Image detection method and system for prolonging second birth process of double-fetal vaginal delivery and SPPH risk based on ultrasonic Speckle-tracing

Abstract

The invention relates to the technical field of ultrasonic image analysis, and discloses an image detection method and system for prolonging the second birth process of double-fetal vaginal delivery and SPPH risk based on ultrasonic Speckle-tracing, and the method comprises the steps: obtaining an ultrasonic two-dimensional gray-scale image sequence; marking a conjoint analysis area between the cervical wall and the front and rear walls of the vagina; constructing a joint area displacement field; constructing a joint region strain trajectory; extracting parturition process characteristic indexes; and executing risk assessment through a fuzzy rule. The technical problem that in the prior art, risk judgment depends on clinical experience, and especially under the condition that continuous stretching occurs on the cervix uteri and vagina combined wall in the second birth process of double-fetal vaginal delivery, early recognition of the evolution process of the tissue from reversible stretching to irreversible instability is difficult to achieve is solved. Due to the fact that joint area displacement field modeling and logarithmic nonlinear strain analysis are introduced, quantitative characterization and continuous risk description of the mechanical behavior of the tissue in the labor process are achieved.

Description

Technical Field

[0001] This invention relates to the field of ultrasound image analysis technology, and in particular to an imaging detection method and system for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking. Background Technology

[0002] Currently, in the clinical management of vaginal delivery of twins, prolonged second stage of labor and the resulting severe postpartum hemorrhage (SPPH) remain significant risk factors affecting maternal and infant safety. Existing techniques for assessing the risk of second stage labor primarily rely on labor duration thresholds, maternal medical history, fetal weight estimation, and clinical experience, supplemented by routine ultrasound observation of fetal head position or cervical morphology. However, these methods mostly focus on macroscopic time indicators or static morphological characteristics, lacking effective quantitative means for the continuous stretching, large displacement, and nonlinear deformation experienced by soft tissues during delivery. In particular, they fail to reflect the dynamic process of key tissues such as the cervix and vaginal wall gradually evolving from reversible stretching to structural instability in the latter part of the second stage of labor.

[0003] For example, in the case of vaginal delivery of twins, the continuous pressure and repeated traction of the presenting part of the fetus will keep the cervical wall and the anterior and posterior vaginal walls in a state of high strain for a long time. However, conventional two-dimensional ultrasound can only provide grayscale morphological information and cannot quantitatively describe the displacement relationship of speckles in the tissue over time, nor can it distinguish the difference between short-term recoverable deformation and long-term cumulative damage. As a result, the existing technology cannot fully meet the need for early, objective and continuous assessment of SPPH risk under the condition of prolonged second stage of labor.

[0004] Therefore, there is an urgent need for a method that can still quantitatively analyze the mechanical behavior of the cervix-vaginal junction region based on ultrasound imaging in the second stage of labor when there are large displacements, nonlinear deformations, and significant individual differences, and further realize dynamic detection and grading assessment of SPPH risk, so as to improve the accuracy of risk identification in the second stage of labor and the ability to support clinical decision-making. Summary of the Invention

[0005] To address the aforementioned technical shortcomings, the purpose of this invention is to propose an imaging detection method based on ultrasound speckle-tracking for the prolongation of the second stage of labor and SPPH risk in twin vaginal deliveries. This method aims to solve the technical problem of relying on clinical experience for risk assessment in existing technologies, especially in cases where continuous traction occurs in the cervix and vaginal commissural wall during the second stage of twin vaginal deliveries, making it difficult to achieve early identification of the evolution of tissue from reversible traction to irreversible instability.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides an imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking.

[0007] The imaging method for detecting prolonged second stage of labor and SPPH risk in twin vaginal deliveries based on ultrasound speckle-tracking includes:

[0008] Step S10: Obtain the transvaginal or transperineal two-dimensional grayscale ultrasound image sequence F during the second stage of labor for twin vaginal delivery. Based on the ultrasound two-dimensional grayscale image sequence F, a multi-layer semantic mask generation mechanism based on structural continuity and region displacement prior is used to perform joint analysis region annotation task, and output the joint analysis region mask between the cervical wall and the anterior and posterior vaginal walls. ;

[0009] Step S20: Based on the joint analysis region mask, a texture sub-block matching mechanism based on multi-scale pyramid and singular value decomposition is used to perform the Speckle pixel block tracking task, and the joint region displacement field is output. ;

[0010] Step S30: Based on the joint region displacement field A logarithmic strain modeling and strain trajectory accumulation extraction mechanism is used to perform finite nonlinear strain calculation tasks, outputting the joint region strain trajectory that varies with time. ;

[0011] Step S40: Based on the joint regional strain trajectory The strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification is used to perform the second stage feature extraction task, and the joint strain accumulation index is output. Duration of monotonic strain growth Total duration of the second stage of labor ;

[0012] Step S50: Based on the joint strain accumulation index Duration of monotonic strain growth Total duration of the second stage of labor Construct the SPPH risk scoring function and output the image risk detection results based on the SPPH risk scoring function.

