An ultrasonic contrast-based precise positioning method and system for jejunum catheterization

Abstract

This invention provides a method and system for precise positioning of jejunal tubes based on contrast-enhanced ultrasound, relating to the field of biomedical technology. The method includes: acquiring ultrasound contrast-enhanced data; preprocessing the ultrasound contrast-enhanced data to obtain a standardized ultrasound contrast-enhanced sequence; generating candidate features and constructing non-tube-body prior constraints based on the standardized ultrasound contrast-enhanced sequence; combining the candidate features and non-tube-body prior constraints to perform joint kernel density estimation disambiguation discrimination in the standardized ultrasound contrast-enhanced sequence to obtain disambiguation discrimination results; evaluating the head-end confidence of the candidate features based on the standardized ultrasound contrast-enhanced sequence to obtain head-end confidence evaluation results; performing time window consistency adjudication analysis on the disambiguation discrimination results and the head-end positioning confidence evaluation results to obtain adjudication results; and obtaining the jejunal head-end position and tube trajectory based on the adjudication results.

Description

Technical Field

[0001] This invention relates to the field of biomedical technology, and in particular to a method and system for precise positioning of jejunal tube placement based on ultrasound contrast imaging. Background Technology

[0002] Image localization of intestinal tubing is an application area that integrates clinical imaging and bedside operation management. Its core task is to provide a visual basis for judging the position of tubing and to support the verification of results under the condition of fluctuations in different individual signs and imaging conditions. The technological development in this field plays an important supporting role in the standardization and precision of clinical bedside operation, and also promotes the implementation and application of medical imaging technology in the bedside setting.

[0003] The existing technological system revolves around medical imaging techniques, encompassing objective imaging confirmation methods represented by radiological imaging, and bedside real-time observation methods represented by ultrasound. Ultrasound, due to its convenience, non-invasiveness, and repeatability, has become a key focus of research and application in this field, driving the development of various enhancement imaging, contrast display, and image quality optimization approaches. Simultaneously, clinical demands for speed and traceability have prompted the formation of standardized practices in this field, including standardized recording of imaging data, verification of observation results across different time periods, and structured expression of localization conclusions. This has resulted in a technological landscape characterized by the parallel application of multiple imaging techniques, the complementarity of bedside observation and objective confirmation, and the gradual standardization of recording and verification mechanisms.

[0004] However, existing in vivo tubing localization technologies lack stable and quantifiable distinguishing criteria, and the continuity and consistency constraints and traceability mechanisms of localization conclusions are insufficient, making it difficult to meet the consistency requirements of accurate localization and result verification. Summary of the Invention

[0005] To address the technical problems of existing in vivo tube positioning technologies lacking stable and quantifiable distinguishing criteria, insufficient continuity and consistency constraints and traceability mechanisms for positioning conclusions, and difficulty in meeting the consistency requirements of accurate positioning and result verification, this invention provides a method and system for accurate positioning of jejunal tubes based on contrast ultrasound.

[0006] The technical solutions provided by the embodiments of the present invention are as follows: The first aspect of this invention provides a method for precise positioning of a jejunal tube based on contrast-enhanced ultrasound, comprising: S1: Acquire relevant data for ultrasound contrast imaging; S2: Preprocess the ultrasound contrast imaging data to obtain a standardized ultrasound contrast imaging sequence; S3: Based on standardized ultrasound contrast imaging sequences, candidate features are generated and non-tubular prior constraints are constructed; S4: Combining candidate features with non-tube prior constraints, joint kernel density estimation and disambiguation discrimination are performed in a standardized ultrasound contrast imaging sequence to obtain the disambiguation discrimination result; S5: Based on standardized ultrasound contrast imaging sequences, head-end confidence assessment is performed on candidate features to obtain head-end confidence assessment results; S6: Perform time window consistency analysis on the disambiguation judgment results and the head-end positioning information evaluation results respectively to obtain the adjudication results; S7: Based on the ruling, the position of the jejunal proximal end and the trajectory of the tube are obtained.

[0007] A second aspect of the present invention provides a precise positioning system for jejunal tube placement based on contrast-enhanced ultrasound, comprising: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the precise positioning method for jejunal tube placement based on ultrasound contrast imaging as described in the first aspect.

[0008] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: This invention addresses the lack of stable and quantifiable distinguishing criteria in existing technologies. By constructing prior constraints for non-tube structures and performing joint kernel density estimation disambiguation, it achieves prior isolation and continuous suppression of non-target strong echo structures, forming a stable and quantifiable distinguishing criterion for echoes, effectively differentiating strong echoes from tube structures and non-tube structures. To address the insufficient continuity and consistency constraints and traceability mechanisms in localization conclusions, it achieves directional, head-end consistency localization and confidence constraints through head-end confidence assessment, and introduces time-window consistency adjudication analysis, ensuring the continuity and traceability of localization outputs. This enables stable issuance and convenient review of localization conclusions, comprehensively meeting the clinical requirements for consistency in accurate localization and result review. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0010] Figure 1 This is a flowchart illustrating a method for precise positioning of jejunal tube placement based on ultrasound contrast imaging, provided in an embodiment of the present invention.

[0011] Figure 2 This is a schematic diagram of a precise positioning system for jejunal tube placement based on ultrasound contrast imaging, provided as an embodiment of the present invention. Detailed Implementation

[0012] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0013] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0014] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0015] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0016] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0017] Reference manual attached Figure 1 The diagram illustrates a flowchart of a precise positioning method for jejunal tube placement based on ultrasound contrast imaging, provided by an embodiment of the present invention.

