Visual intelligent early warning platform for placenta implantable diseases

Abstract

The invention relates to the technical field of placenta implantable disease early warning, and particularly discloses a placenta implantable disease visual intelligent early warning platform which comprises an intestinal gas detection module, an influence analysis module, an intelligent optimization module and an early warning analysis module. Detecting the existence condition of intestinal gas in the ultrasonic image in real time, and calculating the influence precision coefficient of the intestinal gas on placenta imaging; evaluating the influence degree of intestinal gas on the ultrasonic image; when the influence degree of intestinal gas on an ultrasonic image is large, the position and the angle of the ultrasonic probe are automatically adjusted through a three-dimensional motion control system; acoustic wave parameters of ultrasonic acoustic waves are automatically adjusted; the focusing depth and width of the sound beam are accurately adjusted; dynamically optimizing the overall gain and the TGC curve of each depth layer; the early warning coefficient of the placenta is analyzed, the imaging quality can be further optimized, it is ensured that key features of the placenta are accurately captured, the diagnostic value of the ultrasonic image is improved, and the risks of misdiagnosis and missed diagnosis are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of early warning of placenta accreta diseases and relates to a visual intelligent early warning platform for placenta accreta diseases. Background Art

[0002] Placenta accreta diseases refer to the abnormal deep implantation of the placenta into the uterine wall, which may lead to severe postpartum hemorrhage and other complications. Therefore, it is necessary to detect the position and status of the placenta through ultrasonic images to timely identify and manage these potential risks. However, when intestinal gas interferes with the conduction of ultrasonic waves, it will cause the image to be blurred, especially affecting the visibility of the structure behind the placenta, thereby increasing the risk of misdiagnosis or missed diagnosis and making the operation more difficult and time-consuming.

[0003] When the current technology performs ultrasonic image detection on the placenta, it often faces the interference of intestinal gas, which will lead to a decline in image quality, especially the structure at the back may become blurred. Intestinal gas will reflect and scatter ultrasonic waves, forming a shadowing effect, thereby obscuring or distorting the image of the tissue behind. This interference not only affects the clear imaging of the placenta, but also may cause doctors to be unable to accurately evaluate the position, shape and health status of the placenta, increasing the uncertainty of diagnosis and the risk of misdiagnosis.

[0004] When the existing technology performs ultrasonic image detection of the placenta and faces the interference of intestinal gas, it usually adopts techniques such as asking the pregnant woman to lie on her side or capturing the moment of breathing to reduce the impact of gas on image quality. Although these methods can improve the clarity of the image to a certain extent, they still have limitations. The lying-on-side posture may not be suitable for all pregnant women, especially in the late pregnancy, and the method of capturing the moment of breathing requires a high level of skill and experience of the operator. In addition, these methods cannot completely eliminate gas interference, which may lead to the omission of key details, thus affecting the accuracy and timeliness of diagnosis. Therefore, it is particularly important to develop an intelligent early warning platform that can detect and compensate for intestinal gas interference in real time to ensure the accurate diagnosis and management of placenta accreta diseases. Through advanced technical means, automatically optimizing the image quality can improve the detection efficiency, reduce the misdiagnosis rate, and provide more reliable support for clinical decision-making. Summary of the Invention

[0005] In view of the above problems existing in the prior art, the present invention provides a visual intelligent early warning platform for placenta accreta diseases to solve the above technical problems.

