Image-based tumor curative effect prediction method and system

Abstract

The invention relates to the technical field of medical image analysis, in particular to an image-based tumor curative effect prediction method and system, and the method comprises the following steps: based on CT image data of multiple time points before and after treatment, extracting inter-frame pixel position coordinates of the CT image data, calculating an offset vector of an inter-frame pixel position, and implementing inter-frame position calibration; and generating time sequence image correction data. According to the method, accurate calibration of inter-frame positions is achieved through inter-frame pixel position coordinate extraction and offset vector calculation of multi-time-point CT image data, and through a set expansion threshold value and area boundary fitting, the outer contour of a tumor area is recognized, and a pixel point coordinate set of the tumor area boundary is dynamically analyzed, so that accurate calibration of the tumor area is achieved. According to the method, the position change of the tumor in the time sequence is evaluated, key data and density distribution characteristic analysis are provided for predicting the treatment response of the tumor, evaluation of the structural uniformity of the tumor area is supplemented, a basis is provided for individualized treatment prediction, and the change trend of tumor treatment can be effectively predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of medical image analysis, and particularly to an image-based tumor treatment efficacy prediction method and system. Background Art

[0002] The technical field of medical image analysis involves the processes of collecting, analyzing, and interpreting medical image data, aiming to support clinical decision-making and medical research. In this field, professionals use various imaging techniques, such as X-rays, MRI, CT, and ultrasound, etc., to obtain image data under pathological conditions. Through the analysis of image data, medical imaging technology can help doctors diagnose diseases more accurately, evaluate treatment effects, and monitor the disease process.

[0003] Among them, the tumor treatment efficacy prediction method uses the processed and analyzed medical image data to predict the responsiveness of tumor treatment and the patient's status. The technical matters involved in this topic include using image processing techniques and analysis methods to evaluate the tumor characteristics in the images. The methods generally include steps such as image segmentation, image enhancement, feature recognition, and image quality assessment. Through technical processing, quantitative information that has a decisive impact on predicting treatment effects can be extracted from medical images.

[0004] Traditional prediction methods have obvious deficiencies in tracking the dynamic changes of tumors during the treatment process. Especially in terms of the accuracy and real-time nature of tumor treatment effect prediction and monitoring, traditional methods lack sufficient automation and precision in dealing with the inter-frame pixel position offset and tumor region boundary recognition. They often rely on manual intervention for correction and boundary determination, which increases the workload, introduces subjective errors, and limits the optimization of diagnosis and treatment plans. In the continuous monitoring of tumor morphological changes, traditional methods fail to provide sufficient quantitative tools to accurately measure the expansion rate of tumor boundaries and the complexity of structures, which limits the real-time and dynamic assessment of treatment responses, and may thus lead to delays in the best timing for adjusting treatment plans. Summary of the Invention

[0005] The purpose of the present invention is to solve the drawbacks existing in the prior art, and to propose an image-based tumor treatment efficacy prediction method and system.

[0006] To achieve the above purpose, the present invention adopts the following technical scheme: An image-based tumor treatment efficacy prediction method, including the following steps:

[0007] S1: Based on the CT image data at multiple time points before and after treatment, extract the inter-frame pixel position coordinates of the CT image data, calculate the offset vector of the inter-frame pixel positions, perform inter-frame position calibration, and generate time series image correction data;

[0008] S2: Based on the time series image correction data, select the first frame image in the image sequence, calibrate the tumor seed points in the first frame image, and extract the outer contour of the tumor region point by point by calculating the intensity difference and gradient change between the seed points and the neighboring pixels, generating tumor region boundary data;

[0009] S3: Based on the tumor region boundary data, calculate the position change amount of each pixel point in the time series, compare the differences in the boundary coordinates between adjacent frames, and combine the frame sequence change amount to cumulatively analyze the overall change trend, generating tumor boundary dynamic change data;

[0010] S4: Based on the tumor boundary dynamic change data, calculate the gradient direction and gradient magnitude change of each pixel point, statistically analyze the variance of the gradient distribution within the target region, and judge the uniformity of the internal structure of the region through the consistency of the gradient change direction of the pixel points, analyzing the density difference of the target region to obtain density distribution characteristic data;

[0011] S5: Based on the tumor boundary dynamic change data and density distribution characteristic data, analyze the changes of the tumor region over time, calculate the tumor volume expansion rate, the change amplitude of the boundary complexity, and the consistency of the density gradient change direction frame by frame, and combine the change trend in the time dimension to extract the quantitative indicators of the regional morphology and structure changes, predicting the changes in the tumor treatment effect in the future period to obtain the dynamic characteristic indicators of the treatment effect prediction.