[0018] This invention provides a method for precise positioning of jejunal tubes based on contrast-enhanced ultrasound. This method can be implemented using a device for precise positioning of jejunal tubes based on contrast-enhanced ultrasound, which can be a terminal or a server. The processing flow of this method may include the following steps: S1: Acquire relevant data for ultrasound contrast imaging.

[0019] Optionally, the ultrasound contrast-enhanced data may specifically include real-time contrast-enhanced data, injection synchronization timing data, and historical reference data.

[0020] Among them, real-time contrast imaging data is imaging-related data acquired in real time under ultrasound contrast imaging mode, injection synchronization timing data is time-dimensional data related to contrast agent injection operation, and historical reference data is comparative data and statistical data accumulated from past ultrasound contrast imaging.

[0021] S2: Preprocess the ultrasound contrast imaging data to obtain a standardized ultrasound contrast imaging sequence.

[0022] Preprocessing involves standardizing and unifying ultrasound contrast-enhanced data of different types and scales. Standardized ultrasound contrast-enhanced sequences are standardized datasets that integrate various processed data and can be directly used for subsequent localization analysis.

[0023] In one possible implementation, S2 specifically includes sub-steps S201 to S207: S201: The real-time imaging data is aligned using a phase correlation registration algorithm and a pyramid dense optical flow alignment algorithm to obtain an aligned image sequence.

[0024] Among them, the phase correlation registration algorithm is an algorithm for image registration based on phase information, the pyramid dense optical flow alignment algorithm is an algorithm for image alignment by constructing an image pyramid and calculating the dense optical flow field, the alignment processing is an operation to correct inter-frame jitter and achieve pixel-level position matching, and the aligned image sequence is a set of real-time angiographic images after the alignment processing is completed.

[0025] Furthermore, the search window scale and subpixel refinement strategy of the phase correlation registration algorithm are jointly constrained by the imaging depth data and focus position data in the real-time angiography data. The number of layers, iterations, and smoothing weights of the pyramid dense optical flow alignment algorithm are determined by offline calibration of the historical angiography sequence and historical grayscale sequence in the historical reference data, and are limited by the maximum displacement threshold calculated from the acquired frame rate data in the real-time angiography data.

[0026] S202: The aligned image sequence is subjected to noise smoothing and edge fidelity preservation processing by using the nonlocal mean speckle denoising algorithm and the anisotropic diffusion edge-preserving smoothing algorithm to obtain the denoised and enhanced image sequence.

[0027] Among them, the nonlocal mean speckle denoising algorithm is an algorithm that removes speckle noise based on nonlocal similarity, the anisotropic diffusion edge-preserving smoothing algorithm is an algorithm that preserves the image edge features while smoothing noise, noise smoothing is an operation to reduce noise interference in the image, edge fidelity processing is an operation to protect the key features of the image such as edges and contours from being blurred, and the denoised and enhanced image sequence is a set of images after noise processing and edge protection.

[0028] Furthermore, the filtering intensity of the nonlocal mean speckle denoising algorithm is adaptively adjusted by the far-field noise level in the real-time imaging data, and the iteration step number and conduction threshold of the anisotropic diffusion edge-preserving smoothing algorithm are determined by the local gradient statistics of the current frame, so as to avoid excessive smoothing leading to the degradation of candidate fine strip endpoints.

[0029] S203: The contrast enhancement processing is performed on the denoised and enhanced image sequence by using the quantile mapping normalization algorithm and the local contrast-limited histogram equalization algorithm to obtain the contrast-enhanced image sequence.

[0030] Among them, the quantile mapping normalization algorithm is an algorithm that achieves the uniformity of the dynamic range of data through quantile mapping; the local contrast-limited histogram equalization algorithm is a histogram equalization algorithm that limits the contrast in a local area; the contrast enhancement processing is an operation that improves the contrast between light and dark areas of an image and highlights key structures; and the image sequence after contrast enhancement is a collection of images after contrast enhancement.

[0031] Furthermore, the upper and lower bounds of the quantile mapping normalization algorithm are obtained by sliding time window statistics, and the grid scale and clipping threshold of the local contrast-limited histogram equalization algorithm are adjusted in conjunction with the changes in the overall gain data in the real-time imaging data, and are calibrated with reference to the stable distribution of historical grayscale sequences in historical reference data.

[0032] S204: Generate injection event anchors by combining real-time angiography data and injection synchronization time series data using a timestamp alignment algorithm.

[0033] Among them, the timestamp alignment algorithm is an algorithm that matches and binds the timestamps of different data, and the injection event anchor point is a reference node formed by binding the time node of the contrast agent injection operation with the ultrasound contrast imaging time axis.

[0034] Furthermore, the alignment tolerance of the timestamp alignment algorithm is determined by converting the acquisition frame rate data in the real-time angiography data, and the boundary of the injected event anchor point is checked for consistency closure by the injection start time, injection duration and injection end time in the injection synchronization timing data, so as to reduce the impact of clock jitter and recording deviation on the reference node.

[0035] S205: Standardize and normalize the imaging-related parameters in the real-time angiography data and the injected synchronous time-series data to obtain standardized imaging parameters.

[0036] Among them, imaging-related parameters are parameters related to equipment settings and acquisition conditions during ultrasound contrast imaging; standardization is the operation to eliminate differences in data dimensions and achieve a unified data scale; normalization is the operation to map data to a specified range; and standardized imaging parameters are the set of imaging-related parameters after standardization and normalization.

[0037] Specifically, imaging-related parameters include frame rate data, imaging depth data, focus position data, overall gain data, mechanical index data, and injection duration.

[0038] S206: By using a logarithmic compression normalization algorithm, historical reference data is subjected to nonlinear compression processing to obtain standardized historical statistical data.