[0013] Preferably, in step S10, a two-dimensional grayscale ultrasound image sequence F is acquired during the second stage of labor for twin vaginal delivery, either transvaginally or transperineally. Based on the two-dimensional grayscale ultrasound image sequence F, a multi-layer semantic mask generation mechanism based on structural continuity and region displacement prior is used to perform a joint analysis region annotation task, outputting a joint analysis region mask between the cervical wall and the anterior and posterior vaginal walls. The steps specifically include:

[0014] Step S101: Obtain a sequence of two-dimensional grayscale ultrasound images F taken transvaginally or transperineally during the second stage of labor for twin vaginal delivery. Calculate the grayscale distribution similarity between the previous and current frames in the sequence F to obtain the grayscale stability index of the candidate tissue region in the time dimension. Where x is the x-coordinate of the current pixel. The vertical coordinate of the current pixel is denoted as y; the candidate tissue regions include the cervical wall, the anterior vaginal wall, and the posterior vaginal wall.

[0015] Step S102: When the grayscale stability index is within the preset detection period T The grayscale value remains consistently below the preset grayscale stability threshold. At that time, determine the current pixel point It belongs to the high dynamic range region; otherwise, determine the current pixel. It belongs to the structurally stable region; the position of the current pixel corresponding to the initial frame within the preset detection period T is set by an empirical template; a mask sliding window W is introduced, and the spatiotemporal sliding window cumulative averaging method is used for tracking based on the mask sliding window W and the position of the current pixel corresponding to the initial frame, and finally the complete high dynamic region and the complete structurally stable region are output.

[0016] Step S103: The complete high-dynamic region is used as the first layer, and the complete structurally stable region is used as the second layer. A region union fusion is performed on the first and second layers to obtain the joint analysis region. Then, a boundary refinement algorithm based on edge gradients is used to perform Canny edge extraction and morphological closing operations on the final joint analysis region, ultimately outputting a mask of the joint analysis region between the cervical wall and the anterior and posterior vaginal walls. .

[0017] Preferably, in step S20, a texture sub-block matching mechanism based on multi-scale pyramids and singular value decomposition is used to perform the Speckle pixel block tracking task based on the joint analysis region mask, and the joint region displacement field is output. The steps specifically include:

[0018] Step S201: Image pyramid construction stage: Obtain the joint analysis region image corresponding to the joint analysis region mask, and for the k-th frame joint analysis region image of adjacent frames... Joint analysis region image of frame k+1 Joint analysis of the region image with the (k+2)th frame A three-layer Gaussian image pyramid is constructed, with the image resolution decreasing sequentially in each layer of the three-layer Gaussian image pyramid, in order to achieve stepped multi-scale tracking and matching;

[0019] Step S202: Singular Value Decomposition Guided Stage: Singular value decomposition filtering is applied to each layer of the 3-layer Gaussian image pyramid to remove mean drift points and output an optimized Gaussian image pyramid.

[0020] Step S203: Speckle pixel block selection stage: Preset pixel width M, in the optimized Gaussian image pyramid, every... The pixel selection tracking sub-block is used as the Speckle pixel block; the matching strength is calculated for the Speckle pixel block using the normalized cross-correlation coefficient analysis method. Eliminate matching strength Below the preset matching strength threshold After identifying the Speckle pixel blocks, output the Speckle pixel block set;

[0021] Step S204: Based on the Speckle pixel block set, a local median filtering and spatial bilateral constraint fusion mechanism are used to perform displacement smoothing and jump removal processing, and output the joint region displacement field. .

[0022] Preferably, in step S20, displacement smoothing and jump removal are performed based on the Speckle pixel block set using a local median filtering and spatial bilateral constraint fusion mechanism, and the joint regional displacement field is output. The steps specifically include:

[0023] Step S201: Perform local median filtering on the displacement of the Speckle pixel blocks within a 3×3 neighborhood of the Speckle pixel block set to form a median smooth displacement. Where x is the x-coordinate of the current pixel. This represents the ordinate of the current pixel; simultaneously, it calculates the mean and standard deviation of the displacement of the Speckle pixel block within the 3×3 neighborhood.

[0024] Step S202: Execute spatial bilateral constraints: smooth displacement at the median When the deviation from the mean is greater than twice the standard deviation, the median is smoothed out. If an abnormal displacement is detected, boundary interpolation is performed at the location of the abnormal displacement; otherwise, the original speckle pixel block displacement is preserved; the final output is the fused displacement. ;

[0025] Step S203: Calculate the output blended displacement Reorganized into a complete joint region displacement field Output the results.

[0026] Preferably, in step S30, based on the joint regional displacement field A logarithmic strain modeling and strain trajectory accumulation extraction mechanism is used to perform finite nonlinear strain calculation tasks, outputting the joint region strain trajectory that varies with time. The steps specifically include:

[0027] Step S301: Pre-set a reference length Simultaneously, from the joint region displacement field The Euclidean distance between Speckle pixel blocks is obtained as the tissue stretching length of the current frame. ;

[0028] Step S302: Based on reference length and the current frame organization stretch length Inter-frame strain values ​​in the joint region were constructed using a nonlinear logarithmic strain model. , ; and integrate all tissue tensile lengths to output the combined regional strain trajectory as a function of time. Among them, the nonlinear logarithmic strain model is used to adapt to the large deformation scenario under vaginal tension.

