Method and apparatus for monitoring output and performance of mechanical wave generating device on basis of ultrasonic image by using ai

Abstract

The present invention relates to a method and apparatus for automatically inspecting a change in output and performance of a mechanical wave generating device for irradiating mechanical waves, such as shock waves and ultrasonic waves, on the basis of an ultrasonic image by using artificial intelligence technology. A method for monitoring the output and / or the performance of the mechanical wave generating device on the basis of the ultrasonic image by using AI uses an ultrasonic imaging device configured to obtain an ultrasonic image for imaging acoustic characteristics of a medium through which the mechanical waves propagates, and AI for recognizing and learning the ultrasonic image. The output and / or the performance of the mechanical wave generating device can be monitored through the ultrasonic image of the medium to which the mechanical waves are irradiated.

Description

Method and device for monitoring the output and performance of an ultrasonic image-based dynamic wave generator using AI

[0001] The present disclosure relates to a method for automatically auditing changes in the output and performance of a dynamic wave generating device that investigates dynamic waves such as shock waves and ultrasound using ultrasonic imaging and artificial intelligence technology.

[0002] Although a shock wave generating device that generates shock waves is selected as an example of the technology and implementation examples that form the background of the present invention, the content of the invention is directed to a mechanical wave and a mechanical wave generating device including a shock wave or ultrasound.

[0003] Clinically, procedures that use shock waves for therapeutic purposes can be divided into ESWL (Extracorporeal shock wave lithotripsy) and ESWT (Extracorporeal shock wave therapy).

[0004] ESWL utilizes the destructive effects of powerful shock waves to break up stones within the body. ESWT, on the other hand, utilizes shock waves with relatively lower acoustic power than those used in ESWL. Its indications include reducing pain in various degenerative musculoskeletal disorders (e.g., plantar fasciitis, tennis elbow, frozen shoulder), and promoting the healing of damaged tissue. ESWT's clinical efficacy in treating myocardial infarction, erectile dysfunction, and dementia has recently been proven, demonstrating its continued growth in the medical application of shock waves.

[0005] Shockwave therapy devices include a shockwave generator that converts various energies (e.g., electrical, magnetic, chemical, thermal, etc.) into shockwaves, i.e., mechanical energy, and a drive unit that supplies energy to the shockwave converter. The shock generation methods used in ESWL are classified into electrohydraulic (EH), electromagnetic (EM), and piezoelectric (PE) methods depending on the energy type of the drive unit. In addition to the three methods used in ESWL, ESWT additionally uses a ballistic shockwave generation principle that is inexpensive to manufacture and easy to operate. Ballistic shockwave generation methods are classified into pneumatic and electromagnetic methods.

[0006] Shock waves emitted from shock wave converters are often controlled by a focuser or waveguide located around the shock wave converter. Depending on the focuser or waveguide, the shock wave's wavefront varies, and the sound field formed in front of the shock wave generator can be categorized as focused, radial, or planar.

[0007] The acoustic output of shock waves, which are expected to have therapeutic effects, is higher than that of ultrasound used for diagnostic purposes. Shock waves with strong negative pressure can induce cavities in the propagating medium. In water, these cavities are observed as bubbles. The formation of these cavitation bubble clusters can be qualitatively visualized optically using micropulsed LED light (MPLL).

[0008] The effectiveness of shockwave therapy largely depends on the acoustic output of the shockwave delivered. To ensure that the target tissue receives an appropriate amount of shockwaves to achieve the desired therapeutic effect, the practitioner must be able to set and control the acoustic output of the shockwaves delivered to the patient during the treatment.

[0009] Shock wave generators age and deteriorate over time after manufacture, resulting in deterioration in acoustic output. The shock wave converter, which converts the energy of the actuator into shock waves, is a core component of the shock wave generator. It has a relatively short lifespan and is prone to wear and tear. As the number of shock waves generated increases, the acoustic output of the generated shock waves decreases. Shock wave converters should be replaced regularly or after a certain number of shock waves generated to maintain consistent acoustic output.

[0010] Occasionally, shock wave transducers may begin to degrade before their scheduled replacement period, and even after their replacement period has passed, they may remain in use. Problems with the generator other than the shock wave transducer, such as changes in the energy supplied by the drive unit, can also cause a reduction in the acoustic output of the shock wave generator.