[0039] Among them, the logarithmic compression normalization algorithm is an algorithm that achieves nonlinear compression and normalization of data through logarithmic transformation. Nonlinear compression processing is an operation that performs non-uniform compression of data and suppresses the influence of long-tail distribution. Standardized historical statistical data is a set of historical reference data after nonlinear compression.

[0040] S207: Integrate contrast-enhanced image sequences, injection event anchors, standardized imaging parameters, and standardized historical statistical data to generate standardized ultrasound contrast sequences.

[0041] Integration involves merging and encapsulating standardized data of different types.

[0042] In this embodiment of the invention, data alignment, denoising, enhancement, and standardization are achieved through multi-step preprocessing to construct a unified and analyzable standardized ultrasound contrast imaging sequence, laying a high-quality data foundation for accurate localization.

[0043] S3: Based on standardized ultrasound contrast sequences, candidate features are generated and non-tube body prior constraints are constructed.

[0044] Among them, candidate features are a set of various features related to the characteristics of the jejunal tube extracted from standardized ultrasound contrast imaging sequences, while non-tube prior constraints are prior judgment criteria constructed based on historical data to distinguish between tube and non-tube strong echo structures.

[0045] Optionally, the process of generating candidate features specifically includes: By using a multi-scale ridge enhancement algorithm and a thin connected component extraction algorithm, the contrast image sequence in the standardized ultrasound contrast imaging sequence is processed to generate a candidate thin strip set.

[0046] Among them, the multi-scale ridge enhancement algorithm is an algorithm that enhances the ridge structure in an image in a multi-scale space; the elongated connected region extraction algorithm is an algorithm that extracts elongated connected regions from an image; the contrast image sequence is a continuous image frame in the ultrasound contrast mode of a standardized ultrasound contrast sequence; and the candidate strip set is a set of extracted image regions that conform to the tube-like strip features.

[0047] For each candidate thin strip in the candidate thin strip set, calculate the skeleton-related features and the direction change sequence to obtain the basic features of the candidate thin strip.

[0048] Among them, skeleton-related features are features that describe the morphology of the candidate strip skeleton, direction change sequence is a sequence that records the change pattern of the candidate strip in the extension direction, and candidate strip basic features are a set of basic features related to the morphology and direction of the candidate strip.

[0049] A strip sampling window is constructed along the skeleton normal, centered on the set of skeleton pixels of the candidate strip.

[0050] Among them, the skeleton pixel set is the set of pixels that constitute the candidate thin strip skeleton, the skeleton normal is the direction perpendicular to the extension direction of the candidate thin strip skeleton, and the strip sampling window is a strip region constructed to extract the surrounding features of the candidate thin strip.

[0051] Based on the strip sampling window, the median of each frame of the grayscale image sequence in the standardized ultrasound contrast sequence is taken to obtain the grayscale response value of a single frame.

[0052] Among them, the grayscale image sequence is a series of image frames in the two-dimensional grayscale mode of ultrasound in a standardized ultrasound contrast imaging sequence, and the grayscale response value of a single frame is the median of the pixel grayscale within the strip sampling window of a single frame grayscale image.

[0053] Based on the frame timestamp sequence in the standardized ultrasound contrast imaging sequence, the grayscale response values ​​of a single frame are sorted to obtain the grayscale response sequence.

[0054] Among them, the frame timestamp sequence is a sequence that records the acquisition time of ultrasound contrast imaging frames, and the grayscale response sequence is a set of single-frame grayscale response values ​​sorted by time.

[0055] The median of each frame of the contrast-enhanced ultrasound image sequence in the standardized ultrasound contrast-enhanced sequence is taken to obtain the single-frame contrast-enhanced response value.

[0056] Among them, the single-frame angiography effect value is the median of the pixel angiography intensity within the strip sampling window of a single-frame angiography image.

[0057] Based on the overall gain data in the standardized ultrasound contrast imaging sequence, quantile mapping correction is performed on the single-frame contrast effect value.

[0058] Among them, the overall gain data is the overall gain setting parameter of the ultrasound equipment, and the quantile mapping correction is the operation of correcting the influence effect value through quantile mapping.

[0059] Based on the mechanical index data in the standardized ultrasound contrast imaging sequence, the single-frame contrast effect value after quantile mapping correction is subjected to exponential compression correction.

[0060] Among them, mechanical index data is a parameter reflecting the acoustic output conditions of ultrasound contrast imaging, and exponential compression correction is an operation that compresses and corrects the contrast effect value through exponential transformation.

[0061] Based on the frame timestamp sequence in the standardized ultrasound contrast imaging sequence, the single-frame contrast effect values ​​after exponential compression correction are sorted to obtain the candidate thin strip contrast intensity time series.

[0062] Among them, the candidate fine stripe angiography intensity time series is a set of corrected single-frame angiography response values ​​sorted by time.

[0063] Based on the optical flow field output by the pyramid dense optical flow alignment algorithm, the optical flow gradient norm sequence is calculated for the neighborhood of the candidate thin strip.

[0064] Among them, the optical flow field is a vector field describing the motion of pixels in the image, the neighborhood of the candidate strip is the image region surrounding the candidate strip, and the optical flow gradient norm sequence is a sequence recording the changes in the optical flow gradient norm within the neighborhood of the candidate strip.

[0065] Candidate features are obtained by combining the grayscale response sequence, the time series of candidate fine strip imaging intensity, and the optical flow gradient norm sequence.

[0066] Furthermore, the multi-scale ridge enhancement algorithm adopts the Frangi ridge response function based on the eigenvalues ​​of the Hessian matrix, and takes the maximum response in the multi-scale space to form the ridge enhancement result. Then, the enhancement result is labeled with connected components by the slender connected component extraction algorithm and screened by skeletonization and principal axis slenderness constraints, finally obtaining a set of candidate ridges.