[0029] Preferably, in step S40, based on the joint regional strain trajectory The strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification is used to perform the second stage feature extraction task, and the joint strain accumulation index is output. Duration of monotonic strain growth Total duration of the second stage of labor The steps specifically include:

[0030] Step S401: Based on the joint regional strain trajectory The strain rate of change is calculated using the inter-frame difference method. strain rate of change Used to represent the rate of evolution of organizational structure deformation;

[0031] Step S402: Introduce the cumulative deformation period Based on cumulative deformation period and strain rate of change The combined strain cumulative index is calculated using the absolute value time integration method. Combined strain cumulative index Used to indicate the active stress state of tissue structure;

[0032] Step S403: Set the consecutive frame count threshold Continuously monitor the rate of change of strain When the following conditions are met: continuous one cumulative deformation cycle internal strain rate of change All are greater than the preset minimum instability threshold. At that time, determine that the current situation is in the strain instability trend zone and output the duration of monotonous strain growth. , ;

[0033] Step S404: Obtain the total duration of the second stage of labor The final output is the joint strain cumulative index. Duration of monotonic strain growth Total duration of the second stage of labor .

[0034] Preferably, in step S50, the combined strain accumulation index is used. Duration of monotonic strain growth Total duration of the second stage of labor The steps for constructing the SPPH risk scoring function and outputting image risk detection results based on the SPPH risk scoring function include:

[0035] Step S501: Define the first fuzzy rule set, the second fuzzy rule set, and the third fuzzy rule set, and combine the joint strain cumulative index. Duration of monotonic strain growth Total duration of the second stage of labor The input membership triples are generated by mapping them to the first fuzzy rule set, the second fuzzy rule set, and the third fuzzy rule set, respectively.

[0036] Step S502: Based on the preset fuzzy rule base R, perform fuzzy inference on the input membership triples and output a fuzzy output set. ;

[0037] Step S503: Set the fuzzy output The input is fed into a preset deblurring function, which uses the centroid method to blur the output set. Perform numerical processing of risk levels and output continuous risk score values. , Among them, continuous risk score Used to characterize the intensity of labor risk reflected in the current target image segment, when When the time is right, it indicates a low risk during labor; when When the time is right, it indicates a moderate risk during labor; when At this time, it indicates a high risk during labor;

[0038] Step S504: Transfer the continuous risk score value The visualization is overlaid onto the original two-dimensional grayscale ultrasound image sequence, and the final output is the image risk detection result.

[0039] This invention also provides an imaging detection system based on ultrasound speckle-tracking for prolonged second stage of labor and SPPH risk in vaginal delivery of twins, comprising:

[0040] The joint analysis region annotation module is used to acquire a sequence of two-dimensional grayscale ultrasound images F, taken transvaginally or transperineally, during the second stage of labor in twin deliveries. Based on the ultrasound two-dimensional grayscale image sequence F, a multi-layer semantic mask generation mechanism based on structural continuity and region displacement priors is used to perform the joint analysis region annotation task, outputting a joint analysis region mask between the cervical wall and the anterior and posterior vaginal walls. ;

[0041] The displacement field construction module is used to perform Speckle pixel block tracking tasks based on a texture sub-block matching mechanism using multi-scale pyramids and singular value decomposition, based on a joint analysis region mask, and outputs the joint region displacement field. ;

[0042] The nonlinear strain trajectory calculation module is used for calculations based on the joint region displacement field. A logarithmic strain modeling and strain trajectory accumulation extraction mechanism is used to perform finite nonlinear strain calculation tasks, outputting the joint region strain trajectory that varies with time. ;

[0043] The strain evolution feature extraction module is used to extract strain evolution features based on the joint region strain trajectory. The strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification is used to perform the second stage feature extraction task, and the joint strain accumulation index is output. Duration of monotonic strain growth Total duration of the second stage of labor ;

[0044] The risk assessment and output module is used for assessment based on the joint strain cumulative index. Duration of monotonic strain growth Total duration of the second stage of labor Construct the SPPH risk scoring function and output the image risk detection results based on the SPPH risk scoring function.

[0045] The present invention also provides an imaging detection device for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking, comprising: a memory, a processor, and an imaging detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking stored in the memory and executable on the processor. When the imaging detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking is executed by the processor, an imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking is implemented.

[0046] The present invention also provides a computer program product, including an image detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking. When the image detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking is executed by a processor, it realizes an image detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking.

[0047] The beneficial effects of this invention are as follows: This invention constructs the displacement field and nonlinear strain trajectory of the joint analysis area of ​​the cervical wall and the anterior and posterior vaginal walls through ultrasound speckle-tracking, and further extracts the strain accumulation index and the duration of monotonic strain growth. From the perspective of tissue mechanics evolution, it describes the process of soft tissue transformation from reversible stretching to structural instability, thereby enabling early identification of high-risk conditions before the occurrence of severe postpartum hemorrhage, and significantly improving the foresight and sensitivity of risk assessment.

[0048] This invention integrates the cumulative behavior of combined strain, continuous stretch characteristics, and the duration of the second stage of labor into a model, and introduces a risk assessment mechanism based on fuzzy rule reasoning and defuzzification quantification. This effectively avoids the incompatibility of traditional linear weighting or single-threshold judgment methods with individual differences and nonlinear deformation, making the risk scoring results continuous and interpretable, reducing reliance on subjective experience, and improving the stability, repeatability, and clinical decision support value of SPPH risk detection results in the context of twin vaginal delivery. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a flowchart illustrating the first embodiment of an imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking according to the present invention.