[0011] To maintain the acoustic output of shockwave therapy equipment and ensure therapeutic efficacy, technology is needed to monitor the acoustic output and performance of shockwave generators at the user level. The acoustic output of shockwave generators can be measured using hydrophones in accordance with standards such as IEC61486. ​​These measurement and testing methods require expensive measuring equipment, skilled technicians, and significant time. Shockwave generators currently in clinical use lack a built-in monitoring system capable of warning of decreased acoustic output or performance abnormalities.

[0012] The matters described in the technical background of this invention are written to enhance understanding of the background of the invention and may include matters that are not already known in the field to which this technology belongs.

[0013] The problem to be solved by the present invention is to provide a method for easily monitoring, at the user level, the decrease in the acoustic output of shock waves to be irradiated to a patient due to an increase in the number of shock wave generation times in a shock wave generating device or the expected aging of the equipment.

[0014] The technical problems to be solved by the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned can be understood by a person having ordinary skill in the technical field to which the present invention belongs from the description below.

[0015] According to an embodiment of the present invention, a method for monitoring the output and performance of a dynamic wave generator based on an ultrasonic image using AI (Artificial Intelligence) is provided for monitoring the output and performance of a dynamic wave generator that generates a dynamic wave, and the output or performance of the dynamic wave generator can be monitored by utilizing an ultrasonic imaging device configured to acquire an ultrasonic image of a medium through which the dynamic wave propagates and AI that recognizes and learns the ultrasonic image.

[0016] The above-mentioned dynamic waves may collectively refer to ultrasound and shock waves that can be expected to have an effect suitable for medical purposes by inducing acoustic characteristics and biological changes in the investigated target tissue.

[0017] The medium through which the above-mentioned mechanical wave propagates may include a reference medium (e.g., distilled water, oil, gel, or ultrasonic tissue mimicking material) whose acoustic properties are well known or suitable for confirming the propagation properties of the above-mentioned mechanical wave.

[0018] The above ultrasonic imaging device can be triggered at the time when the dynamic wave generating device operates to record a series of ultrasonic images of the medium through which the dynamic wave propagates and transmit the recorded images to the AI.

[0019] The above ultrasound image may include a passive cavitation image (PCI) capable of imaging the dynamic characteristics of a group of bubbles generated in a medium by the above mechanical wave.

[0020] The above ultrasonic imaging device can selectively image a cross-section of a sound field formed in a medium by the above dynamic wave, so that relative position information of the ultrasonic probe included in the dynamic wave generating device and the above ultrasonic imaging device is shared, and the position of the ultrasonic probe can be flexibly controlled.

[0021] The above AI can perform preprocessing to remove background images or other noise from the ultrasound image received from the ultrasound imaging device before the dynamic wave is irradiated.

[0022] The above AI can perform pre-learning to recognize and remember a series of ultrasonic images of the medium according to the output settings of the above dynamic wave generator when the above dynamic wave generator is operating normally.

[0023] The above-mentioned pre-learned AI can receive a series of ultrasonic images collected from the medium when the above-mentioned dynamic wave is irradiated, estimate the output set in the above-mentioned dynamic wave generating device, or determine the degree of inclusion of abnormal images, and provide information on the acoustic output or performance of the above-mentioned dynamic wave generating device, and changes therein.

[0024] According to another embodiment of the present invention, an output and performance monitoring device for a mechanical wave generator using an ultrasonic image-based AI for monitoring the output and performance of a mechanical wave generator that generates a mechanical wave includes an ultrasonic imaging device configured to acquire an ultrasonic image of a medium through which the mechanical wave propagates, and an AI that recognizes and learns the ultrasonic image. The output or performance of the mechanical wave generator can be monitored through the ultrasonic image of the medium through which the mechanical wave is irradiated.

[0025] According to the present invention, during a treatment process using a dynamic wave generating device including shock waves and ultrasound, the practitioner can monitor the output and performance of the dynamic wave generating device by utilizing AI, thereby maintaining the acoustic output or exposure amount of the dynamic wave to be irradiated to the patient within a certain range, thereby making it possible to expect the planned therapeutic effect from the irradiation of the dynamic wave.