[0006] In order to achieve the above object and other objects, the technical solution adopted by the present invention is as follows: On the one hand, the present invention provides a visualization intelligent early warning platform for placenta implantation diseases, including an intestinal gas detection module, an impact analysis module, an intelligent optimization module, and an early warning analysis module. The above-mentioned modules are connected by wired and / or wireless connection methods to achieve data transmission between modules; The intestinal gas detection module is used to detect and locate the intestinal gas area in the ultrasonic image in real time, and then detect the presence of intestinal gas in the ultrasonic image in real time, and calculate its influence precision coefficient on placenta imaging; The impact analysis module is used to evaluate the impact degree of intestinal gas on the ultrasonic image; The intelligent optimization module, when the influence degree of intestinal gas on the ultrasonic image is large, starts the following optimization measures: (i) Probe position and angle adjustment sub-module: Automatically adjust the position and angle of the ultrasonic probe through a three-dimensional motion control system; (ii) Frequency adjustment sub-module: Independently adjust the acoustic wave parameters of the ultrasonic wave; (iii) Sound beam focusing optimization sub-module: Precisely adjust the focusing depth and width of the sound beam; (iv) Gain compensation sub-module: Dynamically optimize the overall gain and the TGC curve of each depth layer; The early warning analysis module extracts the key features of the placenta based on the placenta ultrasonic image, and finally analyzes the early warning coefficient of the placenta.

[0007] Calculate its influence precision coefficient on placenta imaging. The specific calculation process is as follows: Preprocess the ultrasonic image, remove noise through Gaussian filtering, and then apply the Canny edge detection algorithm to identify the edges in the ultrasonic image; Use threshold segmentation technology to divide the ultrasonic image into intestinal gas and background, and use morphological operations to remove artifacts; Mark and analyze the intestinal gas area through connected component analysis, calculate the area of the intestinal gas area, and then obtain the total number of pixels A of the intestinal gas area; Synchronously collect the resolution of the ultrasonic image, denoted as M×N, where M represents the height pixel value of the ultrasonic image and N is the width pixel value of the ultrasonic image; Thus, calculate the influence precision coefficient of the intestinal gas area on placenta imaging , where W is a set weight coefficient used to adjust the influence degree of the intestinal gas area on placenta imaging.

[0008] The logic for evaluating the influence degree of intestinal gas on the ultrasonic image is as follows: Judge the influence precision coefficient of the intestinal gas area on placenta imaging with the set influence upper limit threshold. If the influence precision coefficient of the intestinal gas area on placenta imaging is greater than the set influence upper limit threshold, it is determined that the influence degree of intestinal gas on the ultrasonic image is large, otherwise it is determined that the influence degree of intestinal gas on the ultrasonic image is small.

[0009] The operation logic of the probe position and angle adjustment sub-module is as follows: Obtain the position of the ultrasonic probe , and represent the angle of the ultrasonic probe in Euler angles as (α, β, γ); Define the objective function F to evaluate the quality of the ultrasonic image. F is the quality of the ultrasonic image. The calculation formula of the objective function F is: F = ω1×SNR + ω2×Contrast, where SNR is the signal-to-noise ratio of the ultrasonic image, Contrast is the contrast of the ultrasonic image, and ω1 and ω2 are the set weight coefficients used to balance the influence of different indicators; Furthermore, use the following calculation formula to adjust the position of the ultrasonic probe: , where η is the learning rate, and the partial derivative is calculated by the finite difference method, is the calculated adjustment position of the ultrasonic probe; Similarly, use the calculation formula to adjust the angle of the ultrasonic probe , is the calculated adjustment angle of the ultrasonic probe; Apply the calculated adjustment position and calculated adjustment angle of the ultrasonic probe to the current ultrasonic probe, and use the updated ultrasonic probe position and angle to obtain new ultrasonic image data, thereby obtaining the signal-to-noise ratio and contrast of the new ultrasonic image; Substitute the signal-to-noise ratio and contrast of the new ultrasonic image into the objective function F to obtain a new objective function ; If , it is determined that the new objective function has converged, and stop adjusting the position and angle of the ultrasonic probe; If , continue to iterate on the basis of the calculated adjustment position and calculated adjustment angle of the ultrasonic probe until the new objective function converges; τ is the set threshold.