[0051] Figure 2 This is a schematic diagram of the unconstrained joint analysis region displacement field results of the first embodiment of the imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking according to the present invention.

[0052] Figure 3 This is a schematic diagram of the constrained joint analysis region displacement field results of the first embodiment of the imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking according to the present invention.

[0053] Figure 4 This is a frame-by-frame schematic diagram of the combined region logarithmic strain trajectory in a first embodiment of an imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking, according to the present invention.

[0054] Figure 5 This is a schematic diagram of the equipment used in the present invention for imaging detection of prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Example 1: As Figure 1 The diagram shown is a flowchart of the first embodiment of the imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking according to the present invention. The first embodiment of the imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking according to the present invention is presented.

[0057] In the first embodiment, the imaging detection method for prolonged second stage of labor and SPPH risk in vaginal twin deliveries based on ultrasound speckle-tracking includes:

[0058] Step S10: Obtain the transvaginal or transperineal two-dimensional grayscale ultrasound image sequence F during the second stage of labor for twin vaginal delivery. Based on the ultrasound two-dimensional grayscale image sequence F, a multi-layer semantic mask generation mechanism based on structural continuity and region displacement prior is used to perform joint analysis region annotation task, and output the joint analysis region mask between the cervical wall and the anterior and posterior vaginal walls. ;

[0059] It should be noted that the "multi-layer semantic mask generation mechanism based on structural continuity and regional displacement prior" in this step refers to: in the ultrasound two-dimensional grayscale image sequence, combining the continuous distribution characteristics of the cervical wall, anterior vaginal wall and posterior vaginal wall in anatomical structure, and the overall displacement trend of the joint analysis region between adjacent image frames, multi-level semantic modeling of the target analysis region is performed and a corresponding mask is generated; wherein, the multi-layer semantic includes a structural layer for describing the contour of the cervical wall, a tissue layer for describing the boundary of the anterior vaginal wall and posterior vaginal wall, and a joint region layer for describing the continuous transition relationship between the above tissues; the regional displacement prior includes predictive constraint information formed by the overall translation direction, displacement amplitude and relative spatial position change of the joint analysis region in historical frames.

[0060] It is understandable that by introducing structural continuity constraints and regional displacement priors during the joint analysis region labeling stage, this invention can maintain a stable and consistent spatial range of the joint analysis region between the cervical wall and the anterior and posterior vaginal walls in continuous ultrasound image sequences, in the application scenario of the second stage of vaginal delivery of twins, where tissue morphology changes frequently and stress states continue to evolve. This provides an anatomically significant and temporally continuous analysis region basis for subsequent Speckle pixel block tracking, which is beneficial to improving the reliability of subsequent displacement field construction and strain evolution analysis results.

[0061] It should be understood that, compared with traditional region annotation methods based on single-frame threshold segmentation, manual selection of regions of interest, or fixed template matching, this invention avoids the problems of analysis region drift, boundary fragmentation, or semantic inconsistency caused by slight probe movement, continuous pressure from the presenting fetus, or nonlinear deformation of local tissues through the joint constraint of multi-layer semantic masks and region displacement priors. This makes the annotation results of the joint analysis region no longer dependent on a single grayscale feature or manual experience judgment.

[0062] Step S20: Based on the joint analysis region mask, a texture sub-block matching mechanism based on multi-scale pyramid and singular value decomposition is used to perform the Speckle pixel block tracking task, and the joint region displacement field is output. ;

[0063] It should be noted that the "texture sub-block matching mechanism based on multi-scale pyramid and singular value decomposition" in this step refers to: within the range of the joint analysis region mask output in step S10, constructing multi-scale image pyramids for the joint region images of the current frame and adjacent frames respectively, and performing candidate matching search for Speckle texture sub-blocks with a preset window at each scale; wherein, the multi-scale image pyramid is used to converge displacement estimation from coarse to fine, reducing the probability of matching mismatch in large displacement tension scenarios; singular value decomposition is used to perform low-rank representation of the local grayscale texture matrix of each Speckle texture sub-block, extract its main structural components and suppress noise and local speckle randomness, thereby forming a more stable sub-block similarity measurement basis; the output of the Speckle pixel block tracking task is a set of two-dimensional displacement vectors of each pixel or each tracked block in the joint analysis region between adjacent frames, which is further reorganized to obtain the joint region displacement field.

[0064] It should be understood that, compared to traditional optical flow methods or single-scale template matching methods, which are easily affected by speckle randomness, local brightness fluctuations, and probe micro-jitter in ultrasound speckle scenarios, resulting in displacement noise or mismatches, this invention introduces the hierarchical constraints of multi-scale pyramids and the low-rank stable expression of singular value decomposition during the matching process. This makes the similarity calculation focus more on the main structural components of texture sub-blocks rather than instantaneous noise components, thereby reducing the risk of displacement jumps caused by local speckle rearrangement and echo intensity changes. At the same time, this step uses the pre-sequential joint analysis region mask as a spatial constraint to prevent the tracking algorithm from spreading to other non-target tissue areas in the pelvic floor, further improving the anatomical consistency and repeatability of the displacement field.

