Method for improving ultrasound elasticity imaging and device using same

Abstract

The present invention relates to a method for improving ultrasound elasticity imaging implemented by a processor. The method comprises: a step for receiving ultrasound elasticity RF data for a target part; a first processing step for polarizing impulse noise in ultrasound elasticity imaging on the basis of a statistical distribution of displacement values so as to acquire ultrasound elasticity RF data on which first processing has been performed; a second processing step for applying an adaptive median filter to the ultrasound elasticity RF data on which the first processing has been performed so as to acquire the ultrasound elasticity RF data on which second processing has been performed; and a step for implementing ultrasound elasticity imaging on the basis of the ultrasound elasticity RF imaging on which the first processing and the second processing have been performed.

Description

Method for improving ultrasound elasticity imaging and device using the same

[0001] The present invention relates to a method for improving ultrasonic elasticity imaging, a device using the same, and a system.

[0002] Ultrasound elastography, a technique that uses ultrasound to visualize the difference in stiffness between normal tissue and abnormal tissue such as tumor tissue, is frequently used to diagnose pathological conditions in human organs such as the breast, liver, and prostate.

[0003] There are many different methods for creating ultrasound elastography, and fundamental differences arise in the process of physically moving a target and calculating the amount of target displacement. For example, the acoustic radiation force impulse (ARFI) imaging technique uses ultrasound acoustic radiation force to induce target movement, detect the amount of movement, and visualize it. This ARFI technique generates local tissue movement by transmitting a pushing pulse consisting of hundreds of cycles from an ultrasound transducer, and then transmits a tracking pulse to measure the tissue displacement. The acquired tissue displacement can then be quantitatively calculated using a signal processing technique called a cross-correlation function.

[0004] However, displacement estimation values ​​can easily be inaccurate due to various factors, such as the complex internal structure of the target composed of heterogeneous materials and the resulting non-uniform propagation of ultrasound. In particular, large-amplitude impulse noise can occur, which can significantly degrade the resolution of elasticity images.

[0005] In other words, there is a continuous need for the development of improved ultrasound systems that can acquire higher quality ultrasound elasticity images for more accurate diagnosis.

[0006] The background technology of the invention has been prepared to facilitate a better understanding of the present invention. It should not be construed as an admission that the matters described in the background technology of the invention constitute prior art.

[0007] Meanwhile, to solve the aforementioned problem, the use of a fixed median filter with a fixed window size to remove impulse noise in ultrasound elasticity images has been proposed.

[0008] However, for fixed median filters, image quality improvement performance may vary depending on the window size. For example, if the window size is set small, impulse noise may remain. However, if the window size is set large, impulse noise may disappear, but the overall target shape may become blurred, resulting in a deterioration in ultrasound image quality.

[0009] To solve these problems, an adaptive median filter, rather than a fixed one, has been introduced. However, it is only effective when the noise in the ultrasound elasticity image is unipolar or bipolar, that is, when the noise pattern maintains a constant size. Therefore, it is difficult to apply the adaptive median filter when considering the characteristics of ultrasound elasticity images in which the size of impulse noise generally varies.

[0010] Meanwhile, the inventors of the present invention sought to devise a method for improving ultrasonic elasticity images, which can effectively remove impulse noises of various sizes appearing in ultrasonic elasticity images.

[0011] The inventors of the present invention attempted to achieve image enhancement using RF (Radio Frequency) data used to implement ultrasound elasticity images, unlike existing methods that have attempted image enhancement targeting ultrasound elasticity image files such as JPEG. That is, by obtaining a probability mass function for elasticity RF data and setting and applying a cut-off threshold level based on this, it was recognized that the magnitude of impulse noise can be polarized, and thus the effect is significantly increased when an adaptive median filter is applied. Accordingly, the inventors of the present invention developed a new ultrasound elasticity image enhancement system that can apply an adaptive median filter to ultrasound elasticity images, which could not be applied previously.

[0012] Accordingly, the inventors of the present invention expected that by providing a novel ultrasound elasticity image improvement system, it would be possible to obtain ultrasound elasticity images with improved quality by applying the system to ultrasound elasticity images having impulse noise of various sizes.

[0013] Furthermore, the inventors of the present invention expected that by providing a novel ultrasound elasticity image enhancement system, it would be possible to provide highly reliable analysis results for ultrasound elasticity images regardless of the skill level of the medical staff.

[0014] Accordingly, the problem to be solved by the present invention is to provide a new ultrasound elasticity image improvement method based on a probability mass function and a device using the same.

[0015] The tasks of the present invention are not limited to the tasks mentioned above, and other tasks not mentioned will be clearly understood by those skilled in the art from the description below.

