Endometrial motility auxiliary measurement method and device based on ultrasound technology

By analyzing the historical matching and position distribution of pixel points in endometrial ultrasound images, calculating contribution degree and PCA dimensionality reduction, the reliability problem of endometrial peristalsis information analysis is solved, and the accuracy and credibility of measurement are improved Spend.

CN118924343BActive Publication Date: 2024-12-06BEIJING OBSTETRICS & GYNECOLOGY HOSPITAL CAPITAL MEDICAL UNIV
View PDF 11 Cites 0 Cited by

Patent Information

Application Number
CN202411426253.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-10-14
Publication Date
2024-12-06
Estimated Expiration
2044-10-14

AI Technical Summary

Technical Problem

In the prior art, the pixel points for endometrial peristalsis information analysis are not reliable, resulting in large errors in quantification data and affecting measurement accuracy.

Method used

By obtaining endometrial ultrasound images, the peristalsis analysis area is determined, the initial and final contribution degrees are calculated based on the historical matching of pixel points and the location distribution, and PCA dimensionality reduction is performed to optimize the data to obtain more accurate peristalsis information.

Benefits of technology

It improves the reliability and measurement accuracy of peristaltic information, reduces redundant information, and enhances the credibility of subsequent analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118924343B_ABST
    Figure CN118924343B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of endometrial peristalsis analysis, and specifically to an endometrial peristalsis auxiliary measurement method and device based on ultrasound technology. In the peristalsis analysis area corresponding to the ultrasound image of the current sampling frame, the method obtains the time-series continuous matching points of the pixel points according to the matching of the pixel points in the historical continuous sampling frames; the initial peristalsis contribution of each pixel point is obtained according to the position distribution of each pixel point and the degree of change in the distance of the continuous matching points, and the final peristalsis contribution is obtained by correction according to the consistent change of the adjacent pixel points, and the PCA dimension reduction process is optimized based on the final peristalsis contribution to obtain optimized dimension reduction data to obtain visualized peristalsis information. The present invention characterizes the contribution of peristaltic changes by the pixel points in the ultrasound image, adjusts the dimension reduction process to remove redundant information, and obtains more accurate visualized peristalsis information, so that subsequent measurement results are more accurate and reliable.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of endometrial motility analysis, and in particular to an endometrial motility auxiliary measurement method and device based on ultrasound technology. Background Art

[0002] The measurement of endometrial motility is due to its key role in regulating normal reproductive function and embryo implantation. Among them, endometrial motility changes at different stages of the menstrual cycle, which is consistent with the corresponding reproductive function of the uterus. Abnormal endometrial motility may cause symptoms such as infertility and abnormal uterine bleeding, and may be related to diseases such as uterine fibroids, endometriosis and polycystic ovary syndrome, so it is very important to measure endometrial motility.

[0003] Generally, when measuring endometrial motility, transvaginal ultrasound is used to obtain dynamic videos of the sagittal and coronal planes of the uterus. Since endometrial motility is difficult to identify, a visual quantitative assessment of the direction and amplitude of the motility in the endometrial image is usually performed to improve measurement accuracy. However, in the process of visual quantification of endometrial motility, there are many interference parts in the ultrasound image, and the pixel points involved in the motility information analysis are not reliable, resulting in some motility information being masked, so that the motility data after visual quantification still has large errors, affecting the accuracy of subsequent measurements. Summary of the invention

[0004] In order to solve the technical problem that the pixel points involved in the peristalsis information analysis in the prior art are not reliable enough, so that the peristalsis data after visual quantification still has large errors, the purpose of the present invention is to provide an endometrial peristalsis auxiliary measurement method and device based on ultrasound technology. The technical solution adopted is as follows:

[0005] The present invention provides an endometrial motility auxiliary measurement method based on ultrasound technology, the method comprising:

[0006] Acquire the endometrial ultrasound image of each sampling frame; determine the peristalsis analysis area according to the edge distribution in the endometrial ultrasound image of the current sampling frame; obtain the temporally continuous matching points and form a peristalsis matching point sequence according to the continuous matching of each pixel point in the peristalsis analysis area in the endometrial images of the historical continuous sampling frames;

[0007] In the peristalsis analysis area, according to the position distribution of each pixel point in the peristalsis analysis area, and the significance and period of the time series distance change of the matching point in the corresponding peristalsis matching point sequence, the initial peristalsis contribution of each pixel point is obtained;

[0008] In the peristalsis analysis area, according to the consistency of the position change of each pixel point and the adjacent pixel points in the corresponding peristalsis matching point sequence, the final peristalsis contribution of each pixel point is obtained in combination with the initial peristalsis contribution;

[0009] The PCA dimensionality reduction process of the pixels in the peristalsis analysis area is optimized based on the final peristalsis contribution to obtain optimized dimensionality reduction data of the current sampling frame; and visualized peristalsis information is obtained through the optimized dimensionality reduction data of the current sampling frame.