[0010] The operation logic of the frequency adjustment sub-module is as follows: The acoustic parameters of the ultrasonic wave are divided into axial resolution, lateral resolution, acoustic wave frequency, and wavelength; Obtain the average fat thickness d of the placental mother, in cm, and thus use the empirical formula , estimate the attenuation coefficient ε of the fat tissue corresponding to the placental mother, κ is the set thickness influence factor, in dB / cm^2 / MHz, is the basic attenuation coefficient, in dB / cm / MHz; Based on the acoustic wave attenuation model , A is the total attenuation value of the sound wave, f is the sound wave frequency, and d1 is the tissue thickness; the maximum allowable attenuation is set , and the sound wave frequency of the ultrasonic wave is inversely deduced in combination with the specified imaging depth value D ; Calculate the wavelength of the ultrasonic wave , where c is the speed of sound; Calculate the axial resolution of the ultrasonic wave ; Calculate the lateral resolution of the ultrasonic wave , F1 is the focal length, and D' is the aperture of the transducer.

[0011] The operation logic of the sound beam focusing optimization sub-module is as follows: Step 1: Obtain the tissue density ρ and temperature T of the placental mother, and calculate the adjusted speed of sound , where T0 and ρ0 represent the standard temperature and standard density respectively, represent the correction coefficients of temperature and density respectively; Furthermore, calculate the new focusing depth of the ultrasonic beam ; Step 2: Consider the influence of the absorption amount and scattering amount of the placental mother tissue on the width of the ultrasonic beam, and calculate the width adjustment value of the ultrasonic beam , where W is the initial width of the ultrasonic beam, are the absorption amount and scattering amount of the placental mother tissue respectively; Furthermore, calculate the new focusing width of the ultrasonic beam ; is the set resolution adjustment factor.

[0012] The operation logic of the gain compensation sub-module is as follows: Obtain the echo signal intensities at different depths in the placental mother through the ultrasonic beam, calculate their mean value, and obtain the mean value of the echo signal intensities in the placental mother, which is denoted as Y1; Synchronously record the target echo signal intensity of the placental mother as Y2; Therefore, the adjustment formula for the overall gain G of the ultrasonic beam is; ; is the set gain adjustment function; Obtain the echo signal intensities at different depths in the placental mother , where j is the number of different depths; Therefore, the time gain compensation adjustment formula for the ultrasonic beam corresponding to different depths is: , is the initial time gain value, is a set linear gain compensation coefficient, representing the gain change per unit depth, j is the depth, representing the distance from the ultrasound probe to a certain depth in the image; ξ is a gain adjustment coefficient related to the depth, is the target signal intensity at the j-th depth.

[0013] The analysis process of the warning coefficient of the placenta is as follows: The key features of the placenta are divided into the coordinates of the position center point, area, and perimeter; Evaluate the position offset evaluation index of the placenta , is the coordinate of the position center point of the placenta, is the ideal coordinate of the position center point of the placenta; Calculate the shape factor of the placenta , MJ is the area of the placenta, and ZC is the perimeter of the placenta; Finally, comprehensively analyze the warning coefficient of the placenta , where are the set weight coefficients respectively, to reflect the importance of position and shape to the warning.

[0014] On the other hand, the present invention provides a visual intelligent warning device for placenta implantation diseases, including a processor, a memory, and a communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, it executes to implement a visual intelligent warning platform for placenta implantation diseases as described in the present invention.

[0015] As described above, the visual intelligent warning platform for placenta implantation diseases provided by the present invention has at least the following beneficial effects: The visual intelligent warning platform for placenta implantation diseases provided by the present invention can significantly improve the accuracy and reliability of ultrasonic imaging by real-time detecting and locating the intestinal gas area in the ultrasonic image and calculating its influence accuracy coefficient on placenta imaging. Intestinal gas is a common interference factor in ultrasonic imaging, which will lead to a decline in image quality, thus affecting the accurate assessment of the position and health status of the placenta. By accurately identifying and quantifying the influence of intestinal gas, the parameters and position of the ultrasonic probe can be dynamically adjusted to ensure the best imaging effect. Using a three-dimensional motion control system to automatically adjust the position and angle of the ultrasonic probe can effectively avoid the intestinal gas interference area, thereby improving the clarity and detail performance of the imaging. At the same time, autonomously adjusting the parameters of ultrasonic waves, such as frequency and intensity, and precisely adjusting the focusing depth and width of the sound beam can further optimize the imaging quality and ensure that the key features of the placenta are accurately captured.