[0016] In order to solve the above-described problem, an ultrasound elasticity image enhancement method according to an embodiment of the present invention is provided. The method is an ultrasound elasticity image enhancement method implemented by a processor, comprising: a step of receiving RF data for implementing an ultrasound elasticity image, the RF data including a displacement value for a target region; a first processing step of polarizing an amplitude of impulse noise mixed in the ultrasound elasticity RF data based on a statistical distribution of the displacement value to obtain ultrasound elasticity RF data subjected to a first processing; a second processing step of applying an adaptive median filter to the ultrasound elasticity RF data subjected to the first processing to obtain ultrasound elasticity RF data subjected to a second processing; and a step of implementing an ultrasound elasticity image based on the ultrasound elasticity RF data subjected to the first processing and the second processing, respectively, to obtain an improved ultrasound elasticity image.

[0017] According to a feature of the present invention, the first processing step may include a step of determining a displacement value of a target area based on ultrasonic elastic RF data, a step of determining a probability distribution based on the displacement value, a step of determining a cutoff threshold based on the probability distribution, and a step of polarizing the amplitude of impulse noise based on the cutoff threshold.

[0018] According to another feature of the present invention, the step of determining the displacement value may include a step of determining the displacement value using a correlation function, and the step of determining the probability distribution may include a step of normalizing the displacement value and a step of deriving a probability mass function based on the normalized displacement value. In this case, the step of determining the cutoff threshold value may include a step of determining the cutoff threshold value based on the probability mass function.

[0019] According to another feature of the present invention, the second processing step may include applying an adaptive median filter to the polarized impulse noise.

[0020] According to another feature of the present invention, the method may include a step of finally implementing an ultrasound elasticity image using elastic RF data on which the first and second processing have been performed.

[0021] According to another feature of the present invention, the method may further include the steps of implementing a B-mode (Brightness-mode) image including signal intensity for a target region, and the step of determining a statistical distribution of the signal intensity. Furthermore, the first processing step may include the step of determining a cutoff threshold for removing impulse noise based on the statistical distribution of the signal intensity and the statistical distribution of the displacement value.

[0022] According to another feature of the present invention, the method may further include the step of implementing a B-mode image including signal intensity for a target region, and the step of providing the B-mode image by overlaying it on an improved ultrasound elasticity image.

[0023] According to another feature of the present invention, the first processing step may include a step of determining a cutoff threshold using a pre-trained deep learning algorithm to determine a cutoff threshold for removing impulse noise by learning a statistical distribution, and a step of polarizing the amplitude of the impulse noise based on the cutoff threshold.

[0024] According to another feature of the present invention, the ultrasound elasticity image may be at least one selected from shear wave elastography, acoustic radiation force impulse image, strain elastography, transient elastography, and supersonic shear wave elastography.

[0025] In order to solve the above-described problem, a device for improving an ultrasound elasticity image according to another embodiment of the present invention is provided. The device includes a processor configured to process received ultrasound elasticity RF data including a displacement value for a target region, perform a first process of polarizing impulse noise of the ultrasound elasticity image based on a statistical distribution of the displacement value to obtain an ultrasound elasticity image on which a first process is performed, perform a second process of applying an adaptive median filter to the ultrasound elasticity RF data on which the first process is performed to obtain ultrasound elasticity RF data on which the second process is performed, and perform a third process of implementing an ultrasound elasticity image based on the ultrasound elasticity RF data on which the first process and the second process are performed, respectively, to obtain an improved ultrasound elasticity image.

[0026] According to another feature of the present invention, the processor may be further configured to determine a displacement value of a target region based on ultrasonic elastic RF data, determine a probability distribution based on the displacement value, determine a cutoff threshold based on the probability distribution, and polarize the amplitude of the impulse noise based on the cutoff threshold.

[0027] According to another feature of the present invention, the processor may be further configured to determine a displacement value using a correlation function, normalize the displacement value, determine a probability mass function based on the normalized displacement value, and determine a blocking threshold based on the probability mass function.

[0028] According to another feature of the present invention, the processor may be further configured to polarize the amplitude of impulse noise mixed into the elastic RF data on which the first processing has been performed, and then apply an adaptive median filter to remove the impulse noise.

[0029] According to another feature of the present invention, the processor may be further configured to implement a B-mode image including signal intensities for a target region, determine a statistical distribution of the signal intensities, and determine a cutoff threshold for removing impulse noise based on the statistical distribution of the signal intensities and the statistical distribution of the displacement values.

[0030] According to another aspect of the present invention, the processor is further configured to generate a B-mode image including signal intensities for a target region, and the processor may be further configured to provide the B-mode image by overlaying it on the enhanced ultrasound elastography image.