[0067] Optionally, the process of constructing non-tube prior constraints specifically includes: For the historical contrast-enhanced ultrasound sequence and the historical grayscale sequence in the standardized ultrasound contrast-enhanced sequence, a candidate fine line generation process consistent with the real-time process is performed to obtain the historical fine line feature set.

[0068] Among them, the historical angiography sequence is the angiography image sequence in the historical reference data, the historical grayscale sequence is the grayscale image sequence in the historical reference data, and the historical fine strip feature set is the fine strip feature set extracted from the historical sequence.

[0069] By combining the historical fine line feature set with the number of historical fine lines and the density of historical artifacts in standardized ultrasound contrast sequences, a non-tube strong echo sample set is constructed.

[0070] Among them, the number of historical fine lines is the frequency of occurrence of fine line structures in historical imaging, the density of historical artifacts is the degree of artifact enrichment in historical imaging, and the non-tube strong echo sample set is a sample set containing the characteristics of various non-tube strong echo structures.

[0071] Based on the non-tube strong echo sample set, non-tube prior constraints are formed.

[0072] In this embodiment of the invention, multi-dimensional features of candidate thin strips are extracted and non-tube body prior constraints are constructed to provide morphological, temporal and historical statistical criteria for distinguishing strong echoes between tube bodies and non-tube bodies.

[0073] S4: Combining candidate features with non-tube prior constraints, joint kernel density estimation and disambiguation discrimination are performed in standardized ultrasound contrast imaging sequences to obtain the disambiguation discrimination results.

[0074] Among them, the joint kernel density estimation disambiguation discrimination is a discrimination operation that combines candidate features and non-tube body prior constraints, and uses the kernel density estimation algorithm to distinguish between tube body and non-tube body strong echo structures. The disambiguation discrimination result is the judgment data obtained after the disambiguation discrimination is completed, which is used to distinguish between tube body and non-tube body echoes.

[0075] In one possible implementation, S4 specifically includes sub-steps S401 to S408: S401: The candidate trajectory density is obtained by calculating the historical occurrence density of the corresponding feature vectors of the candidate thin strip feature sequences in the candidate features through the kernel density estimation algorithm.

[0076] Among them, the kernel density estimation algorithm is a nonparametric statistical algorithm that estimates the probability density of data through kernel functions. The candidate thin strip feature sequence is a sequence that records the changes of candidate thin strip features over time. The feature vector is a vector formed by integrating the multi-dimensional features of the candidate thin strip. The historical occurrence density is the probability density of the feature vector appearing in historical data. The candidate trajectory density is the historical occurrence density of the feature vector corresponding to the candidate thin strip feature sequence.

[0077] Furthermore, the kernel density estimation algorithm uses a Gaussian kernel function as the kernel function type. The bandwidth is adaptively determined by the sample variance and sample size of the historical reference data through Silverman's empirical rule, and the bandwidth remains consistent within the same device configuration range, thereby ensuring the uniformity of candidate trajectory density calculation.

[0078] S402: By using the kernel density estimation algorithm, combined with the historical thin strip feature set in the candidate features and the prior constraints of non-tube bodies, the distribution density of non-tube bodies is obtained.

[0079] Among them, the non-tube distribution density is the probability density of the strong echo structure features of non-tubes in the feature space.

[0080] S403: Calculate the candidate trajectory density and the non-tube distribution density to obtain the density logarithmic ratio term.

[0081] The density logarithmic ratio term is the value obtained by taking the natural logarithm of the ratio of the candidate trajectory density to the non-tube distribution density.

[0082] S404: Process the time series of candidate fine strip angiography intensity in the candidate features to obtain the injection time series deviation distance.

[0083] Among them, the injection timing deviation distance is the matching deviation value between the candidate thin strip contrast intensity time series and the contrast agent injection timing.

[0084] Furthermore, the injection timing deviation distance is calculated using a dynamic time warping distance algorithm. This algorithm uses a correlation-based distance metric and employs a Sakhoé-Chiba bandwidth constraint to limit the local offset range of the alignment path, thereby suppressing the influence of non-physical stretching on the calculation results of the injection timing deviation distance.

[0085] S405: Calculate the distance between the density logarithmic ratio term and the injection timing deviation to obtain the corrected density logarithmic ratio value.

[0086] The corrected density logarithmic ratio is a value obtained by combining the density logarithmic ratio term with the injection timing deviation distance correction.

[0087] S406: Process the optical flow gradient norm sequence in the standardized ultrasound contrast sequence to obtain the motion disturbance intensity and noise correction denominator in the trajectory neighborhood.

[0088] Among them, the motion disturbance intensity of the trajectory neighborhood is a value describing the motion instability in the neighborhood of the candidate strip, and the noise correction denominator is a reference value for noise correction calculated in combination with the far-field noise level.

[0089] Furthermore, the calculation area for the far-field noise level is determined by the imaging depth data in the real-time imaging data through a depth ratio segmentation algorithm. That is, after converting the depth axis of a single frame image into physical depth coordinates according to the imaging depth data, the pixel band corresponding to the last 30% to 40% interval of the imaging depth data is taken as the far-field region.

[0090] S407: Calculate the disturbance intensity and noise correction denominator of the trajectory neighborhood to obtain the disturbance noise index suppression term.

[0091] Among them, the disturbance noise exponential suppression term is a multiplicative suppression weight obtained by combining the motion disturbance intensity in the trajectory neighborhood with the noise correction denominator term through exponential transformation.