[0010] Furthermore, the method for obtaining the initial peristaltic contribution includes:

[0011] Obtaining the center point of the peristalsis analysis area; for any pixel point in the peristalsis analysis area, obtaining the near-center index of the pixel point based on the distance between the pixel point and the center point;

[0012] Obtain the matching center point of all matching points in the creeping matching point sequence of the pixel point; obtain the fluctuation degree index of the pixel point according to the degree of deviation between each matching point in the creeping matching point sequence of the pixel point and the matching center point;

[0013] According to the regularity of the position changes of all matching points in the creeping matching point sequence of the pixel point in time sequence, the periodicity index of the pixel point is obtained;

[0014] The initial creep contribution of the pixel is obtained by combining the centroid index, fluctuation index and periodic index of the pixel.

[0015] Furthermore, the method for obtaining the volatility index includes:

[0016] The center point of the corresponding positions of all matching points in the creeping matching point sequence of the pixel point on the image is taken as the matching center point;

[0017] The distance between each matching point and the matching center point in the creeping matching point sequence of the pixel point is used as the fluctuation distance of each matching point;

[0018] The sum of the fluctuation distances of all matching points corresponding to the pixel is calculated to obtain the fluctuation degree index of the pixel.

[0019] Furthermore, the method for obtaining the periodicity index includes:

[0020] Based on the positions of all matching points in the peristaltic matching point sequence of the pixel point in the endometrial ultrasound image, a time series scatter plot of the pixel point is obtained; and the time series scatter plot of the pixel point is subjected to Fourier transformation to obtain a spectrum plot;

[0021] Calculate the mean value of the signal strength in the spectrum of the pixel point to obtain the average signal strength of the pixel point; take the difference between the signal strength corresponding to each peak point in the spectrum and the average signal strength as the strength significance of each peak point;

[0022] The mean of the intensity significance of all peak points is negatively correlated to obtain the periodicity index of the pixel point.

[0023] Furthermore, the method for obtaining the final peristaltic contribution includes:

[0024] For any pixel point, other pixels within the preset neighborhood of the pixel point are regarded as the neighboring pixels of the pixel point;

[0025] According to the position deviation difference between the pixel point and each adjacent pixel point and the matching points in the same sequence in the corresponding creeping matching point sequence, the change consistency of each adjacent pixel point is obtained;

[0026] The final peristaltic contribution of the pixel is obtained by combining the change consistency of all neighboring pixels of the pixel and the initial peristaltic contribution of the pixel.

[0027] Furthermore, the method for obtaining the change consistency includes:

[0028] The difference between the pixel point and the matching point of each sequence number in the corresponding creeping matching point sequence on the horizontal coordinate is taken as the horizontal deviation of the pixel point at each sequence number; the difference between the pixel point and the matching point of each sequence number in the corresponding creeping matching point sequence on the vertical coordinate is taken as the vertical deviation of the pixel point at each sequence number;

[0029] For any adjacent pixel of the pixel, under each sequence number, the difference in horizontal deviation and vertical deviation between the pixel and the adjacent pixel are combined to obtain the deviation difference between the pixel and the adjacent pixel under each sequence number;

[0030] The deviation difference between the pixel and the adjacent pixel under all sequence numbers is combined and negative correlation mapping is performed to obtain the change consistency of the adjacent pixel.

[0031] Furthermore, combining the change consistency of all neighboring pixels of the pixel and the initial peristaltic contribution of the pixel to obtain the final peristaltic contribution of the pixel includes:

[0032] The sum of the change consistency of all adjacent pixels of the pixel is used as the local change adjustment degree of the pixel;

[0033] The product of the local change adjustment degree of the pixel point and the initial peristaltic contribution degree is normalized to obtain the final peristaltic contribution degree of the pixel point.

[0034] Furthermore, the PCA dimensionality reduction process of optimizing the pixels in the peristalsis analysis area based on the final peristalsis contribution to obtain optimized dimensionality reduction data of the current sampling frame includes:

[0035] The product of the sample variance of each pixel in the peristalsis analysis area in PCA dimensionality reduction and the final peristalsis contribution is used as the optimized sample variance of each pixel;

[0036] PCA dimensionality reduction is performed based on the optimized sample variance of each pixel in the creep analysis area to obtain optimized dimensionality reduction data.

[0037] Furthermore, the method of obtaining continuous matching points in time sequence and forming a peristaltic matching point sequence according to the continuous matching of each pixel point in the peristaltic analysis area in the endometrial images of the historical continuous sampling frames includes:

[0038] For any pixel point in the peristalsis analysis area, the matching point of the pixel point in the previous sampling frame in time sequence will be obtained through the feature matching algorithm, and the matching point of each matching point in the previous sampling frame in time sequence will be obtained continuously through the feature matching algorithm, and all the matching points corresponding to the pixel point will be arranged in time sequence to form a peristalsis matching point sequence.

[0039] The present invention also provides an endometrial motility auxiliary measurement device based on ultrasound technology, comprising:

[0040] A data acquisition unit is used to obtain an endometrial ultrasound image of each sampling frame; in the endometrial ultrasound image of the current sampling frame, determine a peristalsis analysis area according to edge distribution; and according to the continuous matching of each pixel point in the peristalsis analysis area in the endometrial images of historical continuous sampling frames, obtain time-series continuous matching points and form a peristalsis matching point sequence;

[0041] A data processing unit, configured to obtain an initial peristaltic contribution of each pixel point in the peristaltic analysis area according to the position distribution of each pixel point in the peristaltic analysis area and the significance and period of the time series distance change of the matching point in the corresponding peristaltic matching point sequence;

[0042] In the peristalsis analysis area, according to the consistency of the position change of each pixel point and the adjacent pixel points in the corresponding peristalsis matching point sequence, the final peristalsis contribution of each pixel point is obtained in combination with the initial peristalsis contribution;

[0043] The visualization display unit is used to optimize the PCA dimensionality reduction process of the pixel points in the peristalsis analysis area based on the final peristalsis contribution to obtain the optimized dimensionality reduction data of the current sampling frame; and obtain visualized peristalsis information through the optimized dimensionality reduction data of the current sampling frame.