[0031] According to another feature of the present invention, the processor may be configured to determine a cutoff threshold for removing impulse noise by using an artificial intelligence algorithm, such as a pre-learned deep learning algorithm, to learn a statistical distribution, and to polarize the amplitude of the impulse noise based on the cutoff threshold.

[0032] According to another feature of the present invention, the ultrasound elasticity image may be at least one selected from shear wave elastography, acoustic radiation force impulse image, strain elastography, transient elastography, and supersonic shear wave elastography.

[0033] Specific details of other embodiments are included in the detailed description and drawings.

[0034] The present invention provides a novel ultrasound elasticity image enhancement system based on a probability mass function applicable to ultrasound elasticity images having impulse noise of various sizes, thereby providing high-quality ultrasound elasticity images.

[0035] In particular, the present invention can provide a remarkable effect for an adaptive median filter that can maximize and preserve information of an original image while reducing the shape distortion side effect of a fixed median filter applied to conventional ultrasound elasticity images.

[0036] That is, the present invention enables acquisition of high-quality ultrasound elasticity images regardless of the skill level of medical staff, and can contribute to establishing more accurate decision-making and treatment plans at the image analysis stage.

[0037] The effects according to the present invention are not limited to those exemplified above, and more diverse effects are included in this specification.

[0038] FIG. 1 illustrates an improvement system for ultrasonic elasticity images using an ultrasonic elasticity image improvement device according to one embodiment of the present invention.

[0039] FIG. 2 illustrates a procedure of a method for improving ultrasound elasticity images according to one embodiment of the present invention.

[0040] Figures 3 to 6 illustrate an example of a procedure for improving ultrasound elasticity images according to one embodiment of the present invention.

[0041] FIG. 7 illustrates a probability mass function graph derived according to a method for improving ultrasonic elasticity images according to one embodiment of the present invention.

[0042] FIGS. 8a to 8h, 9a to 9h, and 10a to 10f illustrate ultrasound elasticity images improved according to a method for improving ultrasound elasticity images according to an embodiment of the present invention.

[0043] The advantages of the invention and the methods for achieving them will become clearer with reference to the embodiments described below in detail with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms. These embodiments are provided solely to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the scope of the invention. The present invention is defined solely by the scope of the claims.

[0044] The shapes, sizes, ratios, angles, numbers, etc. disclosed in the drawings for explaining embodiments of the present invention are exemplary, and therefore the present invention is not limited to the matters illustrated. In addition, in describing the present invention, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. When the terms “includes,” “has,” and “consists of” are used in this specification, other parts may be added unless “only” is used. When a component is expressed in the singular, it includes a case where the plural is included unless there is a specifically explicit description.

[0045] When interpreting components, it is interpreted as including the error range even if there is no separate explicit description.

[0046] The individual features of the various embodiments of the present invention can be partially or wholly combined or combined with each other, and as can be fully understood by those skilled in the art, various technical connections and operations are possible, and each embodiment can be implemented independently of each other or can be implemented together in a related relationship.

[0047] For clarity in the interpretation of this specification, the terms used in this specification are defined below.

[0048] As used herein, the term "target region" can refer to any region for which ultrasound elasticity imaging information is desired. For example, target regions may include the thyroid gland, heart, carotid artery, abdomen, pelvis, and breasts, but are not limited thereto.

[0049] As used herein, the term "ultrasound elastography" refers to an image obtained by measuring the elasticity of an object using a transducer.

[0050] At this time, the ultrasound elasticity image may be a two-dimensional image, a three-dimensional image, a single still image, or a video composed of multiple cuts. For example, if the ultrasound elasticity image is a video composed of multiple cuts, an image enhancement procedure may be performed on each of the multiple images according to an enhancement method according to an embodiment of the present invention.

[0051] As a result, the present invention can be performed within an ultrasound diagnostic device to provide an elastic image with improved quality in real time.

[0052] The term “ultrasonic elastic RF data” used in the present invention means RF data having elasticity information of a target received through an ultrasonic transducer, and may be source data used to implement an ultrasonic elastic image.

[0053] The term "displacement value" as used herein may refer to the amount of movement of a target region following stimulation of the target region. In this case, the displacement value may correspond to the amount of movement of RF data.

[0054] The term "impulse noise" as used herein refers to error noise that occurs in the process of calculating displacement values ​​using a correlation function due to the structure of a target area composed of different materials (e.g., muscle, bone, fat) and non-uniform propagation of ultrasound, which may cause resolution degradation. Meanwhile, impulse noise generally has a constant amplitude in the form of unipolar or bipolar form that is clearly distinguishable from the target object, but can exist in a wider range of sizes within an elastic ultrasound image.

