application first collects the ultrasonic signal of the marker in static water as a specimen signal, which is a wideband original signal without filtering operation and does not lose ultrasonic information; performs convolution operation on the specimen signal and itself to obtain a characteristic convolution kernel for discriminating the marker; then collects the ultrasonic signal of the biological tissue as an original signal, which is also a wideband original signal stream without filtering operation; performs convolution operation on the original signal and the specimen signal to strengthen the marker signal characteristics in the original signal and attenuate non-target signals, and obtains a search signal stream after assigning zero value to the number less than the set threshold; finally, takes the first derivative of the characteristic convolution kernel and the search signal stream to describe the curvature change trend and eliminate the influence of the great difference in signal amplitude on signal comparison, and then searches the search signal stream point by point to see if there is a data segment with the same sign as the first derivative of the characteristic convolution kernel, so as to obtain whether the biological tissue contains the marker, if the biological tissue contains the marker, mark the marker signal segment in the original signal, and obtain the ultrasonic scanning information containing the marker position and depth and other information; in summary, the ultrasonic recognition method of the marker in the biological tissue of the present application can overcome the problem that the existing ultrasonic scanning is difficult to recognize due to the small ultrasonic reflection equivalent of the marker, and can accurately recognize the marker in the biological tissue.

[0025] The ultrasonic recognition method of the marker in the biological tissue of the present application has the following advantages:

[0026] (1) Compared with the traditional B-ultrasound image recognition, the recognition of the present application does not depend on the reflection amplitude of the marker, so the recognition rate is greatly improved, and the recognition rate can reach more than 85%, while the recognition rate of the traditional B-ultrasound image recognition is less than 20%;

[0027] (2) The present application enables the computer to autonomously recognize the marker signal and does not depend on the personal professional experience of the operator on the ultrasonic scanning, which is convenient for large-scale popularization and application;

[0028] (3) The recognition of the present application no longer depends on the image quality generated by the ultrasonic imaging system, and does not need a high-quality and high-priced two-dimensional scanning imaging system, but only uses a one-dimensional line scanning (A-mode) ultrasonic device, which has a great advantage in cost;

[0029] (4) The device of the present application can prompt the operator about the position and depth of the marker by the method of associating sound; and liberates the eyes of the medical staff from observing the two-dimensional ultrasonic image in the traditional mode. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to make the purpose, technical scheme and beneficial effects of the present application clearer, the present application provides the following drawings for illustration:

[0031] Figure 1 Flow chart of the method for ultrasonic recognition of the marker in the biological tissue of the present application;

[0032] Figure 2 Ultrasonic wave signals of the marker in the static water in 5 directions;

[0033] Figure 3 Specimen signals of the marker in 5 directions;

[0034] Figure 4 5 retrieval signal streams obtained after convolution operation of the original signals and the specimen signals in 5 directions;

[0035] Figure 5 Original signals marked with the signal segments of the marker. DETAILED DESCRIPTION

[0036] The present application will be further described below in conjunction with the drawings and specific embodiments so that those skilled in the art can better understand the present application and implement it. The embodiments are not intended to limit the present application.

[0037] As shown in the figure, the method for ultrasonic recognition of the marker in the biological tissue of the present embodiment comprises the following steps: Figure 1 Step 1: Obtain the characteristic convolution kernel of the marker

[0038] 11) Collect the ultrasonic wave signals of the marker in the static water to obtain the specimen signals. Specifically, the ultrasonic scanning signal characteristics of the marker in different directions can be different, and therefore, in actual application, the ultrasonic wave signals of the marker in multiple different directions in the static water can be collected to obtain the specimen signals of the marker in multiple different directions. As shown in the figure, the ultrasonic wave signals of the marker in 5 directions are collected in the present embodiment. Specifically, the ultrasonic wave signals of the marker in the static water collected in the present embodiment are wideband original signals without filtering operation, so as to avoid loss of ultrasonic information.

[0039] Figures 2-3 12) Perform convolution operation of the specimen signals with themselves to obtain the characteristic convolution kernel of the marker, which is used for discriminating the marker. Specifically, when the ultrasonic wave signals of the marker in the static water in multiple different directions are collected, then the ultrasonic wave signals of the marker in each direction in the static water are respectively subjected to convolution operation with themselves to obtain the characteristic convolution kernel of the marker in different directions, and all the characteristic convolution kernels constitute the constructed characteristic convolution kernel set.

[0040] Step 2: Obtain the retrieval signal stream of the biological tissue

[0041] Step 2: Obtain the retrieval signal stream of the biological tissue

[0042] ​21) Collecting the ultrasound signal of the biological tissue to obtain the original signal. Specifically, the original signal of the embodiment is also a wideband original signal stream without filtering operation to avoid loss of ultrasound information.

[0043] 22) After convolution operation of the original signal and the specimen signal, assigning zero value to the number less than the set threshold to obtain the retrieval signal stream. The purpose of the convolution operation of the original signal and the specimen signal is to strengthen the marker signal characteristics in the original signal while attenuating the non-target signal. When collecting the ultrasound signal of the marker in the still water in multiple different directions, the convolution operation of the original signal and the specimen signal of the marker in different directions is required, and then the zero value is assigned to the number less than the set threshold to obtain the retrieval signal stream corresponding to the different directions of the marker, as shown in FIG. 2. Specifically, the threshold of the embodiment can be set to 0. Figure 4

[0044] Step three: identifying the ultrasound scanning information of the marker

[0045] 31) Calculating the first derivative of the feature convolution kernel and the retrieval signal stream, respectively. When collecting the ultrasound signal of the marker in the still water in multiple different directions, the first derivative of all feature convolution kernels and all retrieval signal streams is calculated, respectively.