[0012] As a further solution of the present invention, based on the CT image data at multiple time points before and after treatment, extracting the inter-frame pixel position coordinates of the CT image data, calculating the offset vector of the inter-frame pixel positions, and implementing inter-frame position calibration, the steps of generating the time series image correction data are specifically as follows:

[0013] S101: Based on the CT image data at multiple time points before and after treatment, extract the pixel position coordinates in each frame of the image, extract the corresponding displacement vector data by calculating the spatial position difference of each pixel point between frames, and perform cumulative processing on the displacement vectors of each pixel point to obtain the inter-frame pixel offset vector data;

[0014] S102: Based on the inter-frame pixel offset vector data, adjust the positions of the pixel points in each frame of the image, correct the actual positions of the pixel points by accumulating the offset vectors, and rearrange each frame of the image according to the adjusted positions of the pixels to obtain a position-adjusted image sequence;

[0015] S103: Based on the position-adjusted image sequence, extract the boundary feature points in each frame of the image, compare the position differences of the adjacent frame feature points point by point, adjust the inter-frame alignment error by calculating the average displacement of the feature points, and perform global alignment processing on the boundary feature points of each frame in the image sequence to obtain the time series image correction data.

[0016] As a further solution of the present invention, based on the corrected data of the time-series images, the first frame image in the image sequence is selected, the tumor seed points in the first frame image are calibrated, and by calculating the intensity difference and gradient change between the seed points and the neighboring pixels, the outer contour of the tumor region is extracted point by point, and the steps of generating the boundary data of the tumor region are specifically as follows:

[0017] S201: Based on the corrected data of the time-series images, the first frame image in the image sequence is selected, the center point of the tumor region in the image is calibrated as the seed point, the neighboring pixels around the seed point are scanned point by point, and by calculating the gray intensity difference between the seed point and the neighboring pixels, the neighboring intensity difference data is generated;

[0018] S202: Based on the neighboring intensity difference data, the neighboring scanning range is gradually expanded, the gray gradient change value in the expanded region is calculated, and the set gradient change threshold is used as the expansion stop condition, the expansion range is adjusted, and the critical value of the gray gradient change is matched to obtain the gradient data of the expanded region;

[0019] S203: Based on the gradient data of the expanded region, the boundary pixels of the expanded region are extracted, the boundary pixel points are fitted and analyzed, and by determining the attribution relationship of the boundary pixel points, the outer contour of the tumor region is constructed point by point to obtain the boundary data of the tumor region.

[0020] As a further solution of the present invention, based on the boundary data of the tumor region, the position change amount of each pixel point in the time series is calculated, the difference between the boundary coordinates in adjacent frames is compared, and combined with the frame sequence change amount, the overall change trend is cumulatively analyzed, and the steps of generating the dynamic change data of the tumor boundary are specifically as follows:

[0021] S301: Based on the boundary data of the tumor region, the pixel point coordinates of the tumor boundary in the image are extracted frame by frame, the position difference of each pixel point in adjacent frames is calculated, and by accumulating the position change amount of adjacent frames, the cumulative change amount of the pixel points is statistically obtained to obtain the boundary pixel position change data;

[0022] S302: Based on the boundary pixel position change data, the change amount of the tumor boundary in each frame is statistically analyzed, the expansion rate of the boundary coordinate change is analyzed, and by accumulating the expansion rate in the frame sequence, the boundary expansion trend is analyzed to obtain the boundary expansion rate data;

[0023] S303: Based on the boundary expansion rate data, the complexity difference between the boundary coordinates of adjacent frames is compared frame by frame, and by statistically analyzing the difference in coordinate changes of each frame, the overall change trend is calculated to generate the dynamic change data of the tumor boundary.

[0024] As a further solution of the present invention, the formula for calculating the overall change trend is:

[0025]

[0026] Among them, S is the cumulative trend value of the boundary expansion rate, and w i represents the weighting coefficient of the i-th frame, and C i and C i-1 represent the boundary coordinate sets of the i-th frame and the (i - 1)-th frame respectively, E i and E i-1 represent the boundary expansion rates of the i-th frame and the (i - 1)-th frame respectively, T i and T i-1 represent the timestamps of the i-th frame and the (i - 1)-th frame respectively, α is the expansion rate adjustment parameter, β is the time influence adjustment parameter, and n represents the total number of image frames.