[0055] At this time, impulse noise can be removed from the image by a fixed median filter with a fixed window size, but performance may vary depending on the size of the window.

[0056] For example, if the window size is small, it is difficult to completely remove impulse noise, and if the window size is large, the impulse noise disappears, but the side effect of blurring the overall shape of the target may occur.

[0057] As used herein, the term "adaptive median filter" refers to a nonlinear digital filter whose window size can be adjusted according to image characteristics, and can be more effective in removing impulse noise than a fixed median filter.

[0058] In particular, adaptive median filters can be more effective in removing impulse noise with unipolar or bipolar amplitudes.

[0059] That is, when impulse noise with various amplitude sizes is included in the ultrasound elasticity image, the effect of removing the impulse noise by the adaptive median filter may be minimal.

[0060] Accordingly, according to the features of the present invention, a first processing in which impulse noise is converted into a unipolar or bipolar form for ultrasonic elastic RF data, and a second processing in which an adaptive median filter is applied can be performed.

[0061] Accordingly, the term "first processing" used in this specification may mean a process of polarizing the amplitude of impulse noise according to a setting of a cutoff threshold.

[0062] The term "second processing" as used herein can be interpreted as a process of filtering impulse noise using an adaptive median filter, i.e., removing actual impulse noise.

[0063] That is, according to the ultrasound elasticity image improvement method according to various embodiments of the present invention, an ultrasound elasticity image with impulse noise effectively removed by an adaptive median filter can be obtained.

[0064] As used herein, the term “third processing” may mean imaging of ultrasound elastic RF imaging data.

[0065] That is, it may be possible to obtain improved ultrasound elasticity images through the third processing.

[0066] Hereinafter, with reference to FIG. 1, an improvement system for ultrasonic elasticity images and an ultrasonic elasticity image improvement device using an ultrasonic elasticity image improvement device according to one embodiment of the present invention will be described.

[0067] FIG. 1 illustrates an improvement system for ultrasonic elasticity images using an ultrasonic elasticity image improvement device according to one embodiment of the present invention.

[0068] First, referring to FIG. 1, the improvement system (1000) includes a signal generator (110) that generates a tracking pulse, a pushing pulse, and a reference pulse. At this time, the generated signal passes through a signal beamformer (120) and is amplified by a transmission amplifier (130). Furthermore, the improvement system (1000) includes a transmission / reception switch (140) that switches between transmission and reception, and can be functionally connected to an ultrasonic transducer (150).

[0069] Meanwhile, the improvement system (1000) is switched to the receiving mode by the transmission / reception switch (140), and the ultrasonic elastic RF data received from the ultrasonic converter (150) passes through the receiving amplifier (210) and the receiving beamformer (220) to a plurality of signal processors (310, 320, and 330).

[0070] More specifically, the first signal processor (310) calculates an RF data-based displacement amount based on the received ultrasonic data, determines a probability mass function, and sets a cutoff threshold for noise removal. Furthermore, the first signal processor (310) applies an adaptive median filter.

[0071] According to another feature of the present invention, the second signal processor (320) calculates the reception strength based on RF data.

[0072] According to another feature of the present invention, the third signal processor (330) performs processing on RF data, such as setting a blocking threshold value using the received ultrasonic elastic RF data and a pre-learned artificial intelligence-based prediction model.

[0073] At this time, each signal processor (310, 320 and 330) can be functionally connected to each other.

[0074] According to another feature of the present invention, the improvement system (1000) implements a B-mode image or an improved ultrasound elasticity image based on signals received from signal processors (310 and 320) connected to each of the first image processor (410) and the second image processor (420).

[0075] According to another feature of the present invention, the improvement system (1000) can perform processing to synthesize a B-mode image and an improved elastic ultrasound image received from each of the first image processor (410) and the second image processor (420) through the image synthesizer (510).

[0076] In another improved system according to one embodiment of the present invention, the signal processors (310, 320, 330), the image processors (410, 420) and the image synthesizer (510) may be driven by a single processor, but are not limited thereto, and may be implemented and driven as a set of a plurality of microprocessors or an FPGA (Field-Programmable Gate Array).

[0077] At this time, the signal processors (310 and 320), the image processors (410 and 420), and further the image synthesizer (510) may include a memory for storing various data. For example, the memory (not shown) may store ultrasound B-mode images and elasticity images, or store RF data for ultrasound B-mode images, elasticity RF data for ultrasound elasticity images, probability mass functions, correlation functions, adaptive median filters, and artificial neural network-based prediction models trained to determine cutoff thresholds for elasticity RF data.