[0026] In addition, various effects that can be obtained or expected due to embodiments of the present invention are disclosed directly or implicitly in the detailed description of the embodiments of the present invention.

[0027] The accompanying drawings, which are intended to aid in understanding the present invention, provide embodiments of the present invention along with a detailed description. However, the technical features of the present invention are not limited to any specific drawings, and the features disclosed in each drawing may be combined to form new embodiments. The embodiments of the present specification may be better understood by referring to the following description in conjunction with the accompanying drawings, in which similar reference numerals designate identical or functionally similar elements.

[0028] Figure 1 shows an optical image of a group of bubbles generated at the moment of irradiating a shock wave while the output settings of an electromagnetic shock wave generator are set to three levels, namely minimum, medium, and maximum, and the image was captured by operating the light for 500 μs.

[0029] Figure 2 shows an example of a device for acquiring an ultrasonic image of a medium through which a shock wave generated from a shock wave generating device propagates.

[0030] Figure 3 shows an example of a typical ultrasound image of a shock wave propagation medium immediately after a shock wave has been irradiated by the shock wave generating device of Figure 2.

[0031] Figure 4 shows an example of an ultrasound image (with background image removed) taken 2.4 ms immediately after shock wave generation at a medium power setting of the shock wave generator of Figure 2.

[0032] Figure 5 compares five frames of ultrasound images recorded from the time of shock wave generation when the output setting of the shock wave generating device of Figure 2 is increased.

[0033] FIG. 6 is a schematic block diagram for explaining a method for monitoring the output and performance of an ultrasonic image-based dynamic wave generation device using AI according to an embodiment of the present invention.

[0034] Figure 7 is a flowchart of a method for monitoring the output and performance of an ultrasonic image-based dynamic wave generation device using AI according to an embodiment of the present invention.

[0035] It should be understood that the drawings referenced above are not necessarily drawn to scale and are intended to provide brief representations of various features that illustrate the fundamental principles of the present invention. For example, specific design features of the present invention, including specific dimensions, orientations, positions, and shapes, will be determined in part by the specific intended application and usage environment.

[0036] Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings so that those skilled in the art can easily implement the present invention. However, the present invention may be implemented in various different forms and is not limited to the described embodiments.

[0037] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the present invention. As used herein, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It should also be understood that the terms "comprises" and / or "comprising," as used herein, indicate the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, components, and / or groups thereof. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items. The term "coupled" indicates a physical relationship between two components in which the components are directly connected to one another or are indirectly connected through one or more intervening components.

[0038] In describing the components of the present invention, when it is described that a component is “connected,” “coupled,” or “connected” to another component, it should be understood that the component may be directly connected, coupled, or connected to the other component, but another component may also be “connected,” “coupled,” or “connected” between each component.

[0039] Figure 1 shows optical images of a group of bubbles generated when a shock wave is irradiated underwater, captured by operating the light for 500 μs, with the output settings of an electromagnetic shock wave generator set to three levels: minimum, medium, and maximum. Figures 1 (a), (b), and (c) respectively show optical images of a group of cavitation bubbles generated when the shock wave generated underwater is irradiated with the output settings of the shock wave generator set to minimum, medium, and maximum. As can be seen in Figure 1, the bubble group is concentrated in the focal area where the shock wave is focused, and as the output setting of the shock wave generator increases, the amount of the generated bubble group increases and the generated area widens. These results imply that changes in the acoustic output or performance of a shock wave generator can be identified based on the correlation between the acoustic output of the shock wave generator and the image of the cavitation bubble group.

[0040] When shock wave propagation induces cavitation in a medium, the presence of bubble clusters significantly alters the local acoustic properties of the medium, such as ultrasonic propagation speed, attenuation coefficient, density, and acoustic impedance. Because the acoustic impedance of the gas contained within the bubbles differs significantly from that of the surrounding medium, the bubble clusters strongly reflect ultrasonic signals. Furthermore, the bubbles generated by shock waves are very small compared to the wavelength of the shock wave, measuring only a few micrometers to a few millimeters, and thus act as scatterers that scatter ultrasonic waves.