[0044] The present invention has the following beneficial effects:

[0045] In the peristaltic analysis area corresponding to the ultrasonic image of the current sampling frame, the present invention obtains the temporal continuous matching points of the pixel points to form a peristaltic matching point sequence according to the matching of the pixel points in the historical continuous sampling frames, reflecting the possible peristaltic distribution of the pixel points in the historical sampling frames, thereby further analyzing the degree of peristaltic state characterization of each pixel point according to the continuous peristaltic distribution of each pixel point in history. Therefore, further according to the position distribution of each pixel point and the distance change degree of the continuous matching points, the initial peristaltic contribution of each pixel point is obtained from the critical degree of the position of the pixel point itself to the peristaltic analysis and the significant situation of the pixel point change to the peristaltic analysis, reflecting the influence of the pixel point on the peristaltic analysis, and considering the error of the matching analysis combined with the consistent change of the adjacent pixel points for correction, the final peristaltic contribution is obtained, reflecting the degree of participation of each pixel point in the peristaltic analysis, thereby retaining more features for the pixel points with high contribution. Therefore, the final peristaltic contribution is optimized in the PCA dimensionality reduction process based on the final peristaltic contribution, so that the important part is retained to a greater extent, and the optimized dimensionality reduction data is obtained to obtain more accurate and reliable visual peristaltic information. The present invention characterizes the contribution of peristaltic changes by pixel points in ultrasonic images, adjusts the dimensionality reduction process to remove redundant information, obtains more accurate visualized peristaltic information, and makes subsequent measurement results more accurate and reliable. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0047] Figure 1 A flow chart of an endometrial motility auxiliary measurement method based on ultrasound technology provided by one embodiment of the present invention;

[0048] Figure 2 A schematic diagram of an endometrial ultrasound image provided by an embodiment of the present invention;

[0049] Figure 3 A schematic diagram of a peristalsis analysis area provided by an embodiment of the present invention;

[0050] Figure 4 A flow chart of a method for obtaining initial peristaltic contribution provided by an embodiment of the present invention;

[0051] Figure 5A schematic structural diagram of an endometrial motility auxiliary measurement device based on ultrasound technology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of an endometrial peristalsis auxiliary measurement method and device based on ultrasound technology proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0053] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0054] The specific scheme of the endometrial motility auxiliary measurement method and device based on ultrasound technology provided by the present invention is described in detail below with reference to the accompanying drawings.

[0055] See also Figure 1 , which shows a flow chart of an endometrial motility auxiliary measurement method based on ultrasound technology provided by an embodiment of the present invention, the method comprising the following steps:

[0056] S1: Acquire the endometrial ultrasound image of each sampling frame; in the endometrial ultrasound image of the current sampling frame, determine the peristaltic analysis area based on the edge distribution; based on the continuous matching of each pixel point in the peristaltic analysis area in the endometrial images of historical continuous sampling frames, obtain time-series continuous matching points and form a peristaltic matching point sequence.

[0057] In an embodiment of the present invention, an ultrasonic image of the patient's uterus is obtained through ultrasonic examination. After applying a proper amount of ultrasonic coupling agent on the ultrasonic probe, the probe position is adjusted to perform ultrasonic acquisition of the uterine part. The probe acquires an ultrasonic image of the endometrium including the endometrium. Since peristalsis is a dynamically changing process, ultrasonic images are continuously acquired to analyze the movement. In order to obtain an analyzable peristalsis condition for the current frame, continuous analysis based on historical sampling frames is required. Therefore, in an embodiment of the present invention, the endometrial ultrasonic images of several sampling frames before the current sampling frame are first acquired for preliminary analysis of the change. Please refer to Figure 2 , which shows a schematic diagram of an endometrial ultrasound image provided by an embodiment of the present invention.

[0058] Considering that the endometrium is usually located inside the uterine cavity, when analyzing the current sampling frame, the interference caused by the edge effect can be reduced by narrowing the analysis range. When performing ultrasound scanning, the probe will perform fan-shaped scanning at a certain angle and range. Therefore, the echo-appearing area in the ultrasound image is a fan-shaped area. The range can be determined by fan-shaped contraction of the edge boundary in the image, that is, the peristaltic analysis area of ​​the endometrial ultrasound image of the current sampling frame is determined based on the edge distribution.

[0059] In some possible implementations of the present invention, the size of the edge closest to the boundary of the endometrial ultrasound image of the current sampling frame is used as the fan-shaped image window, and the fan-shaped image window is reduced by a preset ratio through the center point to obtain an analysis window, and the area within the analysis window is used as the peristalsis analysis area of ​​the endometrial peristalsis image of the current sampling frame. The preset ratio can be set to 70%, which can be adjusted by the implementer and is not limited here. Please refer to Figure 3 , which shows a schematic diagram of a peristalsis analysis area provided by an embodiment of the present invention.