[0065] For example, such as Figure 2 As shown, the displacement field constructed within the joint analysis region using the traditional Speckle pixel block tracking method is illustrated. Due to the lack of constraints on the multi-scale stability and principal components of the texture sub-blocks, the displacement field within the joint analysis region exhibits significant directional dispersion, abrupt amplitude changes, and discontinuities in local areas. This is particularly evident at the junction of the cervical wall and the anterior and posterior vaginal walls, and in areas of rapid tissue stress change, where local displacement jumps and anomalous vectors are prone to occur, resulting in poor spatial consistency of the overall displacement field. Figure 3 As shown, this study demonstrates how a multi-scale image pyramid structure is introduced to perform hierarchical matching of Speckle textures at different spatial scales. Singular value decomposition is used at each scale layer to extract the dominant structural components of the texture sub-blocks, effectively suppressing erroneous matching caused by ultrasonic speckle noise, local grayscale fluctuations, and transient tissue nonlinear deformation. The resulting joint regional displacement field exhibits continuous, smooth, and directionally consistent variation characteristics in space, with good synergy and anatomical consistency between adjacent regions in the displacement vectors.

[0066] Step S30: Based on the joint region displacement field A logarithmic strain modeling and strain trajectory accumulation extraction mechanism is used to perform finite nonlinear strain calculation tasks, outputting the joint region strain trajectory that varies with time. ;

[0067] It should be noted that the "logarithmic strain modeling" in this step refers to the large displacement and nonlinear deformation characteristics of the cervical wall and the anterior and posterior vaginal walls under the continuous alternating action of stretching, compression, and release during the second stage of vaginal delivery of twins. Instead of using the linear strain model under the traditional small deformation assumption, a logarithmic strain form is introduced to describe the tissue stretching process based on the local reference length within the joint region. The logarithmic strain is used to characterize the relative length change ratio of the tissue at different time points relative to the initial state, so as to reduce the non-physical error of strain estimation amplified by the displacement amplitude under large deformation conditions.

[0068] Understandably, in the latter part of the second stage of labor, the cervical-vaginal junction often undergoes a complex process of multiple rounds of stretching, partial retraction, and re-stretching. Calculating instantaneous strain values ​​at only a single time point is insufficient to reflect the cumulative load on the tissue structure over time. This invention, by calculating logarithmic strain frame-by-frame on a time series and constructing a continuous strain trajectory, allows the deformation evolution of the junction region to be described as a time function, thus providing a quantitative basis for identifying "short-term reversible stretching" and "continuous irreversible structural evolution."

[0069] For example, such as Figure 4 As shown, the logarithmic strain trajectory of the combined region is calculated at each time point during the second stage of labor. Each step corresponds to the strain increment calculated based on the displacement field of the combined region between two adjacent ultrasound frames. It can be observed that in the early stage of the second stage of labor, although the logarithmic strain trajectory of the combined region fluctuates with time, it generally remains within a relatively stable range, and exhibits a downward trend in some time periods, indicating that the cervical-vaginal combined region tissue remains in a reversible stretching state under alternating stretching and release. As the second stage of labor progresses, the step height of the logarithmic strain trajectory continuously increases with time in subsequent time periods, and no longer shows a significant downward trend, forming a continuously monotonically increasing step characteristic, reflecting that the combined region tissue has entered an irreversible structural evolution stage under continuous stretching.

[0070] Step S40: Based on the joint regional strain trajectory The strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification is used to perform the second stage feature extraction task, and the joint strain accumulation index is output. Duration of monotonic strain growth Total duration of the second stage of labor ;

[0071] It should be noted that the "strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification" in this step refers to: based on the time series of the strain trajectory of the joint region, firstly, the strain change between adjacent time points is differentially calculated to obtain the strain rate of change of the joint region; then, within the time range of the second production process, the strain rate of change is integrated over time to construct a cumulative strain index borne by the joint region throughout the second production process, which is used to characterize the overall deformation load level of the tissue in the time dimension.

[0072] Understandably, during the second stage of vaginal delivery of twins, relying solely on the maximum strain value or instantaneous strain peak is insufficient to distinguish whether tissue is in a state of short-term stretching or continuous structural evolution. This invention, by introducing the integral of the strain rate of change, can accumulate the deformation effects generated during multiple stretching-retraction processes over time, thereby reflecting the overall trend of long-term stress on the tissue. Furthermore, by identifying monotonically increasing strain segments, it can further characterize whether the tissue is continuously subjected to irreversible structural stretching within a certain time interval.

[0073] Step S50: Based on the joint strain accumulation index Duration of monotonic strain growth Total duration of the second stage of labor Construct the SPPH risk scoring function and output the image risk detection results based on the SPPH risk scoring function.