[0078] In various embodiments, the memory may include a volatile or non-volatile recording medium capable of storing various data, commands, and information. The enhancement system (1000) may then display the enhanced ultrasound elasticity image via the display (610).

[0079] According to a feature of the present invention, the display (610) can receive a user's request or input from a keyboard, touch screen, microphone, etc. through an I / O interface and provide a user interface on the display.

[0080] Meanwhile, the configuration of the information providing system according to various embodiments of the present invention is not limited to the above-described one, and can be implemented in more diverse forms as long as filtering for impulse noise based on a probability mass function is performed.

[0081] Hereinafter, with reference to FIGS. 2 to 6, a method for improving ultrasonic elasticity images according to one embodiment of the present invention will be specifically described.

[0082] Figure 2 illustrates a procedure for an ultrasound elasticity image enhancement method according to one embodiment of the present invention. Figures 3 to 6 exemplarily illustrate a procedure for an ultrasound elasticity image enhancement method according to one embodiment of the present invention.

[0083] First, referring to FIG. 2, the information provision procedure according to one embodiment of the present invention is as follows. First, RF data of an ultrasound elasticity image for a target region is received (S310). Next, a first processing step is performed to polarize impulse noise based on the statistical distribution of displacement values ​​(S320). Next, a second processing step is performed, in which an adaptive median filter is applied (S330). Finally, a third processing step is performed to generate an improved ultrasound elasticity image (S340).

[0084] More specifically, in the step (S310) where ultrasonic elastic RF data is received, RF data used to create an ultrasonic elastic image of the target area may be received. That is, elastic RF data from a previous step, rather than a final image file such as a JPEG, may be received, but is not limited thereto. For example, the proposed technology can also be applied to ultrasonic elastic images in JPEG format.

[0085] Next, the first processing step (S320) is performed.

[0086] In various embodiments of the present invention, the first processing step (S320) may be performed based on a probability mass function for RF data.

[0087] More specifically, referring to FIG. 3, elastic RF data having elasticity information about a target is received (S410), and the displacement value, i.e., the movement amount of the target, can be determined by a correlation function (S420). At this time, a plurality of different displacement values ​​can be determined, and after normalization (S430), a probability mass function is derived (S440), and a cutoff threshold for removing impulse noise from the probability mass function is determined (S450). Then, the first processing step can be terminated by applying the cutoff threshold to the RF data so that the amplitude of the impulse noise becomes polarized (S460).

[0088] Typically, a cutoff threshold can be determined by considering sparse signal data as impulse noise. However, if the cutoff threshold is set too low, important signal information may be lost, and if it is set too high, persistent noise may occur. That is, impulse noise can be removed more effectively by determining the cutoff threshold based on the distribution derived in the threshold determination step (S450).

[0089] In various embodiments of the present invention, the first processing step (S320) may be performed with a cutoff threshold determined based on a B-mode image that visualizes the interior by utilizing a difference in acoustic impedance values ​​within the target.

[0090] More specifically, referring to FIG. 4, RF data of a B-mode image is received (S510), and the reception intensity of a target is determined based on an acoustic impedance value (S520). Then, the intensity of the signal of the target is normalized, and at this time, the normalization can be performed based on the maximum value of the intensity (S530). Then, a probability mass function of the normalized signal intensity is determined (S540). At this time, a cutoff threshold can be determined (S550) through a comparison of the probability mass function of the elastic image acquired in the step (S5402) in which the probability mass function is determined based on the elastic RF data and the probability mass function based on the B-mode image. That is, the cutoff threshold is determined based on the B-mode image having relatively little impulse noise, and the amplitude of the impulse noise for the elastic RF data can be polarized while minimizing the loss of image information (S560).

[0091] In various embodiments of the present invention, the first processing step (S320) may be performed by a pre-trained artificial neural network-based model to learn statistical distributions and determine a cutoff threshold for removing the impulse noise.

[0092] More specifically, referring to FIG. 5, an artificial neural network model applicable to impulse noise removal is obtained by performing a step (S6502) of pre-training an artificial neural network model learned to determine a cutoff threshold based on a probability mass function. The pre-trained artificial neural network model determines a new cutoff threshold based on the probability mass function obtained in the step (S640) of determining the probability mass function (S650). Then, in a step (S660) of polarizing impulse noise, primary processing is performed by the determined cutoff threshold, and as a result, improved elastic RF data can be obtained.

[0093] Next, a second processing step is performed in which adaptive median filtering is performed on the improved elastic RF data.

[0094] At this time, an adaptive median filter is applied in the second processing stage, but is not limited thereto, and a wider variety of nonlinear digital filters may be used.