[0060] In the peristalsis analysis area, the pixels participate in the peristalsis to a higher degree, and are more valuable for the motion analysis of the peristalsis. Therefore, in order to further analyze the contribution of the pixels in the peristalsis analysis area to the motion situation, the peristaltic change matching points of each pixel are obtained based on the historical sampling frames, and the possible changes of the pixels are preliminarily characterized.

[0061] In some possible implementations of the embodiments of the present invention, for any pixel point in the peristaltic analysis area, a matching point of the pixel point on the previous sampling frame in time sequence will be obtained through a feature matching algorithm, and the matching point of each matching point on the previous sampling frame in time sequence will be continuously obtained through the feature matching algorithm, and all matching points corresponding to the pixel point are arranged in time sequence to form a peristaltic matching point sequence. Through frame-by-frame matching of pixel points on time-series images, the possible historical change distribution of pixel points on the current frame is obtained, so as to facilitate subsequent change analysis through the peristaltic matching point sequence. It should be noted that the feature matching algorithm will generate a descriptor for each point for matching, and the optimal matching result is the matching point. The feature matching algorithm is a technical means well known to those skilled in the art and will not be described in detail here.

[0062] S2: In the peristalsis analysis area, the initial peristalsis contribution of each pixel point is obtained according to the position distribution of each pixel point in the peristalsis analysis area and the significance and period of the temporal distance change of the matching point in the corresponding peristalsis matching point sequence.

[0063] When analyzing the contribution of pixels to the creep changes, more focused observation is required for points with higher fluctuations. The more significant the fluctuations are, and the closer the pixels are to the center of the area, the higher the need for observation. The more information characterizing the creep situation is, the greater the contribution is. Therefore, the distribution of pixels in the creep analysis area and the changes in matching points in the creep matching point sequence are analyzed.

[0064] Preferably, in some possible implementations of the present invention, the method for obtaining the initial peristaltic contribution degree can be found in Figure 4 , which shows a flow chart of a method for obtaining initial peristaltic contribution provided by an embodiment of the present invention, the method comprising the following steps:

[0065] S201: Acquire the center point of the peristalsis analysis area; for any pixel point in the peristalsis analysis area, obtain the near-center degree index of the pixel point based on the distance between the pixel point and the center point.

[0066] First, from the position of the pixel point itself in the peristaltic analysis area, when the pixel point is located in the central part of the peristaltic analysis, the peristaltic change of the pixel point should be retained, and its reliable contribution to the overall peristaltic analysis is higher. Therefore, in the embodiment of the present invention, the center point of the peristaltic analysis area is first determined as the center point of the area.

[0067] Further, for any pixel point in the peristalsis analysis area, the distance between the pixel point and the center point of the area is negatively correlated to obtain the near-center index of the pixel point. The larger the near-center index, the closer the pixel point is to the center point of the area. The pixel point is more representative of the peristalsis performance of the endometrial area, and the subsequent contribution of the pixel point should be greater. In the embodiment of the present invention, negative correlation means that the dependent variable decreases with the increase of the independent variable. The negative correlation mapping can be performed in the form of inverse proportion, which can be adjusted by the implementer without limitation.

[0068] S202: Obtain the matching center point of all matching points in the creeping matching point sequence of the pixel point, and obtain the fluctuation degree index of the pixel point according to the degree of deviation of the distance between each matching point in the creeping matching point sequence of the pixel point and the matching center point.

[0069] Then, we analyze the changes in the historical matching points of the pixel points. Since peristalsis is a dynamic process, points with large fluctuations need to be observed in detail. The greater the fluctuation, the greater the pressure or stronger contraction of the endometrium represented by the pixel position, which needs to be considered more in the subsequent peristalsis analysis.

[0070] In some possible implementation methods of the embodiments of the present invention, the center point of the corresponding positions of all matching points in the creeping matching point sequence of the pixel point on the image is first taken as the matching center point to reflect the average position of the changing distribution of the matching points. It should be noted that the center point can be obtained through the position coordinates of the point. The method for obtaining the center point is a technical means well known to those skilled in the art and will not be elaborated here.

[0071] Furthermore, the distance between each matching point and the matching center point in the creeping matching point sequence of the pixel point is used as the fluctuation distance of each matching point, and the sum of the fluctuation distances of all matching points corresponding to the pixel point is calculated to obtain the fluctuation degree index of the pixel point. The larger the fluctuation degree index of the pixel point, the greater the overall deviation of the distribution of historical matching points, the pixel point has a more significant fluctuation change feature, and the degree to which the pixel point should participate in the creeping analysis later.

[0072] S203: Obtaining a periodicity index of the pixel point according to the regularity of the position changes of all matching points in the creeping matching point sequence of the pixel point in time sequence.

[0073] At the same time, the regularity of the changes must also be considered when analyzing historical changes. During the peristaltic changes of the endometrium, the more irregular the points are, the more likely they are to represent specific physiological changes or abnormalities in the endometrium, and the more they should be analyzed. Therefore, the degree of regularity of the changes in the matching points in time series can be used to reflect the degree to which the pixels participate in the peristaltic analysis.