[0027] As a further solution of the present invention, based on the dynamic change data of the tumor boundary, calculating the gradient direction and gradient magnitude change of each pixel point, statistically analyzing the variance of the gradient distribution in the target area, and judging the uniformity of the internal structure of the area through the consistency of the gradient change direction of the pixel points, and analyzing the density difference of the target area to obtain the density distribution characteristic data specifically includes the following steps:

[0028] S401: Based on the dynamic change data of the tumor boundary, extract the gray values in the tumor area pixel by pixel, calculate the gradient direction of each pixel point, and obtain the pixel gradient direction change data by comparing the change amount of the pixel gradient direction point by point;

[0029] S402: Based on the pixel gradient direction change data, statistically analyze the local consistency and distribution difference of the gradient direction in each area, calculate the consistency ratio of the local gradient direction, and analyze the uniformity of the internal structure of the tumor area through the change of the cumulative consistency ratio to obtain the gradient consistency distribution data;

[0030] S403: Based on the gradient consistency distribution data, extract the density gradient change range and amplitude in each area, statistically analyze the distribution characteristics of the density difference, and analyze the overall density change trend by combining the uniformity of the gradient distribution in the target area to obtain the density distribution characteristic data.

[0031] As a further solution of the present invention, based on the dynamic change data of the tumor boundary and the density distribution characteristic data, analyze the changes of the tumor area in the time series, calculate the tumor volume expansion rate, the change amplitude of the boundary complexity, and the consistency of the density gradient change direction frame by frame, combine the change trend in the time dimension, extract the quantitative indicators of the regional shape and structure changes, and predict the changes in the tumor treatment effect in the future period to obtain the steps of the dynamic characteristic indicators of the treatment effect prediction specifically include:

[0032] S501: Based on the tumor boundary dynamic change data and density distribution characteristic data, extract the pixel positions and density values within the tumor region frame by frame, calculate the change rate of the tumor volume in each frame of the image, compare the volume change amounts between adjacent frames, and statistically analyze the expansion trend in the time series to obtain the volume expansion rate data;

[0033] S502: Based on the volume expansion rate data, extract the coordinate sets of the tumor boundary frame by frame, calculate the change amount of the complexity of each frame boundary, and analyze the evolution characteristics of the tumor boundary complexity in the time dimension by statistically analyzing the cumulative trend of the complexity change in the time series to obtain the boundary complexity change data;

[0034] S503: Based on the boundary complexity change data and volume expansion rate data, analyze the time series trend of the density gradient change direction pixel by pixel, combine the volume expansion rate and boundary complexity change trends, extract the quantitative indicators of regional morphology and structure changes, analyze the tumor change indicators, predict the changes in the tumor treatment effect in the future period, and generate the dynamic characteristic indicators for efficacy prediction.

[0035] As a further solution of the present invention, the formula for analyzing the tumor change indicators is:

[0036]

[0037] where I is the tumor change quantification indicator, ΔD j represents the change amount of the boundary complexity at the j-th moment, ΔV j represents the change amount of the volume expansion rate at the j-th moment, w D is the weight coefficient of the boundary complexity change amount, w V is the weight coefficient of the volume expansion rate change amount, T j and T j-1 represent the timestamps at the j-th moment and the previous moment respectively, γ is the time smoothing parameter, and N is the total number of time points.

[0038] A tumor efficacy prediction system based on images, the tumor efficacy prediction system based on images is used to execute the above-mentioned tumor efficacy prediction method based on images, and the system includes:

[0039] The image sequence correction module, based on the CT image data at multiple time points before and after treatment, extracts the inter-frame pixel position coordinates of the CT image data, calculates the offset vector of the inter-frame pixel positions, performs inter-frame position calibration, and generates the time series image correction data;

[0040] Based on the time series image correction data, the boundary recognition module selects the first frame image in the image sequence, calibrates the tumor seed points in the first frame image, and extracts the outer contour of the tumor region point by point by calculating the intensity difference and gradient change between the seed points and the neighboring pixels, generating tumor region boundary data;