[0074] It should be noted that the "SPPH risk scoring function" in this step is not constructed based on a simple linear weighted summation of the various features, but rather on a segmented determination and evolutionary consistency constraint mechanism of the strain evolution state in the joint region. Specifically, the risk scoring function uses the joint strain accumulation index to reflect the overall deformation load level borne by the organization during the second stage, the duration of monotonic strain growth to reflect the length of time the organization in the joint region is in a state of sustained structural stretch, and the total duration of the second stage as a temporal background constraint on the above strain evolution results, thus forming a physiologically meaningful multidimensional risk assessment input. In the process of constructing the risk scoring function, the strain evolution state of the joint region is first divided into stages according to the proportion of the duration of monotonic strain growth in the total duration of the second stage, to distinguish between "short-term stretch-dominated state" and "sustained structural evolution-dominated state"; subsequently, only under the premise of satisfying the condition of sustained structural evolution dominance, the joint strain accumulation index is further introduced to refine the risk level, thereby avoiding over-risk assessment when the second stage is long but the organizational strain has not accumulated continuously.

[0075] Understandably, this invention, by introducing the gating condition of "the existence of a sustained monotonic growth phase," shifts the SPPH risk score away from solely relying on the duration of the second stage of labor or the absolute value of accumulated strain. Instead, it emphasizes the consistency between the strain evolution pattern and the time dimension. This mechanism effectively distinguishes between situations where "labor is prolonged but tissues still have the capacity to recover" and high-risk situations where "labor is prolonged and tissues have entered a phase of sustained structural evolution," thereby making the risk assessment results more consistent with actual clinical physiological processes.

[0076] It should be understood that, compared to traditional methods that assess risk based on labor duration statistics, blood loss thresholds, or single imaging indicators, this invention, for the first time, uses the evolutionary behavior of combined regional tissue strain as the core criterion. Through the synergistic constraint of strain accumulation and the duration of monotonic growth, it achieves a shift from "outcome-based statistical risk assessment" to "process-based evolutionary risk identification." This approach can identify high-risk trends in SPPH at the imaging level in advance, rather than making a post-hoc assessment only after a bleeding event occurs. For example, in a case of vaginal delivery of twins, although the total duration of the second stage of labor was relatively long, the duration of the monotonic growth segment in the combined regional strain trajectory was short, the strain accumulation index was at a low level, and the risk scoring function output was a low-risk state. In another case, the combined regional strain trajectory showed a long period of continuous monotonic growth in the latter part of the second stage of labor, the strain accumulation index increased significantly, and the risk scoring function output was a high-risk state. Even if the second stage of labor had not yet exceeded the abnormal duration threshold in the traditional statistical sense, an early warning of imaging risk could still be given.

[0077] Example 2: Furthermore, the present invention provides an imaging detection system for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking. This system employs an imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking as described in the above embodiments, thus solving the technical problem of imaging detection of prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking. The beneficial effects of the imaging detection system for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking provided by the present invention are the same as those of the imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking provided in the above embodiments. Other technical features of the imaging detection system for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0078] Example 3: This invention provides an imaging detection device based on ultrasound speckle-tracking for prolonged second stage of labor and SPPH risk in vaginal delivery of twins. Please refer to... Figure 5An imaging detection device for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking described in Embodiment 1 above. The imaging detection device for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking in this embodiment of the invention may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The imaging detection device for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the invention. An ultrasound speckle-tracking-based imaging detection device for prolonged second stage of labor and SPPH risk in vaginal twin deliveries may include a processing unit 1001 (e.g., a central processing unit, graphics processor, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory 1002 or a program loaded from a storage device 1003 into a random access memory 1004. The random access memory 1004 also stores various programs and data required for the operation of the ultrasound speckle-tracking-based imaging detection device for prolonged second stage of labor and SPPH risk in vaginal twin deliveries. The processing unit 1001, the read-only memory 1002, and the random access memory 1004 are interconnected via a bus 1005. An I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, liquid crystal display (LCD), speaker, vibrator, etc.; storage devices 1003 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009.Communication device 1009 allows an ultrasound speckle-tracking-based imaging device for detecting prolonged second stage of labor and SPPH risk in vaginal delivery of twins to wirelessly or via wired communication with other devices to exchange data. While the figure illustrates an ultrasound speckle-tracking-based imaging device for detecting prolonged second stage of labor and SPPH risk in vaginal delivery of twins with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0079] Example 4: This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the above-described imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking. The computer program product provided by this invention can solve the technical problem of imaging detection for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking. Compared with the prior art, the beneficial effects of the computer program product provided by this invention are the same as those of the imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking provided in the above-described embodiment, and will not be repeated here.

[0080] In particular, according to the embodiments disclosed in this invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from read-only memory 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this invention.

[0081] It should be understood that the various parts disclosed in this invention can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0082] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. An imaging method for detecting prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking, characterized in that, The methods include: Step S10: Obtain the transvaginal or transperineal two-dimensional grayscale ultrasound image sequence F during the second stage of labor for twin vaginal delivery. Based on the ultrasound two-dimensional grayscale image sequence F, a multi-layer semantic mask generation mechanism based on structural continuity and region displacement prior is used to perform joint analysis region annotation task, and output the joint analysis region mask between the cervical wall and the anterior and posterior vaginal walls. ; Step S20: Based on the joint analysis region mask, a texture sub-block matching mechanism based on multi-scale pyramid and singular value decomposition is used to perform the Speckle pixel block tracking task, and the joint region displacement field is output. ; Step S30: Based on the joint region displacement field A logarithmic strain modeling and strain trajectory accumulation extraction mechanism is used to perform finite nonlinear strain calculation tasks, outputting the joint region strain trajectory that varies with time. ; Step S40: Based on the joint regional strain trajectory The strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification is used to perform the second stage feature extraction task, and the joint strain accumulation index is output. Duration of monotonic strain growth Total duration of the second stage of labor ; Step S50: Based on the joint strain accumulation index Duration of monotonic strain growth Total duration of the second stage of labor Construct the SPPH risk scoring function and output the image risk detection results based on the SPPH risk scoring function.