[0041] Ultrasound imaging primarily captures the effects of ultrasound reflection and scattering by scatterers, which are influenced by differences in acoustic impedance. The acoustic properties of a medium altered by shock wave irradiation, such as ultrasound reflection and scattering, are visualized in ultrasound images as strong echogenicity. This suggests that ultrasound images of shock-irradiated media are sensitive to the shock wave irradiation and are closely correlated with the acoustic output of the shock wave.

[0042] Ultrasound imaging may include passive cavitation imaging (PCI), which images the dynamic characteristics of a group of bubbles created in a medium by mechanical waves.

[0043] The present invention discloses a technology for recording changes in the acoustic properties of a medium through which a mechanical wave generated by a mechanical wave generating device including a mechanical wave converter, a driving unit, a focusing unit, and a mechanical wave guide propagates as an ultrasonic image, and monitoring changes in the acoustic output of the mechanical wave generating device and the output and performance of the mechanical wave generating device through an AI algorithm that recognizes the ultrasonic image. Throughout this specification, the mechanical wave may be understood to collectively refer to ultrasound and shock waves that induce acoustic properties and biological changes in the investigated target tissue and are expected to have effects suitable for medical purposes, and the following will exemplarily describe a case where the mechanical wave is a shock wave.

[0044] Fig. 2 shows an example of a device for acquiring an ultrasonic image of a medium through which a shock wave generated from an electromagnetic shock wave generator propagates. Referring to Fig. 2, a shock wave generator (10) includes a shock wave converter (11), a driving unit (13), and a shock wave guide (15). The shock wave converter (11) generates a shock wave by power supplied from the driving unit (13). The shock wave guide (15) may have the form of a focuser capable of focusing the shock wave generated from the shock wave converter (11). Thus, an electromagnetic focused shock wave generator (10) is implemented. The shock wave converter (11) is installed in a tank (17) filled with a shock wave transmission medium, for example, water. Thus, a shock wave focusing area can be formed within the tank (17).

[0045] An ultrasonic imaging device (20) obtains an ultrasonic image for a specific cross-section of a medium through which a generated shock wave propagates. The ultrasonic imaging device (20) includes an ultrasonic imaging probe (21) for obtaining an ultrasonic image. The ultrasonic imaging probe (21) for obtaining an ultrasonic image is configured to be freely movable through a positioning arm (23). The ultrasonic imaging probe (21) for obtaining an ultrasonic image is structurally flexibly connected to the structure of the shock wave converter (11) so as to freely move and share relative position information with each other to obtain an ultrasonic image for a specific cross-section of a medium through which a shock wave propagates. The ultrasonic imaging probe (21) obtains an ultrasonic image from immediately before the generation of a shock wave in conjunction with an operating signal of a shock wave generating device, and the obtained image is recorded in the ultrasonic imaging device (20).

[0046] Figure 3 shows an example of a typical ultrasonic image of a shock wave propagation medium immediately after a shock wave is irradiated by the shock wave generator of Figure 2. Figure 3 (a) is a background image before the shock wave is irradiated, and Figure 3 (b) is an ultrasonic image recorded after the shock wave is irradiated. Figure 3 (c) is an image after the background image of (a) is removed from the image of (b). In the background image of Figure 3 (a), the horizontal line is the central axis of the shock wave sound field that is symmetrical vertically, and the point orthogonal to the vertical axis indicates the focus (convergence) position of the shock wave. The recorded image undergoes a preprocessing step that removes the initial background image formed by the anatomical structure of the object to be irradiated with the shock wave before the shock wave is generated, and then an ultrasonic image that selectively visualizes the acoustic changes in the propagation medium induced by the shock wave is extracted.

[0047] Figure 4 shows an example of an ultrasound image (with background image removed) captured 2.4 ms immediately after shock wave generation at a medium power setting of the shock wave generator of Figure 2. Figure 4 shows 12 ultrasound images with background image removed. The ultrasound images are recorded from the time when a strong electromagnetic signal due to the operation signal of the shock wave generator appears in the ultrasound image until after the cavitation bubble cluster disappears and no further significant changes in the ultrasound image are detected (e.g., ~ 5.4 ms).