[0092] S408: Calculate the logarithmic ratio of the correction density and the disturbance noise index suppression term to obtain the disambiguation judgment result.

[0093] The disambiguation result is the echo disambiguation value, which is the core of the disambiguation process. The calculation formula is as follows: Where J(t) represents the echo disambiguation discriminant value at time t, p t Let q represent the candidate trajectory density at time t. t d represents the non-tube volume distribution density at time t. t G represents the injection timing deviation at time t. t n represents the intensity of motion disturbance in the trajectory neighborhood at time t.t Let t represent the far-field noise level at time t, ln() represents the natural logarithm operation with base e, and exp() represents the natural exponential operation.

[0094] Specifically, the calculation of the echo disambiguation discrimination value requires solving the density logarithmic ratio term, the time-corrected density logarithmic ratio term, and the disturbance noise index suppression term in sequence. Then, the time-corrected density logarithmic ratio term is multiplied by the disturbance noise index suppression term, and the final value is the echo disambiguation discrimination value used to distinguish strong echoes from tubes and non-tubes.

[0095] In this embodiment of the invention, by combining kernel density estimation disambiguation discrimination and formula calculation, a stable quantitative distinction between strong echoes from pipes and non-pipes is achieved, thereby improving the anti-interference capability and discrimination accuracy of pipeline positioning.

[0096] S5: Based on standardized ultrasound contrast imaging sequences, head-end confidence assessment is performed on candidate features to obtain head-end confidence assessment results.

[0097] Among them, the cephalic end confidence assessment is an operation that quantitatively evaluates the confidence level of the cephalic end position of the jejunal tube corresponding to the candidate feature. The cephalic end confidence assessment result is a numerical value that quantitatively represents the confidence level of the cephalic end positioning.

[0098] In one possible implementation, S5 specifically includes sub-steps S501 to S509: S501: Define the set of candidate strips in the candidate features as the tube body travel input set.

[0099] Among them, the tube body course input set is a set of candidate strips used to analyze the course trajectory of the jejunal tube body.

[0100] S502: Based on the standardized ultrasound contrast imaging sequence, the contrast image sequence is used to perform cross-frame correlation processing on the tube body course input set to obtain the candidate thin strip end point sequence.

[0101] Among them, cross-frame association processing is an operation of matching and tracking the same candidate thin strip in different frames. The candidate thin strip end point sequence is a set of end point coordinates of the candidate thin strip in each frame, ordered by time.

[0102] S503: Determine the end of the tube trajectory based on the sequence of candidate thin strip end points.

[0103] Among them, the end of the tube trajectory is the end position of the tube trajectory corresponding to the candidate thin strip.

[0104] S504: The enhanced distribution entropy is obtained by calculating the end of the pipe trajectory using the information entropy calculation algorithm.

[0105] Among them, the information entropy calculation algorithm is an algorithm for calculating the information entropy of data, and the enhanced distribution entropy is an information entropy value that describes the complexity of the distribution of the enhanced intensity of the angiography in the forward neighborhood at the end of the tube trajectory.

[0106] Furthermore, the calculation of the enhanced distribution entropy requires first obtaining the enhanced intensity distribution of the forward neighborhood at the end of the tube trajectory through the forward fan-shaped neighborhood statistical mapping algorithm. That is, a forward fan-shaped neighborhood is constructed with the end point of the tube trajectory as the center and the tangential direction of the end as the fan-shaped axis. Adaptive binning histogram statistics are performed on the imaging intensity of the pixels in the neighborhood and normalized. Then, the enhanced distribution entropy is obtained by solving the enhanced intensity distribution through the information entropy calculation algorithm.

[0107] S505: Based on the standardized ultrasound contrast imaging sequence, clustering and position analysis are performed on the end of the tube trajectory to obtain the fluctuation of the cluster center position and the end alignment distance.

[0108] Among them, cluster analysis is an analysis method that aggregates pixels with similar features into classes; position analysis is an analysis operation on the changes in the position of the cluster center; cluster center position fluctuation describes the degree of fluctuation in the position of the contrast enhancement cluster center over time; and end alignment distance is the geometric alignment error value between the end point of the tube trajectory and the enhancement cluster center point.

[0109] Furthermore, the end alignment distance is solved by the vector projection consistency error calculation algorithm. That is, the unit vector of the end point of the tube trajectory in the end tangential direction is used as the direction basis vector. The displacement vector of the enhanced aggregation center point relative to the end point is projected onto the direction basis vector to obtain the forward projection amount. At the same time, the absolute offset of the displacement vector in the normal component is calculated. The ratio of the absolute offset of the normal component to the forward projection amount is the end alignment distance.

[0110] S506: Calculate the enhanced distribution entropy to obtain the entropy exponential suppression term.

[0111] Among them, the entropy exponential suppression term is a multiplicative confidence weight obtained by taking the negative of the enhanced distribution entropy and then taking the exponent.

[0112] S507: Calculate the fluctuation at the location of the aggregation center to obtain the inverse fluctuation suppression term.

[0113] Among them, the reciprocal suppression term of fluctuation is a multiplicative confidence weight obtained by adding one to the fluctuation at the center of aggregation and taking the reciprocal.

[0114] S508: Calculate the end alignment distance to obtain the alignment index suppression term.

[0115] The alignment index suppression term is a multiplicative confidence weight obtained by negativeening the end alignment distance and taking the exponent.

[0116] S509: Calculate the entropy exponent suppression term, the inverse fluctuation suppression term, and the alignment exponent suppression term to obtain the head-end confidence assessment result.