2. The imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking as described in claim 1, characterized in that, In step S10, a two-dimensional grayscale ultrasound image sequence F is acquired during the second stage of labor for twin vaginal delivery, either transvaginally or transperineally. Based on the ultrasound two-dimensional grayscale image sequence F, a multi-layer semantic mask generation mechanism based on structural continuity and region displacement priors is used to perform a joint analysis region annotation task, outputting a joint analysis region mask between the cervical wall and the anterior and posterior vaginal walls. The steps specifically include: Step S101: Obtain a sequence of two-dimensional grayscale ultrasound images F taken transvaginally or transperineally during the second stage of labor for twin vaginal delivery. Calculate the grayscale distribution similarity between the previous and current frames in the sequence F to obtain the grayscale stability index of the candidate tissue region in the time dimension. Where x is the x-coordinate of the current pixel. The vertical coordinate of the current pixel is denoted as y; the candidate tissue regions include the cervical wall, the anterior vaginal wall, and the posterior vaginal wall. Step S102: When the grayscale stability index is within the preset detection period T The grayscale value remains consistently below the preset grayscale stability threshold. At that time, determine the current pixel point It belongs to the high dynamic range region; otherwise, determine the current pixel. It belongs to the structurally stable region; the current pixel position corresponding to the initial frame within the preset detection period T is set by an empirical template; a mask sliding window W is introduced, and the spatiotemporal sliding window cumulative averaging method is used for tracking based on the mask sliding window W and the current pixel position corresponding to the initial frame, and finally the complete high dynamic region and the complete structurally stable region are output; Step S103: The complete high-dynamic region is used as the first layer, and the complete structurally stable region is used as the second layer. A region union fusion is performed on the first and second layers to obtain the joint analysis region. Then, a boundary refinement algorithm based on edge gradients is used to perform Canny edge extraction and morphological closing operations on the final joint analysis region, ultimately outputting a mask of the joint analysis region between the cervical wall and the anterior and posterior vaginal walls. .

3. The imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking as described in claim 1, characterized in that, In step S20, based on the joint analysis region mask, a texture sub-block matching mechanism based on multi-scale pyramids and singular value decomposition is used to perform the Speckle pixel block tracking task, and the joint region displacement field is output. The steps specifically include: Step S201: Image pyramid construction stage: Obtain the joint analysis region image corresponding to the joint analysis region mask, and for the k-th frame joint analysis region image of adjacent frames... Joint analysis region image of frame k+1 Joint analysis of the region image with the (k+2)th frame A three-layer Gaussian image pyramid is constructed, with the image resolution decreasing sequentially in each layer of the three-layer Gaussian image pyramid, in order to achieve stepped multi-scale tracking and matching; Step S202: Singular Value Decomposition Guided Stage: Singular value decomposition filtering is applied to each layer of the 3-layer Gaussian image pyramid to remove mean drift points and output an optimized Gaussian image pyramid. Step S203: Speckle pixel block selection stage: Preset pixel width M, in the optimized Gaussian image pyramid, every The pixel selection process uses tracking sub-blocks as Speckle pixel blocks; the matching strength is calculated for the Speckle pixel blocks using normalized cross-correlation coefficient analysis. Eliminate matching strength Below the preset matching strength threshold After identifying the Speckle pixel blocks, output the Speckle pixel block set; Step S204: Based on the Speckle pixel block set, a local median filtering and spatial bilateral constraint fusion mechanism are used to perform displacement smoothing and jump removal processing, and output the joint region displacement field. .

4. The imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking as described in claim 3, characterized in that, In step S20, based on the Speckle pixel block set, a local median filtering and spatial bilateral constraint fusion mechanism are used to perform displacement smoothing and jump removal processing, and the joint region displacement field is output. The steps specifically include: Step S201: Perform local median filtering on the displacement of the Speckle pixel blocks within a 3×3 neighborhood of the Speckle pixel block set to form a median smooth displacement. Where x is the x-coordinate of the current pixel. This represents the ordinate of the current pixel; simultaneously, it calculates the mean and standard deviation of the displacement of the Speckle pixel block within the 3×3 neighborhood. Step S202: Execute spatial bilateral constraints: smooth displacement at the median When the deviation from the mean is greater than twice the standard deviation, the median is smoothed out. If an abnormal displacement is detected, boundary interpolation is performed at the location of the abnormal displacement; otherwise, the original speckle pixel block displacement is preserved; the final output is the fused displacement. ; Step S203: Calculate the output blended displacement Reorganized into a complete joint region displacement field Output the results.