[0048] Figure 5 compares ultrasonic images recorded from the time of shock wave generation when the output setting of the shock wave generator of Figure 2 is increased. Figures 5 (a), (b), and (c) show five frames of ultrasonic images acquired from the time of shock wave generation when the output setting of the shock wave generator is increased to minimum, medium, and maximum. As can be expected, as the output increases, the intensity and range of echogenicity in the ultrasonic image in the shock wave focus area increases, similar to what was observed in the photographic images of the cavitation bubble group generated by the shock wave exemplified in Figure 1.

[0049] FIG. 6 is a schematic block diagram illustrating a method for monitoring the output and performance of an ultrasonic image-based dynamic wave generator using AI according to an embodiment of the present invention. Referring to FIG. 6, when the shock wave generator (10) operates normally, the ultrasonic imaging device (20) records an ultrasonic image into a DB (25), and the recorded ultrasonic image DB (25) is transmitted to an artificial intelligence AI (30) and used as data for training the AI. FIG. 6 illustrates a functional overview of AI learning and verification. The artificial intelligence AI (30) recognizes or memorizes an ultrasonic image captured for a medium while allowing the mechanical wave generated from the normally operating dynamic wave generator to propagate through a standard medium with well-known acoustic characteristics or a reference medium suitable for verifying the propagation characteristics of the dynamic wave (e.g., distilled water, oil, gel, or ultrasonic tissue mimicking material), and learns (deep learning) the pattern of the ultrasonic image according to the output setting of the shock wave generator. AI coding for deep learning can be implemented through programs such as TensorFlow, which has image recognition capabilities.

[0050] The learned AI then goes through a verification step to determine whether the presented ultrasound image is an image collected from a medium through which shock waves propagate, or to estimate the set output of the shock wave generator from the presented ultrasound image. This allows for additional learning to ensure that the error (output setting estimation error or abnormal ultrasound image identification error) remains within a set threshold, thereby maintaining the quality of learning. Through this process, the AI ​​is trained to estimate the operating state or output conditions of the shock wave generator when an ultrasound image is input.

[0051] Figure 7 is a flowchart of a method for monitoring the output and performance of an ultrasonic image-based dynamic wave generator using AI according to an embodiment of the present invention. Figure 7 illustrates a functional flowchart for using learned AI to observe an ultrasonic image of a medium through which a shock wave generated from a shock wave generator propagates, thereby determining, within a margin of error, whether the shock wave generator is operating normally or whether there is a difference between the actual generated acoustic output and the settings of the shock wave generator.

[0052] The learned artificial intelligence AI (30) receives an ultrasonic image of the medium through which the shock wave propagates when the shock wave generating device (10) operates from the ultrasonic imaging device (20). The AI ​​determines whether the performance (R) of the shock wave generating device (10) is normal within the allowed error (Th(R), Th(X)) from the provided ultrasonic image (if R <Th(R))(S21 단계), 충격파 발생 장치(10)의 성능(R)이 정상인 경우(R<Th(R))에 음향 출력 설정 정보(X)를 추정한다(S22 단계). 여기서 충격파 발생 장치(10)의 성능(R, %)은 "n / N * 100"을 의미하고, 여기서 N은 제공된 초음파 영상의 총 개수를 나타내며, n은 제공된 영상 중 비정상적인 영상의 개수를 나타낸다. 한편, 충격파 발생 장치(10)의 성능(R)이 정상이 아닌 경우(R<Th(R)), 성능 이상으로 판단한다.

[0053] After estimating the sound output setting information (X), it is determined whether the difference (X-X0) between the estimated sound setting information (X) and the actual output information (X0) is within the allowed error (Th(X)) (step S23). If the difference (X-X0) between the estimated sound setting information (X) and the actual output information (X0) is within the allowed error (Th(X)), it is determined to be normal operation. On the other hand, if the difference (X-X0) between the estimated sound setting information (X) and the actually set output information (X0) is not within the allowed error (Th(X)), it is determined to be an output abnormality.