[0074] In some possible implementations of the embodiments of the present invention, first, based on the positions of all matching points in the peristaltic matching point sequence of the pixel point in the endometrial ultrasound image, a time series scatter plot of the pixel point is obtained, and the time series position change of the matching point is reflected by the scatter plot, which is convenient for subsequent analysis of regularity. The time series scatter plot of the pixel point is further subjected to Fourier transformation to obtain a spectrum diagram, and the spectrum is used to perform time series regularity analysis. It should be noted that the method of Fourier transforming the time series scatter plot is a technical means well known to those skilled in the art, and will not be described in detail here.

[0075] Furthermore, the mean of the signal intensity in the spectrum graph of the pixel point is calculated to obtain the average signal intensity of the pixel point. The difference between the signal intensity corresponding to each peak point in the spectrum graph and the average signal intensity is used as the intensity significance of each peak point. The intensity of the peak point in the spectrum graph reflects the signal change characteristics of the position time series change in the time series scatter diagram. In the spectrum graph, when the signal intensity of the main component is more significant, it means that the change law is more concentrated in the main frequency part and has a regular degree. It should be noted that the acquisition of peak points in the spectrum graph is a technical means well known to those skilled in the art, such as obtaining peak points by the local maximum search method, which will not be elaborated here.

[0076] Finally, the mean of the intensity significance of all peak points is negatively correlated to obtain the periodicity index of the pixel point. The higher the intensity significance of the peak point, the more regular the change of the position of the matching point in the time series is. Therefore, the smaller the periodicity index is, the weaker the pixel point is in terms of periodicity and needs to be paid attention to.

[0077] S204: combining the near-center degree index, the fluctuation degree index and the period degree index of the pixel point to obtain the initial creep contribution of the pixel point.

[0078] Finally, combined with the change analysis of the pixel itself and the historical matching points, the three indicators are combined to reflect the degree of the pixel's participation in the subsequent peristaltic analysis, that is, the initial peristaltic contribution of the pixel. In the embodiment of the present invention, the product of the pixel's proximal index, fluctuation index and periodic index is used as the initial peristaltic contribution of the pixel. The larger the pixel's proximal index, the closer the pixel is to the center part, and the more meaningful the observation and analysis is. The larger the pixel's fluctuation index, the larger the pixel's historical matching point changes, and the more meaningful the observation and analysis is. The larger the pixel's periodic index, the more irregular the position change of the pixel's historical matching point, and the more meaningful the observation and analysis is, so the initial peristaltic contribution will also be larger.

[0079] At this point, the preliminary analysis of the pixels in the peristalsis analysis area is completed, and the initial peristalsis contribution of each pixel in the subsequent peristalsis analysis is obtained.

[0080] S3: In the peristalsis analysis area, according to the consistency of the position change of each pixel point and the adjacent pixel points in the corresponding peristalsis matching point sequence, the final peristalsis contribution of each pixel point is obtained in combination with the initial peristalsis contribution.

[0081] Since the peristalsis and contraction in the uterine tissue are usually relatively uniform, the local tissue cells or structures have similar physiological and functional characteristics, and are also similar in motion state patterns. In order to ensure the coordination of the local tissue in the overall function, the motion state is usually coherent, that is, the changes of the local pixels in the image have a certain degree of consistency. Therefore, each pixel is further analyzed and corrected through the local consistency to obtain a more reliable contribution.

[0082] In some possible implementation methods of the embodiments of the present invention, for any pixel point, other pixel points within the preset neighborhood range of the pixel point are regarded as neighboring pixel points of the pixel point, and the neighboring pixel points are first determined. In a specific embodiment of the embodiments of the present invention, the preset neighborhood range can be set to the eight neighborhood ranges of the pixel point, which is not limited here.

[0083] Furthermore, based on the position deviation difference between the pixel point and each adjacent pixel point and the corresponding matching point in the same sequence in the creeping matching point sequence, the change consistency of each adjacent pixel point is obtained, and the change deviation and consistency in time series are reflected by the difference between the deviation between the pixel point and the matching point and the deviation between the adjacent pixel point in the same sequence matching point.

[0084] In some possible implementations of the present invention, the difference between the pixel point and the matching point of each sequence number in the corresponding peristaltic matching point sequence on the horizontal coordinate is first used as the horizontal deviation of the pixel point at each sequence number, and the difference between the pixel point and the matching point of each sequence number in the corresponding peristaltic matching point sequence on the vertical coordinate is used as the vertical deviation of the pixel point at each sequence number. Among them, the matching point of each sequence number is also the corresponding matching point under each historical sampling frame, and the deviation of the position change under each historical sampling frame is reflected by the deviation of the pixel point and the matching point on the coordinates.

[0085] Furthermore, for any adjacent pixel point of the pixel point, at each serial number, the difference in lateral deviation and the difference in longitudinal deviation between the pixel point and the adjacent pixel point are combined to obtain the deviation difference between the pixel point and the adjacent pixel point at each serial number. In an embodiment of the present invention, the difference in lateral deviation and the difference in longitudinal deviation are accumulated between the pixel point and the adjacent pixel point at each serial number to obtain the deviation difference between the pixel point and the adjacent pixel point at each serial number, reflecting the difference in the change position of the matching point between the pixel point and the adjacent pixel point in each sampling frame in the horizontal and vertical directions.