5. The imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking as described in claim 3, characterized in that, In step S30, based on the joint region displacement field A logarithmic strain modeling and strain trajectory accumulation extraction mechanism is used to perform finite nonlinear strain calculation tasks, outputting the joint region strain trajectory that varies with time. The steps specifically include: Step S301: Pre-set a reference length Simultaneously, from the joint region displacement field The Euclidean distance between Speckle pixel blocks is obtained as the tissue stretching length of the current frame. ; Step S302: Based on reference length and the current frame organization stretch length Inter-frame strain values ​​in the joint region were constructed using a nonlinear logarithmic strain model. , ; and integrate all tissue tensile lengths to output the combined regional strain trajectory as a function of time. Among them, the nonlinear logarithmic strain model is used to adapt to the large deformation scenario under vaginal tension.

6. The imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking as described in claim 1, characterized in that, In step S40, based on the joint regional strain trajectory The strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification is used to perform the second stage feature extraction task, and the joint strain accumulation index is output. Duration of monotonic strain growth Total duration of the second stage of labor The steps specifically include: Step S401: Based on the joint regional strain trajectory The strain rate of change is calculated using the inter-frame difference method. strain rate of change Used to represent the rate of evolution of organizational structure deformation; Step S402: Introduce the cumulative deformation period Based on cumulative deformation period and strain rate of change The cumulative index of joint strain is calculated using the absolute value time integration method. Combined strain cumulative index Used to indicate the active stress state of tissue structure; Step S403: Set the consecutive frame count threshold Continuously monitor the rate of change of strain When the following conditions are met: continuous one cumulative deformation cycle Internal strain rate of change All are greater than the preset minimum instability threshold. At that time, determine that the current situation is in the strain instability trend zone and output the duration of monotonous strain growth. , ; Step S404: Obtain the total duration of the second stage of labor The final output is the joint strain cumulative index. Duration of monotonic strain growth Total duration of the second stage of labor .

7. The imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking as described in claim 1, characterized in that, In step S50, based on the joint strain accumulation index Duration of monotonic strain growth Total duration of the second stage of labor The steps for constructing the SPPH risk scoring function and outputting image risk detection results based on the SPPH risk scoring function include: Step S501: Define the first fuzzy rule set, the second fuzzy rule set, and the third fuzzy rule set, and combine the joint strain cumulative index. Duration of monotonic strain growth Total duration of the second stage of labor The input membership triples are generated by mapping them to the first fuzzy rule set, the second fuzzy rule set, and the third fuzzy rule set, respectively. Step S502: Based on the preset fuzzy rule base R, perform fuzzy inference on the input membership triples and output a fuzzy output set. ; Step S503: Set the fuzzy output The input is fed into a preset deblurring function, which uses the centroid method to blur the output set. Perform numerical processing of risk levels and output continuous risk score values. , Among them, continuous risk score Used to characterize the intensity of labor risk reflected in the current target image segment, when When the time is right, it indicates a low risk during labor; when When the time is right, it indicates a moderate risk during labor; when At this time, it indicates a high risk during labor; Step S504: Transfer the continuous risk score value The visualization is overlaid onto the original two-dimensional grayscale ultrasound image sequence, and the final output is the image risk detection result.

8. An imaging detection system for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking, applied to the imaging detection method for prolonged second stage of labor and SPPH risk in vaginal delivery of twins based on ultrasound speckle-tracking as described in any one of claims 1 to 7, characterized in that, The imaging detection system for prolonged second stage of labor and SPPH risk in twin vaginal deliveries based on ultrasound speckle-tracking includes: The joint analysis region annotation module is used to acquire a sequence of two-dimensional grayscale ultrasound images F transvaginally or transperineally during the second stage of labor in twin vaginal deliveries. Based on the ultrasound two-dimensional grayscale image sequence F, a multi-layer semantic mask generation mechanism based on structural continuity and region displacement priors is used to perform the joint analysis region annotation task, and outputs a joint analysis region mask between the cervical wall and the anterior and posterior vaginal walls. ; The displacement field construction module is used to perform Speckle pixel block tracking tasks based on a texture sub-block matching mechanism using multi-scale pyramids and singular value decomposition, using a joint analysis region mask, and outputs the joint region displacement field. ; The nonlinear strain trajectory calculation module is used for calculations based on the joint region displacement field. A logarithmic strain modeling and strain trajectory accumulation extraction mechanism is used to perform finite nonlinear strain calculation tasks, outputting the joint region strain trajectory that varies with time. ; The strain evolution feature extraction module is used to extract strain evolution features based on the joint region strain trajectory. The strain evolution assessment mechanism based on strain rate of change integral and monotonic growth segment identification is used to perform the second stage feature extraction task, and the joint strain accumulation index is output. Duration of monotonic strain growth Total duration of the second stage of labor ; The risk assessment and output module is used to assess and output risks based on the joint strain cumulative index. Duration of monotonic strain growth Total duration of the second stage of labor Construct the SPPH risk scoring function and output the image risk detection results based on the SPPH risk scoring function.

9. An imaging detection device based on ultrasound speckle-tracking for the risk of prolonged second stage of labor and SPPH in vaginal delivery of twins, characterized in that, The imaging detection device for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking includes: a memory, a processor, and an imaging detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking, stored in the memory and executable on the processor. When the imaging detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking is executed by the processor, it implements the imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking according to any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes an imaging detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking. When the imaging detection program for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking is executed by the processor, it implements an imaging detection method for prolonged second stage of labor and SPPH risk in twin vaginal delivery based on ultrasound speckle-tracking as described in any one of claims 1 to 7.

Images