[0016] Dynamically optimizing the overall gain and the time gain compensation (TGC) curves of each depth layer can compensate for signal losses caused by acoustic attenuation of intestinal gas and other tissues, making the brightness and contrast of the image more uniform and clear. This multi-level optimization strategy not only improves the diagnostic value of the image but also reduces the risks of misdiagnosis and missed diagnosis. Finally, by analyzing the warning coefficient of the placenta, more comprehensive and accurate clinical information can be provided to help doctors better evaluate the health status and potential risks of the placenta. This series of automated and intelligent adjustment measures not only improve the efficiency of ultrasound examinations but also enhance the safety and comfort of patients, having important clinical application value and necessity. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 Schematic diagram of the connection of each module of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The above content is only an example and explanation of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by the claims of the present invention, they should fall within the protection scope of the present invention.

[0020] Embodiment 1: Please refer to Figure 1 As shown, a visual intelligent warning platform for placenta implantation diseases includes an intestinal gas detection module, an impact analysis module, an intelligent optimization module, and a warning analysis module. The above-mentioned modules are connected by wired and / or wireless connection methods to achieve data transmission between the modules; The intestinal gas detection module is used to detect and locate the intestinal gas area in the ultrasound image in real time, and then detect the presence of intestinal gas in the ultrasound image in real time and calculate its influence accuracy coefficient on placenta imaging; Calculating its influence accuracy coefficient on placenta imaging, the specific calculation process is as follows: Preprocess the ultrasound image, remove noise through Gaussian filtering, and then apply the Canny edge detection algorithm to identify the edges in the ultrasound image; Use threshold segmentation technology to divide the ultrasound image into intestinal gas and background, and use morphological operations to remove artifacts; Mark and analyze the intestinal gas area through connected component analysis, calculate the area of the intestinal gas area, and then obtain the total number of pixels A of the intestinal gas area; Synchronously collect the resolution of the ultrasonic image, denoted as M×N, where M represents the height pixel value of the ultrasonic image and N is the width pixel value of the ultrasonic image; Thus, calculate the influence precision coefficient of the intestinal gas area on placenta imaging , where W is a set weight coefficient used to adjust the influence degree of the intestinal gas area on placenta imaging.

[0021] An influence analysis module for evaluating the influence degree of intestinal gas on the ultrasonic image; The logic for evaluating the influence degree of intestinal gas on the ultrasonic image is as follows: Judge the influence precision coefficient of the intestinal gas area on placenta imaging and the set upper influence threshold. If the influence precision coefficient of the intestinal gas area on placenta imaging is greater than the set upper influence threshold, it is determined that the influence degree of intestinal gas on the ultrasonic image is large; otherwise, it is determined that the influence degree of intestinal gas on the ultrasonic image is small.

[0022] When it is determined that the influence degree of intestinal gas on the ultrasonic image is small, it is recommended that the patient adjust the body position or control breathing to further improve the imaging conditions.

[0023] An intelligent optimization module that, when the influence degree of intestinal gas on the ultrasonic image is large, starts the following optimization measures: (i) Probe position and angle adjustment sub-module: Automatically adjust the position and angle of the ultrasonic probe through a three-dimensional motion control system; The operation logic of the probe position and angle adjustment sub-module is as follows: Obtain the position of the ultrasonic probe , and represent the angle of the ultrasonic probe in Euler angles as (α, β, γ); Define an objective function F for evaluating the quality of the ultrasonic image. F is the quality of the ultrasonic image, and the calculation formula of the objective function F is: F = ω1×SNR + ω2×Contrast, where SNR is the signal-to-noise ratio of the ultrasonic image, Contrast is the contrast of the ultrasonic image, and ω1 and ω2 are respectively set weight coefficients used to balance the influence of different indicators; Furthermore, use the following calculation formula to adjust the position of the ultrasonic probe: , where η is the learning rate, and the partial derivative is calculated by the finite difference method, is the calculated adjustment position of the ultrasonic probe; Similarly, use the calculation formula to adjust the angle of the ultrasonic probe , is the calculated adjustment angle of the ultrasonic probe; Apply the calculated adjustment position and calculated adjustment angle of the ultrasound probe to the current ultrasound probe, and use the updated position and angle of the ultrasound probe to obtain new ultrasound image data, thereby obtaining the signal-to-noise ratio and contrast of the new ultrasound image; Substitute the signal-to-noise ratio and contrast of the new ultrasound image into the objective function F to obtain a new objective function ; If , it is determined that the new objective function has converged, and stop adjusting the position and angle of the ultrasound probe; If , continue to iterate based on the calculated adjustment position and calculated adjustment angle of the ultrasound probe until the new objective function converges; τ is a set threshold.