[0086] Finally, the deviation difference between the pixel point and the adjacent pixel point under all sequence numbers is combined and negative correlation mapping and normalization processing are performed to obtain the consistency of change of the adjacent pixel points. In an embodiment of the present invention, the cumulative value of the deviation difference between the pixel point and the adjacent pixel point under all sequence numbers is negatively correlated and normalized to obtain the consistency of change of the adjacent pixel points. In an embodiment of the present invention, a negative exponential function with a natural constant as the base can be selected for negative correlation mapping and normalization processing. Normalization is a technical means well known to those skilled in the art. The choice of normalization can also be linear normalization or standard normalization, etc., which is not limited here.

[0087] At this point, the analysis of the consistency of changes in adjacent pixels is completed.

[0088] Finally, the final peristaltic contribution of the pixel is obtained by combining the change consistency of all neighboring pixels of the pixel and the initial peristaltic contribution of the pixel.

[0089] The degree of consistency in the changes of adjacent pixels reflects the credibility of the pixel contribution analysis. The higher the consistency of the local changes of the pixel, the smaller the matching point representation error of the pixel in the matching point analysis, and the more reliable the result of the pixel contribution analysis.

[0090] In the embodiment of the present invention, the sum of the change consistency of all adjacent pixels of the pixel is used as the local change adjustment degree of the pixel to comprehensively reflect the local change consistency. The product of the local change adjustment degree of the pixel and the initial peristaltic contribution is normalized to obtain the final peristaltic contribution of the pixel. The larger the local change adjustment degree, the more credible the change analysis of the matching point is, and the higher the final peristaltic contribution is.

[0091] At this point, the analysis and adjustment of the contribution of the pixels in the peristalsis analysis area is completed, and the final peristalsis contribution is obtained.

[0092] S4: Optimizing the PCA dimensionality reduction process of the pixel points in the peristalsis analysis area based on the final peristalsis contribution to obtain optimized dimensionality reduction data of the current sampling frame; and obtaining visualized peristalsis information through the optimized dimensionality reduction data of the current sampling frame.

[0093] When performing auxiliary measurement of uterine peristalsis, redundant information will be removed through PCA dimensionality reduction to perform characterization analysis of the peristaltic movement. During PCA dimensionality reduction, the variance gain can be enhanced based on the contribution situation, so that the information in important pixel points can be retained to a greater extent, thereby improving the accuracy of subsequent peristaltic movement characterization.

[0094] In the embodiment of the present invention, the product of the sample variance of each pixel point in the peristalsis analysis area in PCA dimensionality reduction and the final peristalsis contribution is used as the optimized sample variance of each pixel point to enhance the peristalsis characterization ability of the pixel point. PCA dimensionality reduction is performed based on the optimized sample variance of each pixel point in the peristalsis analysis area to obtain optimized dimensionality reduction data.

[0095] Among them, the sample variance is an information measurement indicator used by PCA. When the variance of a feature is large, it means that this feature carries more information, so the sample variance can also be called the interpretable variance. It should be noted that the PCA dimensionality reduction process and the acquisition of the sample variance are well-known technical means well known to those skilled in the art. Only the sample variance value is weighted and adjusted, which will not be further elaborated here.

[0096] In a specific implementation method of an embodiment of the present invention, the PCA dimensionality reduction process reduces the dimension of each pixel point and the corresponding peristaltic matching point sequence as a dimension, selects the eigenvectors corresponding to the principal components to form a projection matrix, and the projection matrix is ​​used to project the original data set into a low-dimensional space. The projection matrix is ​​used to project the original pixel point data set into the low-dimensional space, and a reduced three-dimensional data set is obtained. As optimized reduced dimensionality data, it can reduce interference and more accurately characterize the motion change characteristics of the peristalsis.

[0097] Visual peristalsis information is obtained by optimizing the dimensionality reduction data of the current sampling frame. The visual peristalsis information can be represented by a visualization curve. By optimizing the dimensionality reduction data, a peristalsis visualization curve reflecting the endometrial peristalsis can be drawn. The peristalsis visualization curve can not only intuitively display the dynamic changes of endometrial peristalsis, but also provide richer perspectives and details, and can more comprehensively understand and analyze the patterns and laws of peristalsis, providing data support for the measurement of endometrial peristalsis.

[0098] In summary, in the peristaltic analysis area corresponding to the ultrasonic image of the current sampling frame, the present invention obtains the time-series continuous matching points of the pixel points to form a peristaltic matching point sequence according to the matching of the pixel points in the historical continuous sampling frames, reflecting the possible peristaltic distribution of the pixel points in the historical sampling frames, thereby further analyzing the degree of peristaltic state characterization of each pixel point according to the continuous peristaltic distribution of each pixel point in history. Therefore, further according to the position distribution of each pixel point and the distance change degree of the continuous matching points, the initial peristaltic contribution of each pixel point is obtained from the critical degree of the position of the pixel point itself to the peristaltic analysis and the significant situation of the pixel point change to the peristaltic analysis, reflecting the influence of the pixel point on the peristaltic analysis, and considering the error of the matching analysis combined with the consistent change of the adjacent pixel points for correction, the final peristaltic contribution is obtained, reflecting the degree of participation of each pixel point in the peristaltic analysis, thereby retaining more features for the pixel points with high contribution. Therefore, the final peristaltic contribution is optimized in the PCA dimensionality reduction process based on the final peristaltic contribution, so that the important part is retained to a greater extent, and the optimized dimensionality reduction data is obtained to obtain more accurate and reliable visual peristaltic information. The present invention characterizes the contribution of peristaltic changes by pixel points in ultrasonic images, adjusts the dimensionality reduction process to remove redundant information, obtains more accurate visualized peristaltic information, and makes subsequent measurement results more accurate and reliable.