[0046] 32) Point by point searching in the first derivative of the retrieval signal stream to find whether there is a data segment with the same sign as the first derivative of the feature convolution kernel: if yes, the marker signal segment is obtained, and step four is executed; if no, it indicates that there is no marker in the biological tissue.

[0047] Specifically, when collecting the ultrasound signal of the marker in the still water in multiple different directions, point by point searching in the first derivative of the retrieval signal stream corresponding to the ultrasound scanning in different directions of the marker to find whether there is a data segment with the same sign as the first derivative of any one of the feature convolution kernel set: if yes, the marker signal segment is obtained, and step four is executed; if no, it indicates that there is no marker in the biological tissue.

[0048] Step four: marking the marker signal segment in the original signal to obtain the ultrasound scanning information corresponding to the marker. Through the ultrasound scanning information of the marker, the position and depth information of the marker can be obtained, so as to realize the identification of the marker, as shown in FIG. 3. Figure 5

[0049] ​​The ultrasonic recognition method of the marker in the biological tissue of the embodiment first collects the ultrasonic signal of the marker in the still water as a specimen signal, the specimen signal is a wideband original signal without filtering operation, and the ultrasonic information is not lost; the convolution operation of the specimen signal and itself is performed to obtain a characteristic convolution kernel for discriminating the marker; then the ultrasonic signal of the biological tissue is collected as an original signal, the original signal is also a wideband original signal stream without filtering operation; the convolution operation of the original signal and the specimen signal is performed to strengthen the marker signal characteristics in the original signal and attenuate the non-target signal, and after the number less than the set threshold is assigned to zero, a retrieval signal stream is obtained; finally, the first derivative of the characteristic convolution kernel and the retrieval signal stream is calculated to describe the curvature change trend and eliminate the influence of the great difference of the signal amplitude on the signal comparison, then whether there is a data segment with the same sign as the first derivative of the characteristic convolution kernel in the retrieval signal stream is searched point by point, so as to obtain whether the biological tissue contains the marker, if the biological tissue contains the marker, the marker signal segment in the original signal is marked, and the ultrasonic scanning information containing the marker position and depth and the like is obtained; in summary, the ultrasonic recognition method of the marker in the biological tissue of the embodiment can overcome the problem that the existing ultrasonic scanning is difficult to recognize due to the small ultrasonic reflection equivalent of the marker, and can accurately recognize the marker in the biological tissue.

[0050] Note: when the ultrasonic scanning signal characteristics of the marker in each direction are the same, such as when the marker is spherical, only the ultrasonic signal of the marker in one direction in the still water needs to be collected in step 11); when the ultrasonic scanning signal characteristics of the marker in each direction are different, the ultrasonic signal of the marker in at least two directions in the still water needs to be collected, and the meaning of "multiple" in this paper is "at least two".

[0051] The above-described embodiments are only preferred embodiments for fully illustrating the present application, and the protection scope of the present application is not limited thereto. The equivalent substitutions or transformations made by the person skilled in the art on the basis of the present application are within the protection scope of the present application. The protection scope of the present application is subject to the claims.

Claims

1. A method for ultrasound identification of markers within biological tissues, characterized in that: Includes the following steps: Step 1: Obtain the feature convolution kernel of the markers 11) Collect ultrasonic signals of the marker in still water to obtain the specimen signal; 12) Perform a convolution operation between the specimen signal and itself to obtain the feature convolution kernel of the marker; Step 2: Obtain the retrieval signal stream of biological tissues 21) Acquire ultrasound signals from biological tissues to obtain raw signals; 22) After performing a convolution operation between the original signal and the sample signal, assign zero values ​​to numbers less than a set threshold to obtain the retrieval signal stream; Step 3: Ultrasonic scanning information for identifying markers 31) Calculate the first derivatives of the feature convolution kernel and the retrieval signal stream, respectively; 32) Search point by point in the first derivative of the retrieved signal stream for a data segment whose sign matches the first derivative of the feature convolution kernel: if yes, the marker signal segment is obtained, and step four is executed; if no, it indicates that there is no marker in the biological tissue. Step 4: Mark the marker signal segment in the original signal to obtain the ultrasonic scanning information corresponding to the marker.

2. The ultrasonic identification method for markers in biological tissues according to claim 1, characterized in that: In step 11), ultrasonic signals of the marker in still water in multiple different directions are collected to obtain the specimen signals of the marker in multiple different directions.

3. The ultrasonic identification method for markers in biological tissues according to claim 2, characterized in that: In step 12), the ultrasonic signals of the marker in each direction in still water are convolved with itself to obtain the feature convolution kernels of the marker in different directions. All the feature convolution kernels are used to construct a feature convolution kernel set.

4. The ultrasound identification method for markers in biological tissues according to claim 3, characterized in that: In step 22), after performing convolution operations on the original signal and the specimen signals of the marker in different directions, the numbers less than a set threshold are assigned zero values ​​to obtain the retrieval signal streams corresponding to the different directions of the marker.

5. The ultrasound identification method for markers in biological tissues according to claim 4, characterized in that: In step 31), the first derivatives of all feature convolution kernels and all retrieval signal streams are calculated respectively.

6. The ultrasound identification method for markers in biological tissues according to claim 5, characterized in that: In step 32), the first derivative of the retrieval signal stream corresponding to the ultrasound scan in different directions of the marker is searched point by point to see if there is a data segment whose first derivative sign is consistent with the first derivative of any feature convolution kernel in the feature convolution kernel set: if yes, the marker signal segment is obtained and step four is executed; if no, it indicates that there is no marker in the biological tissue.

Images