[0095] More specifically, referring to FIG. 6, the improved ultrasonic RF data obtained as a result of the step (S710) in which the first processing based on the cutoff threshold is performed is subjected to adaptive median filtering of levels A and B (S720 to S770).

[0096] Adaptive median filtering within level A first takes the median (I) of the elements within the kernel. med ) when calculating I min < I med < I max Check if the condition is met (S720). If this condition is not met, the kernel size (W xy ) and check the condition again (S730). The kernel size is the predefined maximum kernel size (W max ) is greater than I med Outputs (S740, S750). At this time, I min < I med < I max If is satisfied, I med This may mean that the value of the adjacent element is within a reasonable range. If this condition is met, the process of Level B can be performed.

[0097] Within level B, the filter considers the original value of the central element inside the kernel (I ) as the value being replaced. xy ) can be examined in detail. That is, from level B to I min < I xy < I max A check is performed to see if the condition is satisfied (S760). If this condition is satisfied, I xy This may mean that the value of other nearby elements is within a reasonable range, in which case the filter is I xyIt can be concluded that the filter is unlikely to be damaged by noise. Consequently, the I corresponding to the improved ultrasonic RF data xy Outputs (S770). Meanwhile, I min < I med < I max or I min < I xy < I max If one of the conditions is not met I med or I xy This may mean that the neighboring element is outside the reasonable element value range. In this case, level B is I min < I med < I max After the condition is met I xy Since it is meaningful to evaluate the validity of the filter, the median (I med ) and the original value (I xy ) can be used to balance noise removal and image detail preservation.

[0098] For example, the process may be implemented with an improved ultrasound elasticity image by selecting either the Ixy value in step S770 or the Imed value in step S750.

[0099] Hereinafter, the performance of an ultrasound elasticity image enhancement system according to various embodiments of the present invention will be described with reference to FIGS. 7, 8a to 8h, 9a to 9h, and 10a to 10f.

[0100] Figure 7 illustrates a probability mass function graph derived according to an ultrasound elasticity image enhancement method according to an embodiment of the present invention. Figures 8a to 8h, 9a to 9h, and 10a to 10f illustrate ultrasound elasticity images enhanced according to an ultrasound elasticity image enhancement method according to an embodiment of the present invention.

[0101] First, referring to FIG. 7, a probability mass function graph derived according to a method for enhancing ultrasound elasticity images according to one embodiment of the present invention is exemplarily illustrated. Based on this graph, a cutoff threshold for more effectively removing impulse noise is set, processing is performed to polarize the impulse noise, and the graph is then used as an input signal for the next step, an adaptive median filter.

[0102] Next, Figures 8a to 8h show experimental results comparing the conventional filter and the proposed probability distribution function-based elasticity image enhancement method applied to a tissue-simulating phantom.

[0103] Figure 8a is the original elastic image with impulse noise, and Figure 8b is the result of applying a fixed median filter with a window size of 5 x 5. In the case of a fixed filter, the impulse noise disappears to some extent, but it may be difficult to completely remove it. On the other hand, Figure 8e is the result of applying a fixed median filter with a window size of 11 x 11. In this case, the impulse noise is completely removed, but the side effect of distorting the shape of the original image appears.

[0104] Referring further to FIG. 8f, the results of applying the elastic ultrasound image enhancement method according to various embodiments of the present invention are illustrated. That is, when the elastic ultrasound image method proposed in the present invention is applied, impulse noise can be removed while maintaining the shape of the target.

[0105] At this time, referring to the images that emphasize the boundary of the elastic images of FIGS. 8c, 8d, 8g and 8h, it appears that when the elastic ultrasound image improvement method according to various embodiments of the present invention was used (FIG. 8h), the shape of the target was maintained while the impulse noise was removed more effectively.

[0106] Figures 9a to 9h are the results of evaluating a small eye using the same method as Figures 8a to 8h described above.

[0107] Figure 9a is the original elastic image with impulse noise, and Figure 9b is the result of applying a fixed median filter with a window size of 5 x 5. In the case of a fixed filter, the impulse noise disappears to some extent, but it may be difficult to completely remove it. On the other hand, Figure 9e is the result of applying a fixed median filter with a window size of 11 x 11. In this case, the impulse noise is completely removed, but the side effect of distorting the shape of the original image appears.

[0108] Referring further to FIG. 9f, the results of applying the elastic ultrasound image enhancement method according to various embodiments of the present invention are illustrated. That is, when the elastic ultrasound image method proposed in the present invention is applied, impulse noise can be removed while maintaining the shape of the target.