[0099] The present invention also provides an endometrial peristalsis auxiliary measurement device based on ultrasound technology, see Figure 5 , which shows a schematic structural diagram of an endometrial peristalsis auxiliary measurement device based on ultrasound technology provided by an embodiment of the present invention, the device comprising:

[0100] A data acquisition unit is used to obtain an endometrial ultrasound image of each sampling frame; in the endometrial ultrasound image of the current sampling frame, determine a peristalsis analysis area according to edge distribution; and according to the continuous matching of each pixel point in the peristalsis analysis area in the endometrial images of historical continuous sampling frames, obtain time-series continuous matching points and form a peristalsis matching point sequence;

[0101] A data processing unit, configured to obtain an initial peristaltic contribution of each pixel point in the peristaltic analysis area according to the position distribution of each pixel point in the peristaltic analysis area and the significance and period of the time series distance change of the matching point in the corresponding peristaltic matching point sequence;

[0102] In the peristalsis analysis area, according to the consistency of the position change of each pixel point and the adjacent pixel points in the corresponding peristalsis matching point sequence, the final peristalsis contribution of each pixel point is obtained in combination with the initial peristalsis contribution;

[0103] The visualization display unit is used to optimize the PCA dimensionality reduction process of the pixel points in the peristalsis analysis area based on the final peristalsis contribution to obtain the optimized dimensionality reduction data of the current sampling frame; and obtain visualized peristalsis information through the optimized dimensionality reduction data of the current sampling frame.

[0104] Since the specific implementation process of the endometrial motility auxiliary measurement device based on ultrasound technology in this embodiment is the same as the specific implementation process of the endometrial motility auxiliary measurement method based on ultrasound technology mentioned above, it will not be described in detail here.

[0105] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0106] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for auxiliary measurement of endometrial motility based on ultrasound technology, It is characterized in that The method comprises: Acquire the endometrial ultrasound image of each sampling frame; determine the peristalsis analysis area according to the edge distribution in the endometrial ultrasound image of the current sampling frame; obtain the temporally continuous matching points and form a peristalsis matching point sequence according to the continuous matching of each pixel point in the peristalsis analysis area in the endometrial images of the historical continuous sampling frames; In the peristalsis analysis area, according to the position distribution of each pixel point in the peristalsis analysis area, and the significance and period of the time series distance change of the matching point in the corresponding peristalsis matching point sequence, the initial peristalsis contribution of each pixel point is obtained; In the peristalsis analysis area, according to the consistency of the position change of each pixel point and the adjacent pixel points in the corresponding peristalsis matching point sequence, the final peristalsis contribution of each pixel point is obtained in combination with the initial peristalsis contribution; The PCA dimensionality reduction process of the pixels in the peristalsis analysis area is optimized based on the final peristalsis contribution to obtain optimized dimensionality reduction data of the current sampling frame; and visualized peristalsis information is obtained through the optimized dimensionality reduction data of the current sampling frame.

2. According to claim 1, a method for auxiliary measurement of endometrial motility based on ultrasound technology, It is characterized in that The method for obtaining the initial peristaltic contribution comprises: Obtaining the center point of the peristalsis analysis area; for any pixel point in the peristalsis analysis area, obtaining the near-center index of the pixel point based on the distance between the pixel point and the center point; Obtain the matching center point of all matching points in the creeping matching point sequence of the pixel point; obtain the fluctuation degree index of the pixel point according to the degree of deviation between each matching point in the creeping matching point sequence of the pixel point and the matching center point; According to the regularity of the position changes of all matching points in the creeping matching point sequence of the pixel point in time sequence, the periodicity index of the pixel point is obtained; The initial creep contribution of the pixel is obtained by combining the centroid index, fluctuation index and periodic index of the pixel.

3. According to claim 2, a method for auxiliary measurement of endometrial motility based on ultrasound technology, It is characterized in that The method for obtaining the volatility index includes: The center point of the corresponding positions of all matching points in the creeping matching point sequence of the pixel point on the image is taken as the matching center point; The distance between each matching point and the matching center point in the creeping matching point sequence of the pixel point is used as the fluctuation distance of each matching point; The sum of the fluctuation distances of all matching points corresponding to the pixel is calculated to obtain the fluctuation degree index of the pixel.

4. According to claim 2, a method for auxiliary measurement of endometrial motility based on ultrasound technology, It is characterized in that The method for obtaining the cycle degree index includes: Based on the positions of all matching points in the peristaltic matching point sequence of the pixel point in the endometrial ultrasound image, a time series scatter plot of the pixel point is obtained; and the time series scatter plot of the pixel point is subjected to Fourier transformation to obtain a spectrum plot; Calculate the mean value of the signal strength in the spectrum of the pixel point to obtain the average signal strength of the pixel point; take the difference between the signal strength corresponding to each peak point in the spectrum and the average signal strength as the strength significance of each peak point; The mean of the intensity significance of all peak points is negatively correlated to obtain the periodicity index of the pixel point.