[0054] According to the present invention, an AI is trained to recognize and remember an ultrasonic image that visualizes the acoustic characteristics of a reference medium (e.g., distilled water, oil, gel, or ultrasonic tissue mimicking material) through which a mechanical wave generated under output setting conditions of a mechanical wave generating device, for example, a shock wave generating device, propagates, and the trained AI can be utilized to monitor changes in the acoustic output and / or performance of the mechanical wave generating device.

[0055] Although the embodiments of the present invention have been described above, the scope of the present invention is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concept of the present invention defined in the following claims also fall within the scope of the present invention.

Claims

1. A method for monitoring the output and performance of a dynamic wave generator using AI based on ultrasonic imaging for generating dynamic waves, A method for monitoring the output and performance of a dynamic wave generator based on an ultrasonic image using AI, wherein the method comprises: an ultrasonic imaging device configured to acquire an ultrasonic image of a medium through which the above-mentioned dynamic wave propagates; and AI that recognizes and learns the ultrasonic image, wherein at least one of the output and performance of the dynamic wave generator is monitored.

2. In paragraph 1, The above-mentioned dynamic waves are ultrasound and shock waves that can be expected to have an effect suitable for medical purposes by inducing acoustic characteristics and biological changes in the investigated target tissue. This is a method for monitoring the output and performance of an AI-based ultrasound image-based dynamic wave generation device.

3. In paragraph 1, A method for monitoring the output and performance of an AI-based ultrasonic image-based dynamic wave generator, wherein the medium through which the above-mentioned dynamic wave propagates includes a reference medium having well-known acoustic properties or suitable for confirming the propagation properties of the above-mentioned dynamic wave.

4. In paragraph 1, A method for monitoring the output and performance of an ultrasonic imaging-based dynamic wave generator using AI, wherein the ultrasonic imaging device is triggered at the time when the dynamic wave generator operates to record a series of ultrasonic images of the medium through which the dynamic wave propagates and transmit the recorded images to the AI.

5. In paragraph 4, A method for monitoring the output and performance of an AI-based ultrasonic image-based dynamic wave generator, wherein the ultrasonic image includes a passive cavitation image (PCI) capable of visualizing the dynamic characteristics of a group of bubbles generated in a medium by the dynamic wave.

6. In paragraph 1, A method for monitoring the output and performance of an ultrasonic wave generator using AI, wherein the relative position information of an ultrasonic probe included in the ultrasonic wave generator and the ultrasonic imaging device is shared so that the ultrasonic imaging device can selectively image a cross-section of a sound field formed by the ultrasonic wave, and the position of the ultrasonic probe can be flexibly controlled.

7. In paragraph 1, A method for monitoring the output and performance of an ultrasonic image-based dynamic wave generation device using AI, wherein the AI performs preprocessing to remove background images or other noise from the ultrasonic image received from the ultrasonic imaging device before the dynamic wave is irradiated.

8. In paragraph 1, A method for monitoring the output and performance of a dynamic wave generator based on an ultrasonic image using AI, wherein the AI performs pre-learning to recognize and remember a series of ultrasonic images of the medium according to the output settings of the dynamic wave generator when the dynamic wave generator is operating normally.

9. In paragraph 8, The above-mentioned pre-learned AI receives a series of ultrasonic images collected from the medium when the above-mentioned dynamic wave is irradiated, estimates the output set in the above-mentioned dynamic wave generating device or determines the degree of inclusion of abnormal images, and provides information on the acoustic output or performance of the above-mentioned dynamic wave generating device and changes in each thereof, wherein the above-mentioned pre-learned AI provides a method for monitoring the output and performance of a dynamic wave generating device based on ultrasonic images using AI.

10. In a device for monitoring the output and performance of a dynamic wave generator that generates dynamic waves, an ultrasonic image-based device for monitoring the output and performance of a dynamic wave generator that generates dynamic waves, An ultrasonic imaging device configured to obtain an ultrasonic image of a medium through which the above-mentioned dynamic wave propagates, and Including AI that recognizes and learns the above ultrasound image, An output and performance monitoring device for a dynamic wave generator based on an ultrasonic image utilizing AI, capable of monitoring at least one of the output and performance of the dynamic wave generator through the ultrasonic image of the medium in which the dynamic wave is irradiated.

Images