[0041] Based on the tumor region boundary data, the boundary change analysis module calculates the position change amount of each pixel point in the time series, compares the differences in boundary coordinates between adjacent frames, and combines the frame sequence change amount to cumulatively analyze the overall change trend, generating tumor boundary dynamic change data;

[0042] Based on the tumor boundary dynamic change data, the density distribution analysis module calculates the gradient direction and gradient amplitude change of each pixel point, statistically analyzes the gradient distribution variance in the target region, and determines the uniformity of the internal structure of the region through the consistency of the gradient change direction of the pixel points, analyzing the density difference in the target region to obtain density distribution characteristic data;

[0043] Based on the tumor boundary dynamic change data and density distribution characteristic data, the effect prediction analysis module analyzes the changes in the tumor region over time, calculates the tumor volume expansion rate, the change amplitude of the boundary complexity, and the consistency of the density gradient change direction frame by frame, combines the change trend in the time dimension, extracts the quantitative indicators of the regional morphology and structure changes, predicts the changes in the tumor treatment effect in the future period, and obtains the dynamic characteristic indicators of the treatment effect prediction.

[0044] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0045] In the present invention, through the extraction of the inter-frame pixel position coordinates and the calculation of the offset vector of the multi-time point CT image data, the precise calibration of the inter-frame position is realized, providing a more accurate benchmark for the image sequence, ensuring the consistency and alignment of the time series images, effectively improving the accuracy and efficiency of subsequent image processing. By setting the expansion threshold and region boundary fitting, the outer contour of the tumor region is recognized, and the set of pixel coordinates of the tumor region boundary is dynamically analyzed to evaluate its position change in the time series, promoting the precise evaluation of the tumor expansion rate and morphological changes, providing key data for predicting the treatment response of the tumor, making medical decisions more scientific and precise. The analysis of the density distribution characteristics supplements the evaluation of the structural uniformity of the tumor region, is crucial for understanding the biological characteristics of the tumor, provides a basis for individualized treatment prediction, and can effectively predict the change trend of tumor treatment. Description of the Drawings

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0047] Figure 1 It is a schematic diagram of the working process of the present invention;

[0048] Figure 2 It is a detailed flowchart of S1 of the present invention;

[0049] Figure 3 It is a detailed flowchart of S2 of the present invention;

[0050] Figure 4 It is a detailed flowchart of S3 of the present invention;

[0051] Figure 5 It is a detailed flowchart of S4 of the present invention;

[0052] Figure 6 It is a detailed flowchart of S5 of the present invention;

[0053] Figure 7 It is a system flowchart of the present invention. Specific Embodiments

[0054] The following will describe the technical solutions in the present invention in conjunction with the accompanying drawings.

[0055] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as an "example" in the present invention should not be construed as being more preferred or more advantageous than other embodiments or design solutions. Exactly speaking, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0056] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when their differences are not emphasized, their intended meanings are the same. "(of)", "corresponding", and "corresponding" can sometimes be used interchangeably. It should be noted that when their differences are not emphasized, their intended meanings are the same.

[0057] In the embodiments of the present invention, sometimes subscripts such as W1 may be written in a non-subscript form such as W1. When their differences are not emphasized, their intended meanings are the same.

[0058] To make the technical problems, technical solutions, and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0059] Please refer to Figure 1 , the present invention provides a technical solution: an image-based tumor treatment effect prediction method, including the following steps:

[0060] S1: Based on the CT image data at multiple time points before and after treatment, extract the inter-frame pixel position coordinates of the CT image data, calculate the offset vector of the inter-frame pixel positions, adjust the positions of each frame in the image sequence, align with the image boundary feature points, implement inter-frame position calibration, and generate time series image correction data;

[0061] S2: Based on the time series image correction data, select the first frame image in the image sequence, calibrate the tumor seed points in the first frame image, calculate the intensity difference and gradient change between the seed points and the neighboring pixels, define the expansion area by setting an expansion threshold, and perform region boundary fitting. Combining the attribution analysis of the region boundary pixels, extract the outer contour of the tumor region point by point to generate tumor region boundary data;

[0062] S3: Based on the tumor region boundary data, extract the set of pixel point coordinates of the tumor boundary in each frame of the image, calculate the position change amount of each pixel point in the time series, statistically analyze the expansion rate of the tumor region boundary and the change trend of each frame boundary coordinate, statistically compare the differences in the boundary coordinates between adjacent frames, and combine the frame sequence change amount to cumulatively analyze the overall change trend to generate tumor boundary dynamic change data;