[0117] Among them, the head-end confidence assessment result is the head-end location confidence value, which is the core quantitative result of the head-end confidence assessment, and its calculation formula is as follows: Where C(t) represents the head-end positioning information value at time t, e t Let s represent the enhanced distribution entropy at time t. t This represents the fluctuation of the aggregation center position at time t, and exp() represents the natural exponent operation. This represents the end alignment distance at time t.

[0118] Specifically, the calculation of the head-end positioning confidence value requires first solving the entropy exponent suppression term, the inverse fluctuation suppression term, and the alignment exponent suppression term, and then multiplying the three suppression terms in sequence. The final value obtained is the head-end positioning confidence value that characterizes the reliability of head-end positioning.

[0119] In this embodiment of the invention, the confidence of head-end positioning is quantitatively characterized by multi-dimensional head-end confidence assessment and formula calculation, providing a reliable positioning confidence basis for subsequent consistency decisions.

[0120] S6: Perform time window consistency analysis on the disambiguation judgment results and the head-end positioning information evaluation results respectively to obtain the adjudication results.

[0121] Among them, the time window consistency adjudication analysis is an operation that judges and analyzes the consistency between the disambiguation judgment result and the head-end confidence assessment result within a continuous time window. The adjudication result is a value obtained after the consistency analysis is completed, which is used to determine whether to output the positioning result.

[0122] In one possible implementation, S6 specifically includes sub-steps S601 to S607: S601: Based on the frame timestamp sequence of the standardized ultrasound contrast imaging sequence, the temporal consistency deviation is obtained by calculating the candidate head position sequence within adjacent time windows through the position distribution difference calculation algorithm.

[0123] Among them, the time window is a time interval divided according to time and used for analysis; the candidate head position sequence is the set of candidate head position coordinates within each time window; the position distribution difference calculation algorithm is an algorithm to calculate the degree of difference in position distribution between different sets; and the time consistency deviation is a value that describes the stability of the candidate head position distribution within adjacent time windows.

[0124] S602: Process the echo disambiguation discriminant value in the disambiguation result to obtain the echo disambiguation logarithmic enhancement term.

[0125] Among them, the echo disambiguation discriminant value is the core judgment value in the disambiguation result, and the echo disambiguation logarithmic enhancement term is the enhancement value obtained after performing a logarithmic transformation on the echo disambiguation discriminant value.

[0126] S603: Process the head-end positioning confidence value in the head-end confidence assessment results to obtain the head-end positioning log enhancement term.

[0127] Among them, the head-end positioning confidence value is the core quantitative value in the head-end confidence assessment result, and the head-end positioning logarithmic enhancement term is the enhancement value obtained after logarithmically transforming the head-end positioning confidence value.

[0128] S604: Calculate the logarithmic enhancement term for echo disambiguation and the logarithmic enhancement term for head-end localization to obtain the joint logarithmic enhancement product term.

[0129] Among them, the joint logarithmic enhancement product term is a numerical value that comprehensively represents the disambiguation and localization reliability obtained by multiplying the two logarithmic enhancement terms.

[0130] S605: Calculate the time consistency deviation to obtain the consistency deviation denominator.

[0131] The consistency deviation denominator is a baseline value used for correction, obtained by adding one to the time consistency deviation.

[0132] S606: Calculate the joint logarithmic enhancement product term and the consistency deviation denominator term to obtain the positioning and issuance value.

[0133] Among them, the location issuance value is a core value obtained after comprehensively considering disambiguation, location credibility, and time consistency, used to determine whether to issue a location result.

[0134] Among them, the location issuance value is the core judgment value for time window consistency adjudication analysis, and its calculation formula is as follows: Where P(t) represents the positioning and issuance value at time t, J(t) represents the echo disambiguation judgment value at time t, C(t) represents the head-end positioning signal value at time t, and r t Let t represent the time consistency deviation at time t, and ln() represent the natural logarithm operation with base e.

[0135] Specifically, the calculation of the positioning issuance value requires first solving the echo disambiguation logarithmic enhancement term and the head-end positioning logarithmic enhancement term. The two are multiplied to obtain the joint logarithmic enhancement product term. Then, the joint logarithmic enhancement product term is divided by the consistency deviation denominator term, and the final value is the positioning issuance value.

[0136] S607: Use the location issuance value as the adjudication result obtained from the time window consistency adjudication analysis.

[0137] In this embodiment of the invention, by combining time window consistency adjudication and formula calculation, disambiguation judgment, location information and time stability, a core judgment value that can be directly used to issue location results is output.

[0138] S7: Based on the ruling, the position of the jejunal proximal end and the trajectory of the tube are obtained.

[0139] Among them, the jejunal proximal end position is the specific spatial location of the proximal end of the jejunal tube in the body, and the tube trajectory is the path of the jejunal tube in the body.

[0140] Alternatively, S7 specifically refers to: Determine if the location issuance value in the ruling result is less than the location issuance threshold. If so, maintain the frozen marker for the non-tube strong echo set and review the candidate head position set until the location issuance threshold is met, then output the jejunal head position and tube trajectory. Otherwise, directly output the jejunal head position and tube trajectory, while keeping the frozen marker for the non-tube strong echo set valid.

[0141] Among them, the positioning issuance threshold is the critical value for determining whether the positioning issuance value meets the requirements of the output positioning result; the non-tube strong echo set freeze mark is a mark for non-tube strong echo structures to prevent them from participating in subsequent positioning calculations; and the verification is an operation to analyze and verify the candidate head end position set again.

[0142] It should be noted that those skilled in the art can set the location issuance threshold according to actual needs, and this invention does not limit it.