[0024] The signal-to-noise ratio (SNR) and contrast are important indicators for measuring image quality. SNR can be calculated by the ratio of signal intensity to noise intensity, while contrast reflects the difference between different gray values in the image. By constructing the objective function F to synthesize these indicators, the system can automatically adjust the probe parameters to optimize the imaging effect. By calculating the gradient in real time and adjusting the probe position and angle, the system can dynamically respond to different imaging conditions, reduce human operation errors, and improve the diagnostic efficiency and accuracy.

[0025] (ii) Frequency adjustment sub-module: Autonomously adjust the acoustic parameters of the ultrasonic wave; The acoustic parameters of the ultrasonic wave are divided into axial resolution, lateral resolution, acoustic frequency, and wavelength; Obtain the average fat thickness d of the placental mother, in cm, and thus use the empirical formula to estimate the attenuation coefficient ε of the adipose tissue corresponding to the placental mother. κ is a set thickness influence factor, in dB / cm^2 / MHz, which can be estimated according to specific experimental data or literature, usually between 0.01 and 0.05; is the basic attenuation coefficient, in dB / cm / MHz, usually set to 0.5 dB / cm / MHz; Based on the acoustic attenuation model , A is the total attenuation value of the acoustic wave, f is the acoustic frequency, and d1 is the tissue thickness; ensure the effective penetration of the acoustic wave signal within the specified depth, set the maximum allowable attenuation , and inversely deduce the acoustic frequency of the ultrasonic wave in combination with the specified imaging depth value D ; In the above formula, tissue thickness and fat thickness refer to different concepts: Tissue thickness: Usually refers to the depth of the entire tissue that the acoustic wave needs to penetrate, including all layers (such as skin, fat, muscle, etc.).

[0026] Fat thickness: Specifically refers to the thickness of the fat layer, and is only used to calculate the attenuation effect of fat on sound waves.

[0027] When calculating the attenuation coefficient, the fat thickness is used to estimate the specific attenuation effect of the fat layer on sound waves. The tissue thickness is used to determine the total depth that the sound wave needs to penetrate.

[0028] In the sound wave attenuation model, f represents the sound wave frequency, and the imaging depth D is the depth at which it is desired for the sound wave to penetrate effectively. There is no direct conversion relationship between the two, but rather factors that need to be considered simultaneously when selecting an appropriate frequency; The relationship between the sound wave frequency f and the imaging depth D is mainly reflected in the following points: 1. Ultrasonic sound waves provide higher resolution, but their attenuation is faster and the penetration depth is shallower. Low-frequency sound waves have slower attenuation and can penetrate deeper tissues, but the resolution is lower; 2. In order to obtain a clear image at a specific depth, an appropriate frequency must be selected so that the attenuation does not exceed the signal processing capacity of the device. This is why it is necessary to calculate the frequency inversely according to the imaging depth; 3. In practical applications, when selecting the frequency, the acoustic characteristics of the tissue (such as the attenuation coefficient) and the depth of the imaging target need to be considered to ensure sufficient signal strength and image quality; Therefore, the relationship between the frequency and the imaging depth is not a direct conversion, but rather they affect each other through the attenuation model to help select the most suitable imaging parameters.