[0063] S4: Based on the tumor boundary dynamic change data, analyze the gray scale distribution characteristics within the tumor region pixel by pixel, calculate the gradient direction and gradient amplitude change of each pixel point, statistically analyze the gradient distribution variance within the target region, and judge the uniformity of the internal structure of the region through the consistency of the gradient change direction of the pixel points. Combining the density gradient change range and amplitude, analyze the density difference of the target region to obtain density distribution characteristic data;

[0064] S5: Based on the tumor boundary dynamic change data and density distribution characteristic data, analyze the changes of the tumor region in the time series, calculate the tumor volume expansion rate, the change amplitude of the boundary complexity, and the consistency of the density gradient change direction frame by frame. Combining the change trend in the time dimension, analyze the dynamic characteristics of the tumor region, extract the quantitative indicators of the regional morphology and structure changes, and predict the changes in the tumor treatment effect in the future period to obtain the dynamic characteristic indicators for treatment effect prediction.

[0065] The time - series image correction data includes the corrected pixel spatial coordinates, the gray - scale intensity matching values between sequential frames, and the position consistency index after overall correction. The tumor region boundary data includes the coordinate set of boundary pixel points, the intensity difference value inside and outside the boundary, and the cumulative area of regional expansion. The tumor boundary dynamic change data includes the amount of boundary coordinate change, the expansion rate in the time dimension, and the complexity change value. The density distribution characteristic data includes the pixel gradient direction consistency value, the variance of gradient magnitude distribution, and the density change amplitude range. The dynamic characteristic index for efficacy prediction includes the volume expansion rate, the density gradient change rate, and the boundary complexity change trend.

[0066] Please refer to Figure 2 , based on the CT image data at multiple time points before and after treatment, extract the inter - frame pixel position coordinates of the CT image data, calculate the offset vector of the inter - frame pixel positions, adjust the position of each frame in the image sequence, align with the image boundary feature points, and implement inter - frame position calibration. The steps to generate the time - series image correction data are specifically as follows:

[0067] S101: Based on the CT image data at multiple time points before and after treatment, extract the pixel position coordinates in each frame of the image. By calculating the spatial position difference of each pixel point between frames, extract the corresponding displacement vector data, and perform cumulative processing on the displacement vectors of each pixel point to obtain the inter - frame pixel offset vector data;

[0068] Based on the CT image data at multiple time points before and after treatment, obtain the pixel coordinate positions in each frame of the CT image. Load the image data into an image - processing tool, such as ImageJ or MATLAB. Extract the pixel point positions through image segmentation and recognition algorithms. Automatically identify the coordinates of each pixel point through the change of image gray - scale values. After extraction, generate a data matrix composed of the number of frames and pixel positions. Then, calculate the spatial position difference of each pixel point between frames. Analyze the difference point - by - point using the Euclidean distance change formula of pixel coordinates between frames. By traversing the position changes of each pixel point, generate the corresponding displacement vector data. Perform cumulative processing on all displacement vectors. Obtain the overall offset vector data of inter - frame pixels through superposition. The data can be directly exported in vector - field representation for correcting the position changes in the image sequence.

[0069] S102: Based on the inter - frame pixel offset vector data, adjust the pixel point positions in each frame of the image. Correct the actual positions of the pixel points by accumulating the offset vectors, and rearrange each frame of the image according to the positions after pixel adjustment to obtain the position - adjusted image sequence;

[0070] Based on the inter-frame pixel offset vector data, adjust the positions of pixel points in each frame of the image. Normalize the offset vector data to make the offset value ranges consistent among different images. By analyzing the magnitude and direction of the offset vector of each pixel point, adjust the position of each pixel point in the corresponding frame of the image point by point, including starting from the first frame, accumulating the offset vectors and correcting frame by frame to form an adjusted position mapping matrix. At the same time, reapply the corrected coordinates to each frame of the image. The adjusted image sequence needs to be verified by visual comparison to ensure that the geometric position changes of each frame of the image conform to the actual observation situation, and generate an image sequence with adjusted positions.