[0143] Furthermore, when the location issuance value is less than the location issuance threshold, there is an upper limit to the number of consecutive review time windows for the candidate head position set. This upper limit is the number of time windows covered by twice the injection duration in the injected synchronous time sequence data. If the location issuance threshold is not met even after reaching this upper limit, the review will be terminated and a review termination flag will be output to avoid infinite waiting.

[0144] In this embodiment of the invention, the position of the jejunal head and the trajectory of the tube are determined based on the positioning and issuance value, and a review upper limit mechanism is set to ensure positioning accuracy while avoiding infinite waiting and improving clinical applicability.

[0145] Reference manual attached Figure 2 The diagram shows a structural schematic of a jejunal tube placement precision positioning system based on ultrasound contrast imaging provided by the present invention.

[0146] The present invention also provides a jejunal tube placement precision positioning system 20 based on contrast-enhanced ultrasound, applied to the above-mentioned jejunal tube placement precision positioning method based on contrast-enhanced ultrasound, comprising: Processor 201.

[0147] The memory 202 stores computer-readable instructions. When the computer-readable instructions are executed by the processor 201, they implement the precise positioning method for jejunal tube placement based on ultrasound contrast imaging as described in the method embodiment.

[0148] The jejunal tube placement precision positioning system 20 based on ultrasound contrast imaging provided by this invention can perform the above-mentioned jejunal tube placement precision positioning method based on ultrasound contrast imaging and achieve the same or similar technical effects. To avoid repetition, this invention will not elaborate further.

[0149] It should be understood that the processor in the embodiments of the present invention can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0150] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0151] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0152] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0153] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0154] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0155] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0156] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0157] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0158] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0159] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0160] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0161] This invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the precise positioning method for jejunal tube placement based on ultrasound contrast imaging as described in the method embodiment.

[0162] The present invention provides a computer-readable storage medium that can realize the steps and effects of the jejunal tube placement precision positioning method based on ultrasound contrast imaging in the above-described method embodiments. To avoid repetition, the present invention will not repeat the details.

[0163] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0164] The following points need to be explained: (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0165] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the invention, i.e., these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element or there may be intermediate elements.

[0166] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0167] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An ultrasonic contrast-based precise positioning method for jejunum catheterization, characterized in that, include: S1: Acquire relevant data for ultrasound contrast imaging; S2: Preprocess the ultrasound contrast-enhanced data to obtain a standardized ultrasound contrast-enhanced sequence; S3: Based on the standardized ultrasound contrast imaging sequence, generate candidate features and construct non-tube prior constraints; S4: Combining the candidate features with the non-tube prior constraints, perform joint kernel density estimation disambiguation discrimination in the standardized ultrasound contrast imaging sequence to obtain the disambiguation discrimination result; S5: Based on the standardized ultrasound contrast imaging sequence, perform head-end confidence assessment on the candidate features to obtain the head-end confidence assessment result; S6: Perform time window consistency adjudication analysis on the disambiguation judgment result and the head-end positioning information evaluation result respectively to obtain the adjudication result; S7: Based on the aforementioned decision, the position of the jejunal proximal end and the trajectory of the tube are obtained.

2. The ultrasound contrast-based precise positioning method for jejunum catheterization according to claim 1, characterized in that, The ultrasound contrast-enhanced data specifically includes real-time contrast-enhanced data, injection synchronization timing data, and historical reference data.

3. The method for precise positioning of jejunal tube placement based on contrast-enhanced ultrasound according to claim 2, characterized in that, S2 specifically includes: S201: The real-time imaging data is aligned using a phase correlation registration algorithm and a pyramid dense optical flow alignment algorithm to obtain an aligned image sequence; S202: The aligned image sequence is subjected to noise smoothing and edge fidelity preservation processing by nonlocal mean speckle denoising algorithm and anisotropic diffusion edge-preserving smoothing algorithm to obtain a denoised and enhanced image sequence. S203: The denoised and enhanced image sequence is subjected to contrast enhancement processing by using the quantile mapping normalization algorithm and the local contrast-limited histogram equalization algorithm to obtain the contrast-enhanced image sequence. S204: Generate injection event anchor points by combining the real-time angiography data and the injection synchronization timing data using a timestamp alignment algorithm; S205: Standardize and normalize the imaging-related parameters in the real-time angiography data and the injection synchronization time series data to obtain standardized imaging parameters; S206: The historical reference data is subjected to nonlinear compression processing using a logarithmic compression normalization algorithm to obtain standardized historical statistical data; S207: Integrate the contrast-enhanced image sequence, the injection event anchor point, the standardized imaging parameters, and the standardized historical statistical data to generate a standardized ultrasound contrast sequence.

4. The method for precise positioning of jejunal tube placement based on contrast-enhanced ultrasound according to claim 1, characterized in that, The process of generating candidate features specifically includes: The contrast image sequence in the standardized ultrasound contrast imaging sequence is processed by a multi-scale ridge enhancement algorithm and a thin connected component extraction algorithm to generate a candidate thin strip set. For each candidate strip in the candidate strip set, calculate the skeleton-related features and the direction change sequence to obtain the basic features of the candidate strip; A strip sampling window is constructed along the skeleton normal, centered on the set of skeleton pixels of the candidate thin strip; Based on the strip sampling window, the median of the grayscale image sequence in the standardized ultrasound contrast sequence is taken frame by frame to obtain the single-frame grayscale response value. Based on the frame timestamp sequence in the standardized ultrasound contrast imaging sequence, the grayscale response values ​​of the single frame are sorted to obtain the grayscale response sequence. The median of each frame of the contrast-enhanced ultrasound image sequence in the standardized ultrasound contrast-enhanced sequence is taken to obtain the single-frame contrast-enhanced response value. Based on the overall gain data in the standardized ultrasound contrast sequence, the quantile mapping correction is performed on the single-frame contrast effect value. Based on the mechanical index data in the standardized ultrasound contrast sequence, the single-frame contrast effect value after quantile mapping correction is subjected to exponential compression correction. Based on the frame timestamp sequence in the standardized ultrasound contrast imaging sequence, the single-frame contrast response values ​​after exponential compression correction are sorted to obtain the candidate thin strip contrast intensity time series. Based on the optical flow field output by the pyramid dense optical flow alignment algorithm, the optical flow gradient norm sequence is calculated for the neighborhood of the candidate thin strip; The candidate features are obtained by combining the grayscale response sequence, the candidate fine strip imaging intensity time series, and the optical flow gradient norm sequence.