[0109] At this time, referring to the images that emphasize the boundary of the elastic images of FIGS. 9c, 9d, 9g and 9h, it appears that when the elastic ultrasound image improvement method according to various embodiments of the present invention was used (FIG. 9h), the shape of the target was maintained while the impulse noise was removed more effectively.

[0110] Figures 10a, 10b, and 10c show experimental results of applying a pure adaptive median filter without the first processing, i.e., polarization of the probability distribution-based impulse noise, to a small eye. Furthermore, Figures 10d, 10e, and 10f show experimental results of applying an adaptive median filter after polarization of the probability distribution-based impulse noise, according to various embodiments of the present invention. Comparing the two results, it appears that when only the adaptive median filter is applied, a large amount of impulse noise (red graph) remains.

[0111] These results may imply that when applying the ultrasound elasticity image enhancement method according to various embodiments of the present invention, impulse noise is more effectively removed while maintaining the shape of the target than when using a single adaptive median filter.

[0112] That is, the present invention can provide a remarkable effect for an adaptive median filter that can maximize and preserve information of an original image while reducing the shape distortion side effect of a fixed median filter applied to conventional ultrasound elasticity images.

[0113] Furthermore, the present invention enables acquisition of high-quality ultrasound elasticity images regardless of the level of skill of medical staff, and can contribute to establishing more accurate decision-making and treatment plans at the image analysis stage.

[0114] Although the embodiments of the present invention have been described in more detail with reference to the attached drawings, the present invention is not necessarily limited to these embodiments, and various modifications may be implemented without departing from the technical spirit of the present invention. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention, but to explain it, and the scope of the technical spirit of the present invention is not limited by these embodiments. Therefore, it should be understood that the embodiments described above are exemplary in all aspects and not restrictive. The protection scope of the present invention should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of the rights of the present invention.

[0115] [Explanation of symbols]

[0116] 1000: System

[0117] 110: Signal generator 120: Transmission beamformer

[0118] 130: Transmitter amplifier 140: Transmit / receive switch

[0119] 150: Ultrasonic transducer

[0120] 210: Receiver amplifier 220: Receiver beamformer

[0121] 310: First signal processor 320: Second signal processor

[0122] 330: Third signal processor

[0123] 410: First image processor 420: Second image processor

[0124] 510: Video synthesizer

[0125] 610: Display

[0126] [National Research and Development Project Supporting This Invention]

[0127] [Project ID] 1711192864

[0128] [Assignment Number] 00141273

[0129] [Ministry Name] Ministry of Science and ICT

[0130] [Name of Project Management (Specialist) Institution] (Foundation) Inter-Ministry Full-cycle Medical Device Research and Development Project Group

[0131] [Research Project Name] Inter-Ministry Full-cycle Medical Device Research and Development (Ministry of Science and Technology)

[0132] [Research Project Title] High-Performance Air-Coupled Ultrasound for Noninvasive Hearing Function Improvement

[0133] Probe module development

[0134] [Contribution rate] 1 / 2

[0135] [Name of the project performing organization] Dongguk University Industry-Academic Cooperation Foundation

[0136] [Research Period] January 1, 2023 - December 31, 2023

[0137] [Project ID] 1711180351

[0138] [Assignment Number] 2021R1A2C1004329

[0139] [Ministry Name] Ministry of Science and ICT

[0140] [Name of Project Management (Specialist) Institution] National Research Foundation of Korea

[0141] [Research Project Name] Individual Basic Research (Ministry of Science and ICT)

[0142] [Research Project Title] Development of Core Elastic Ultrasound-Based Technology for Precise Cataract Diagnosis

[0143] [Contribution rate] 1 / 2

[0144] [Name of Project Performing Organization] Dongguk University

[0145] [Research Period] March 1, 2023 - February 29, 2024

Claims

1. A method for improving ultrasound elasticity images implemented by a processor, A step of receiving ultrasonic elastic RF data including displacement values ​​for a target area; A first processing step of polarizing impulse noise of an ultrasonic elasticity image based on a statistical distribution of the displacement values ​​to obtain ultrasonic elasticity RF data on which first processing has been performed; A second processing step of applying an adaptive median filter to the ultrasonic elasticity RF data on which the first processing has been performed, to obtain ultrasonic elasticity RF data on which the second processing has been performed, and A method for improving ultrasound elasticity, comprising a third processing step of implementing an ultrasound elasticity image based on ultrasound elasticity RF data on which the first processing and the second processing have been performed, respectively, to obtain an improved ultrasound elasticity image.