5. According to claim 1, a method for auxiliary measurement of endometrial motility based on ultrasound technology, It is characterized in that The method for obtaining the final creep contribution comprises: For any pixel point, other pixels within the preset neighborhood of the pixel point are regarded as the neighboring pixels of the pixel point; According to the position deviation difference between the pixel point and each adjacent pixel point and the matching points in the same sequence in the corresponding creeping matching point sequence, the change consistency of each adjacent pixel point is obtained; The final peristaltic contribution of the pixel is obtained by combining the change consistency of all neighboring pixels of the pixel and the initial peristaltic contribution of the pixel.

6. The method for auxiliary measurement of endometrial motility based on ultrasound technology according to claim 5, It is characterized in that The method for obtaining the change consistency includes: The difference between the pixel point and the matching point of each sequence number in the corresponding creeping matching point sequence on the horizontal coordinate is taken as the horizontal deviation of the pixel point at each sequence number; the difference between the pixel point and the matching point of each sequence number in the corresponding creeping matching point sequence on the vertical coordinate is taken as the vertical deviation of the pixel point at each sequence number; For any adjacent pixel of the pixel, under each sequence number, the difference in horizontal deviation and vertical deviation between the pixel and the adjacent pixel are combined to obtain the deviation difference between the pixel and the adjacent pixel under each sequence number; The deviation difference between the pixel and the adjacent pixel under all sequence numbers is combined and negative correlation mapping is performed to obtain the change consistency of the adjacent pixel.

7. According to claim 5, a method for auxiliary measurement of endometrial motility based on ultrasound technology, It is characterized in that The step of combining the change consistency of all adjacent pixels of the pixel and the initial peristaltic contribution of the pixel to obtain the final peristaltic contribution of the pixel includes: The sum of the change consistency of all adjacent pixels of the pixel is used as the local change adjustment degree of the pixel; The product of the local change adjustment degree of the pixel point and the initial peristaltic contribution degree is normalized to obtain the final peristaltic contribution degree of the pixel point.

8. According to claim 1, a method for auxiliary measurement of endometrial motility based on ultrasound technology, It is characterized in that The PCA dimensionality reduction process of optimizing the pixels in the peristalsis analysis area based on the final peristalsis contribution to obtain optimized dimensionality reduction data of the current sampling frame includes: The product of the sample variance of each pixel in the peristalsis analysis area in PCA dimensionality reduction and the final peristalsis contribution is used as the optimized sample variance of each pixel; PCA dimensionality reduction is performed based on the optimized sample variance of each pixel in the creep analysis area to obtain optimized dimensionality reduction data.

9. The method for auxiliary measurement of endometrial motility based on ultrasound technology according to claim 1, It is characterized in that The method of obtaining continuous matching points in time sequence and forming a peristaltic matching point sequence according to the continuous matching of each pixel point in the peristaltic analysis area in the endometrial images of the historical continuous sampling frames includes: For any pixel point in the peristalsis analysis area, the matching point of the pixel point in the previous sampling frame in time sequence will be obtained through the feature matching algorithm, and the matching point of each matching point in the previous sampling frame in time sequence will be obtained continuously through the feature matching algorithm, and all the matching points corresponding to the pixel point will be arranged in time sequence to form a peristalsis matching point sequence.

10. An endometrial motility auxiliary measurement device based on ultrasound technology, It is characterized in that include: A data acquisition unit is used to obtain an endometrial ultrasound image of each sampling frame; in the endometrial ultrasound image of the current sampling frame, determine a peristalsis analysis area according to edge distribution; and according to the continuous matching of each pixel point in the peristalsis analysis area in the endometrial images of historical continuous sampling frames, obtain time-series continuous matching points and form a peristalsis matching point sequence; A data processing unit, configured to obtain an initial peristaltic contribution of each pixel point in the peristaltic analysis area according to the position distribution of each pixel point in the peristaltic analysis area and the significance and period of the time series distance change of the matching point in the corresponding peristaltic matching point sequence; In the peristalsis analysis area, according to the consistency of the position change of each pixel point and the adjacent pixel points in the corresponding peristalsis matching point sequence, the final peristalsis contribution of each pixel point is obtained in combination with the initial peristalsis contribution; A visualization display unit, used for optimizing the PCA dimensionality reduction process of the pixel points in the peristalsis analysis area based on the final peristalsis contribution, and obtaining optimized dimensionality reduction data of the current sampling frame; Visualized peristalsis information is obtained through optimized dimension reduction data of the current sampling frame.

Citation Information

Patent Citations

  • Improved capillary bloodstream imaging method

    CN107862724A

  • Uterus ultrasound micro peristalsis video processing method

    CN108537785A

  • Method for measuring endometrial peristalsis and equipment for measuring endometrial peristalsis

    CN113040822A

  • Algorithm for automatically identifying and quantifying endometrial peristalsis of magnetic resonance image

    CN116468650A

  • Endometrial peristalsis data processing method and ultrasonic equipment

    CN116763352A