5. The method for precise positioning of jejunal tube placement based on contrast-enhanced ultrasound according to claim 1, characterized in that, The process of constructing non-tube volume prior constraints specifically includes: For the historical contrast-enhanced ultrasound sequence and the historical grayscale sequence in the standardized ultrasound contrast-enhanced sequence, a candidate fine strip generation process consistent with the real-time process is executed to obtain a set of historical fine strip features. By combining the historical fine strip feature set with the number of historical fine strips and the density of historical artifacts in the standardized ultrasound contrast imaging sequence, a non-tube strong echo sample set is constructed. Based on the non-tube strong echo sample set, the non-tube prior constraints are formed.

6. The method for precise positioning of jejunal tube placement based on contrast-enhanced ultrasound according to claim 1, characterized in that, S4 specifically includes: S401: Using a kernel density estimation algorithm, calculate the historical occurrence density of the corresponding feature vectors for the candidate thin strip feature sequences in the candidate features to obtain the candidate trajectory density; S402: By using the kernel density estimation algorithm, combining the historical thin strip feature set in the candidate features with the non-tube body prior constraints, the non-tube body distribution density is obtained; S403: Calculate the density of the candidate trajectory and the distribution density of the non-tube body to obtain the density logarithmic ratio term; S404: Process the time series of the imaging intensity of the candidate thin strips in the candidate features to obtain the injection time sequence deviation distance; S405: Calculate the distance between the density logarithmic ratio term and the injection timing deviation to obtain the corrected density logarithmic ratio value; S406: Process the optical flow gradient norm sequence in the standardized ultrasound contrast sequence to obtain the trajectory neighborhood motion disturbance intensity and noise correction denominator. S407: Calculate the motion disturbance intensity in the trajectory neighborhood and the noise correction denominator term to obtain the disturbance noise index suppression term; S408: Calculate the logarithmic ratio of the correction density and the disturbance noise index suppression term to obtain the disambiguation discrimination result.

7. The method for precise positioning of jejunal tube placement based on contrast-enhanced ultrasound according to claim 1, characterized in that, S5 specifically includes: S501: Define the set of candidate strips in the candidate features as the tube body travel input set; S502: Based on the contrast image sequence of the standardized ultrasound contrast sequence, perform cross-frame correlation processing on the tube travel input set to obtain the candidate thin strip end point sequence; S503: Determine the end of the tube trajectory based on the sequence of candidate thin strip end points; S504: The enhanced distribution entropy is obtained by calculating the end of the pipe trajectory using an information entropy calculation algorithm; S505: Based on the contrast image sequence of the standardized ultrasound contrast imaging sequence, perform clustering and position analysis on the end of the tube trajectory to obtain the fluctuation of the cluster center position and the end alignment distance; S506: Calculate the enhanced distribution entropy to obtain the entropy exponential suppression term; S507: Calculate the fluctuation of the aggregation center position to obtain the fluctuation reciprocal suppression term; S508: Calculate the end alignment distance to obtain the alignment index suppression term; S509: Calculate the entropy index suppression term, the inverse fluctuation suppression term, and the alignment index suppression term to obtain the head-end confidence assessment result.

8. The method for precise positioning of jejunal tube placement based on contrast-enhanced ultrasound according to claim 1, characterized in that, S6 specifically includes: S601: Based on the frame timestamp sequence of the standardized ultrasound contrast imaging sequence, the temporal consistency deviation is obtained by calculating the candidate head position sequence within adjacent time windows through the position distribution difference calculation algorithm. S602: Process the echo disambiguation discriminant value in the disambiguation result to obtain the echo disambiguation logarithmic enhancement term; S603: Process the head-end positioning confidence value in the head-end confidence assessment result to obtain the head-end positioning log enhancement term; S604: Calculate the logarithmic enhancement term for echo disambiguation and the logarithmic enhancement term for head-end positioning to obtain a joint logarithmic enhancement product term; S605: Calculate the time consistency deviation to obtain the consistency deviation denominator; S606: Calculate the joint logarithmic enhancement product term and the consistency deviation denominator term to obtain the positioning and issuance value; S607: Use the location issuance value as the adjudication result obtained from the time window consistency adjudication analysis.

9. The method for precise positioning of jejunal tube placement based on contrast-enhanced ultrasound according to claim 1, characterized in that, Specifically, S7 is: Determine whether the location issuance value in the ruling result is less than the location issuance threshold; if so, maintain the frozen mark of the non-tube strong echo set and review the candidate head position set until the location issuance threshold is met, and output the jejunal head position and the tube trajectory; otherwise, directly output the jejunal head position and the tube trajectory, and keep the frozen mark of the non-tube strong echo set valid.

10. A precise positioning system for jejunal tube placement based on contrast-enhanced ultrasound, characterized in that, include: processor; A memory storing computer-readable instructions, which, when executed by the processor, implement the precise positioning method for jejunal tube placement based on ultrasound contrast imaging as described in any one of claims 1 to 9.

Images