[0029] Calculating the wavelength of ultrasonic sound waves , where c is the speed of sound, which is approximately 1540 m / s in human tissues; Calculating the axial resolution of ultrasonic sound waves ; Calculating the lateral resolution of ultrasonic sound waves , where F1 is the focal length and D' is the aperture of the transducer.

[0030] (iii) Beam focusing optimization sub-module: Precisely adjust the focusing depth and width of the beam; The operation logic of the beam focusing optimization sub-module is as follows: Step 1: Obtain the tissue density ρ and temperature T of the placental mother, and calculate the adjusted speed of sound , where T0 and ρ0 respectively represent the standard temperature and standard density, respectively represent the correction coefficients for temperature and density; Furthermore, calculate the new focusing depth of the ultrasonic beam ; Step 2: Consider the influence of the absorption and scattering amounts of the placental maternal tissue on the width of the ultrasonic beam, and calculate the width adjustment value of the ultrasonic beam , where W is the initial width of the ultrasonic beam, are the absorption and scattering amounts of the placental maternal tissue respectively; Furthermore, calculate the new focused width of the ultrasonic beam ; is the set resolution adjustment factor.

[0031] (iv) Gain compensation sub-module: Dynamically optimize the overall gain and the TGC curves of each depth layer; The operation logic of the gain compensation sub-module is as follows: Obtain the echo signal intensities at different depths in the placental mother through the ultrasonic beam, calculate their mean values, obtain the mean value of the echo signal intensities in the placental mother, and denote it as Y1; Synchronously denote the target echo signal intensity of the placental mother as Y2; Therefore, the adjustment formula for the overall gain G of the ultrasonic beam is; ; is the set gain adjustment function; Obtain the echo signal intensities at different depths in the placental mother , where j is the number of different depths; Therefore, the time gain compensation adjustment formula for the ultrasonic beam corresponding to different depths is: , is the initial time gain value, is the set linear gain compensation coefficient, indicating the gain change per unit depth, j is the depth, representing the distance from the ultrasonic probe to a certain depth in the image; ξ is the gain adjustment coefficient related to the depth, is the target signal intensity at the j-th depth.

[0032] The above calculation formulas effectively cope with the changes in different depths and signal intensities by dynamically adjusting the overall gain and time gain compensation (TGC). The advantage of this method is that it can automatically compensate for the signal attenuation caused by intestinal gas, ensuring the clarity and contrast of the image at each depth. By analyzing the echo signal intensity, the system can optimize the gain setting in real time, making the image more uniform, reducing diagnostic errors, and improving the accuracy of detecting placental implantation diseases.

[0033] Early warning analysis module, based on the placental ultrasound image, extracts the key features of the placenta, and finally analyzes the early warning coefficient of the placenta.

[0034] The analysis process of analyzing the early warning coefficient of the placenta is as follows: The key features of the placenta are divided into the coordinates of the central point of the position, the area, and the perimeter; The position offset evaluation index of the placenta is evaluated , is the coordinate of the central point of the placenta's position, is the coordinate of the ideal central point of the placenta's position; Calculate the shape factor of the placenta , where MJ is the area of the placenta and ZC is the perimeter of the placenta; Finally, comprehensively analyze the warning coefficient of the placenta , where are the set weight coefficients respectively, to reflect the importance of position and shape for warning.

[0035] Embodiment 2: A visual intelligent warning device for placenta implantation diseases shown according to an exemplary embodiment includes a processor, a memory, and a communication bus; A computer-readable program executable by the processor is stored on the memory; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, it executes to implement a visual intelligent warning platform for placenta implantation diseases as described in the present invention.

[0036] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0037] It should be understood that determining B based on A does not mean determining B only based on A, and B can also be determined based on A and / or other information.