2. In paragraph 1, The above first processing step is, A step of determining a displacement value according to the amount of movement of the target portion based on the ultrasonic elastic RF data; A step of determining a probability distribution based on the above displacement value; A step of determining a cut-off threshold level based on the above probability distribution, and A method for enhancing ultrasound elasticity images, comprising the step of polarizing the impulse noise based on the above-described cutoff threshold.

3. In paragraph 2, The step of determining the above displacement value is: A step of determining a displacement value using a correlation function is included, The step of determining the above probability distribution is: a step of normalizing the above displacement value, and A step of determining a probability mass function based on the above normalized displacement value is included, The step of determining the above blocking threshold is: A method for improving ultrasound elasticity images, comprising the step of determining a cutoff threshold based on the probability mass function.

4. In paragraph 1, The second processing step is: A method for improving ultrasound elasticity images, comprising the step of applying the adaptive median filter to the polarized impulse noise.

5. In paragraph 1, The above method, A step of implementing a B-mode (Brightness-mode) image including signal intensity for the target area, and further comprising a step of determining a statistical distribution of the signal strength; The above first processing step is, A method for enhancing ultrasound elasticity images, comprising the step of determining a cutoff threshold for removing the impulse noise based on the statistical distribution of the signal intensity and the statistical distribution of the displacement value.

6. In paragraph 1, A step of implementing a B-mode image including signal intensity for the target area, and A method for enhancing ultrasound elasticity images, further comprising the step of providing the B-mode image by overlaying it on the enhanced ultrasound elasticity image.

7. In paragraph 1, The above first processing step is, A step of determining the cutoff threshold by using a pre-trained deep learning algorithm to learn the statistical distribution and determine the cutoff threshold for removing the impulse noise, and A method for improving ultrasound elasticity images, comprising the step of polarizing the amplitude of the impulse noise based on the above-mentioned cutoff threshold.

8. In paragraph 1, The above ultrasound elasticity image is, A method for enhancing ultrasound elastography, wherein the method comprises at least one of shear wave elastography, acoustic radiation force impulse imaging, strain elastography, transient elastography, and supersonic shear wave elastography.

9. Processing the received ultrasonic elastic RF data, including the displacement value for the target area, To obtain an ultrasound elasticity image on which the first processing has been performed, the first processing is performed to polarize the impulse noise of the ultrasound elasticity image based on the statistical distribution of the displacement values, To obtain ultrasonic elasticity RF data on which second processing has been performed, a second process is performed by applying an adaptive median filter to the ultrasonic elasticity RF data on which the first processing has been performed, A device for improving ultrasound elasticity images, comprising a processor configured to perform third processing to implement an ultrasound elasticity image based on ultrasound elasticity RF data on which the first processing and the second processing have been performed, respectively, to obtain an improved ultrasound elasticity image.

10. In paragraph 9, The above processor, Based on the above ultrasonic elastic RF data, the displacement value according to the amount of movement of the target area is determined, Determine the probability distribution based on the above displacement values, Determine the blocking threshold based on the above probability distribution, A device for enhancing ultrasound elasticity imaging, further configured to polarize the impulse noise based on the above-mentioned cutoff threshold.

11. In paragraph 10, The above processor, Determine the displacement value using the correlation function, Normalize the above displacement values, Determine the probability mass function based on the above normalized displacement value, A device for improving ultrasound elasticity imaging, further configured to determine a cutoff threshold based on the probability mass function.

12. In paragraph 9, The above processor, A device for enhancing ultrasound elasticity images, further configured to apply the adaptive median filter to the amplitude-polarized noise after the first processing is performed.

13. In paragraph 9, The above processor, Receive RF data for B-mode imaging including signal intensity for the target area, Determine the statistical distribution of the above signal strength, A device for improving ultrasound elasticity imaging, further configured to polarize the amplitude of the impulse noise based on the statistical distribution of the signal intensity and the statistical distribution of the displacement value.

14. In paragraph 9, The above processor is configured to implement an image and synthesize an image, The above processor implements a B-mode image including signal intensity for the target area, A device for enhancing ultrasound elasticity images, further configured to provide the B-mode image by overlaying it on the enhanced elasticity ultrasound image.

15. In paragraph 9, The above processor, The cutoff threshold is determined by using a pre-trained deep learning algorithm to learn the statistical distribution and determine the cutoff threshold for removing the impulse noise. A device for enhancing ultrasound elasticity imaging, further configured to polarize the impulse noise based on the above-mentioned blocking threshold.

16. In paragraph 9, The above ultrasound elasticity image is, A device for improving ultrasound elasticity imaging, wherein the device comprises at least one selected from transverse wave elasticity imaging, acoustic radiation force impulse imaging, strain elasticity imaging, transient elasticity imaging, and supersonic transverse wave elasticity imaging.

Images