[0038] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present application, and all should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0039] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A visual intelligent early warning platform for placenta accreta disease, characterized in that: It includes an intestinal gas detection module, an impact analysis module, an intelligent optimization module and an early warning analysis module, and the above modules are connected by wired and / or wireless connection to achieve data transmission between the modules; The intestinal gas detection module is used to detect and locate the intestinal gas area in the ultrasound image in real time, and then detect the presence of intestinal gas in the ultrasound image in real time, and calculate the accuracy coefficient of its influence on the placental imaging; Impact analysis module, used to evaluate the degree of influence of intestinal gas on ultrasound images; Intelligent optimization module, when the intestinal gas has a significant impact on the ultrasound image, the following optimization measures are initiated: (i) Probe position and angle adjustment submodule: Automatically adjust the position and angle of the ultrasound probe through a three-dimensional motion control system; (ii) Frequency adjustment submodule: autonomously adjusts the acoustic wave parameters of the ultrasonic sound wave; (iii) Beam focusing optimization submodule: precisely adjusts the focusing depth and width of the acoustic beam; (iv) Gain compensation submodule: dynamically optimizes the overall gain and TGC curve of each depth layer; The early warning analysis module extracts the key features of the placenta based on the placental ultrasound image and finally analyzes the early warning coefficient of the placenta.

2. A visual intelligent early warning platform for placenta accreta disease according to claim 1, characterized in that: Calculate the accuracy coefficient of its impact on placental imaging. The specific calculation process is as follows: The ultrasound image is preprocessed by removing noise through Gaussian filtering, and then the Canny edge detection algorithm is applied to identify the edges in the ultrasound image; The ultrasound images were separated into bowel gas and background using threshold segmentation techniques, and artifacts were removed using morphological operations; The intestinal gas region is marked and analyzed by connected component analysis, the area of ​​the intestinal gas region is calculated, and then the total number of pixels A in the intestinal gas region is obtained; The resolution of the synchronously acquired ultrasound image is recorded as M×N, where M represents the height pixel value of the ultrasound image and N represents the width pixel value of the ultrasound image; The accuracy coefficient of the effect of intestinal gas area on placental imaging is calculated , W is the set weight coefficient, which is used to adjust the influence of intestinal gas area on placental imaging.

3. A visual intelligent early warning platform for placenta accreta disease according to claim 2, characterized in that: The logic for assessing the extent to which bowel gas affects ultrasound images is as follows: The accuracy coefficient of the influence of the intestinal gas area on the placental imaging is judged against the set upper limit threshold of influence. If the accuracy coefficient of the influence of the intestinal gas area on the placental imaging is greater than the set upper limit threshold of influence, it is judged that the influence of the intestinal gas on the ultrasound image is greater; otherwise, it is judged that the influence of the intestinal gas on the ultrasound image is less.

4. A visual intelligent early warning platform for placenta accreta disease according to claim 1, characterized in that: The operation logic of the probe position and angle adjustment submodule is as follows: Obtaining the location of the ultrasound probe , the angle of the ultrasound probe is expressed as (α, β, γ) using Euler angles; The objective function F is defined to evaluate the quality of ultrasound images. F is the quality of ultrasound images. The calculation formula of the objective function F is: F=ω1×SNR+ω2×Contrast, where SNR is the signal-to-noise ratio of ultrasound images, Contrast is the contrast of ultrasound images, and ω1 and ω2 are respectively the set weight coefficients used to balance the influence of different indicators. Then the position of the ultrasound probe is adjusted using the following calculation formula: , where η is the learning rate and the partial derivative Calculated by the finite difference method, Calculate and adjust the position of the ultrasound probe; Similarly, use the calculation formula to adjust the angle of the ultrasound probe , Adjust the angle for ultrasound probe calculations; Applying the calculated adjusted position and calculated adjusted angle of the ultrasound probe to the current ultrasound probe, and acquiring new ultrasound image data using the updated ultrasound probe position and angle, thereby acquiring a signal-to-noise ratio and contrast of a new ultrasound image; Substitute the signal-to-noise ratio and contrast of the new ultrasound image into the objective function F to obtain the new objective function ; like , then the new objective function is judged to have reached convergence, and the adjustment of the position and angle of the ultrasonic probe is stopped; like , then the iteration continues based on the calculated adjustment position and the calculated adjustment angle of the ultrasonic probe until the new objective function converges; τ is the set threshold.

5. A visual intelligent early warning platform for placenta accreta disease according to claim 1, characterized in that: The operating logic of the frequency adjustment submodule is as follows: The acoustic wave parameters of the ultrasonic sound wave are divided into axial resolution, lateral resolution, acoustic wave frequency and wavelength; Get the average maternal fat thickness of the placenta d, in cm, and use the empirical formula , the attenuation coefficient ε of the maternal placenta corresponding to the fat tissue is estimated, κ is the set thickness influence factor, the unit is dB / cm^2 / MHz, is the basic attenuation coefficient in dB / cm / MHz; Based on the sound wave attenuation model , A is the total attenuation value of the sound wave, f is the sound wave frequency, d1 is the tissue thickness; set the maximum allowable attenuation , combined with the specified imaging depth value D, the sound wave frequency of the ultrasonic sound wave is obtained by reverse calculation ; Calculate the wavelength of ultrasonic sound waves , c is the speed of sound; Calculating the axial resolution of ultrasound sound waves ; Calculating the lateral resolution of ultrasound sound waves , F1 is the focal length, and D' is the aperture of the transducer.

6. A visual intelligent early warning platform for placenta accreta disease according to claim 1, characterized in that: The operation logic of the beam focusing optimization submodule is as follows: Step 1: Obtain the tissue density ρ and temperature T of the placenta and calculate the adjusted sound speed , T0 and ρ0 represent standard temperature and standard density respectively, denote the correction factors for temperature and density respectively; Then the new focal depth of the ultrasonic beam is calculated ; Step 2: Consider the effect of maternal tissue absorption and scattering on the width of the ultrasound beam and calculate the width adjustment value of the ultrasound beam , W is the initial width of the ultrasonic beam, are the absorption and scattering of the placental maternal tissue, respectively; Then the new focal width of the ultrasonic beam is calculated ; The resolution adjustment factor is set.

7. A visual intelligent early warning platform for placenta accreta disease according to claim 1, characterized in that: The operation logic of the gain compensation submodule is as follows: The echo signal strength at different depths in the placenta and the mother is obtained by ultrasonic beam, and the mean value is calculated to obtain the mean value of the echo signal strength in the placenta and the mother, which is recorded as Y1; The target echo signal intensity of the placenta and mother is simultaneously recorded as Y2; Therefore, the adjustment formula for the overall gain G of the ultrasonic beam is: ； is the set gain adjustment function; Obtain echo signal strength at different depths within the placenta , j is the number of different depths; Therefore, the time gain compensation adjustment formula for the ultrasonic beam corresponding to different depths is: , is the initial time gain value, is the set linear gain compensation coefficient, which represents the gain change per unit depth; j is the depth, which represents the distance from the ultrasound probe to a certain depth in the image; ξ is the gain adjustment coefficient related to the depth, is the target signal strength at the jth depth.

8. A visual intelligent early warning platform for placenta accreta disease according to claim 1, characterized in that: The analysis process of the warning coefficient of the placenta is as follows: The key features of the placenta are the coordinates of the center point, area and perimeter; Assessment of placental position deviation index , is the coordinate of the center point of the placenta, The coordinates of the ideal center point of the placenta; Calculating the shape factor of the placenta , MJ is the area of ​​the placenta, ZC is the circumference of the placenta; Finally, the warning coefficient of the placenta was comprehensively analyzed ,in The weight coefficients are set respectively to reflect the importance of location and shape to the warning.

9. A visual intelligent early warning device for placenta accreta disease, characterized by: It is implemented based on a visual intelligent early warning platform for placenta accreta disease according to any one of claims 1 to 8, comprising a processor, a memory and a communication bus; The memory stores a computer-readable program executable by the processor; The communication bus realizes the connection and communication between the processor and the memory; When the processor executes the computer-readable program, it is implemented to implement a visual intelligent early warning platform for placenta accreta disease as described in any one of claims 1-8.

Images