[0071] S103: Based on the image sequence with adjusted positions, extract the boundary feature points in each frame of the image, compare the position differences of adjacent frame feature points point by point, adjust the inter-frame alignment error by calculating the average displacement of the feature points, and perform global alignment processing on the boundary feature points of each frame in the image sequence to obtain time series image correction data;

[0072] Based on the image sequence with adjusted positions, it is necessary to extract the boundary feature points in each frame of the image, including applying edge detection algorithms such as Canny or Sobel, locating the edge pixel points by calculating the gradient change of each frame of the image, analyzing the coordinates of the extracted edge feature points, comparing the position changes of adjacent frame feature points one by one, calculating the average displacement of each feature point through an algorithm and adjusting the inter-frame alignment error, adjusting the positions of the global boundary feature points of each frame in the image sequence by accumulating the vector field to ensure the global alignment of the feature points. At the same time, perform a global consistency check on the adjusted data, and use image stitching software or programming tools to output the corrected time series image data to form time series image correction data.

[0073] Please refer to Figure 3 , based on the time series image correction data, select the first frame image in the image sequence, calibrate the tumor seed points in the first frame image, by calculating the intensity difference and gradient change between the seed points and the neighborhood pixels, set an expansion threshold to limit the expansion area, and perform regional boundary fitting. Combining the attribution analysis of the regional boundary pixels, the steps to extract the outer contour of the tumor area point by point and generate the tumor area boundary data are as follows:

[0074] S201: Based on the time series image correction data, select the first frame image in the image sequence, calibrate the center point of the tumor area in the image as the seed point, scan the neighborhood pixels around the seed point point by point, and generate neighborhood intensity difference data by calculating the gray intensity difference between the seed point and the neighborhood pixels;

[0075] Based on the time-series image correction data, select the first frame image in the image sequence, calibrate the center point of the tumor area in the image as the seed point, load the first frame image through an image processing tool, select the center point of the tumor as the seed point by clicking with the mouse or using an automatic segmentation algorithm, then extract the neighborhood pixel data around the seed point, adopt a neighborhood scanning algorithm to check the gray values of the surrounding pixel points one by one, generate neighborhood intensity difference data by calculating the gray intensity difference between the seed point and each surrounding pixel point. The process can be completed based on the gray difference calculation formula, and corresponding calculations are performed on each pixel point to generate neighborhood intensity difference data.

[0076] S202: Based on the neighborhood intensity difference data, gradually expand the neighborhood scanning range, calculate the gray gradient change value within the expanded area, and use the set gradient change threshold as the expansion stop condition to adjust the expansion range and match the critical value of the gray gradient change to obtain the expanded area gradient data;

[0077] Based on the neighborhood intensity difference data, gradually expand the neighborhood scanning range, expand the neighborhood by setting an initial neighborhood radius and an increment step size, calculate the gray gradient change value within the expanded area, perform point-by-point calculations on the pixel gradient changes in the expanded area using the gray gradient change formula, set the gradient change threshold as the expansion stop condition, and stop expanding when the gray gradient change value in the expanded neighborhood is lower than the set threshold. Dynamically adjust the expansion range to match the critical value of the gray gradient change to obtain the gradient change data of the expanded area, which indicates the possible position of the boundary of the tumor area and provides a data basis for contour fitting.

[0078] S203: Based on the expanded area gradient data, extract the boundary pixels of the expanded area, perform fitting analysis on the boundary pixel points, and gradually construct the outer contour of the tumor area by determining the attribution relationship of the boundary pixel points to obtain the tumor area boundary data;

[0079] Based on the expanded area gradient data, extract the boundary pixels of the expanded area, apply edge detection algorithms such as Canny or Sobel algorithms to locate the boundary pixel points, perform fitting analysis on the extracted boundary pixel points one by one, construct a fitting curve based on the coordinate data of the boundary pixel points, optimize the boundary fitting result by determining the attribution relationship of the boundary pixel points, gradually construct the outer contour of the tumor area, and at the same time smooth the fitted boundary to reduce mutation points or noise interference to obtain the tumor area boundary data, providing complete boundary information for subsequent data analysis and visualization.

[0080] Please refer to Figure 4, based on the tumor region boundary data, extract the set of pixel coordinates of the tumor boundary in the image frame by frame, calculate the position change amount of each pixel in the time series, statistically analyze the expansion rate of the tumor region boundary and the change trend of the boundary coordinates in each frame, statistically compare the differences in the boundary coordinates between adjacent frames, and combine the frame sequence change amount to cumulatively analyze the overall change trend. The specific steps for generating the dynamic change data of the tumor boundary are as follows:

[0081] S301: Based on the tumor region boundary data, extract the pixel coordinates of the tumor boundary in the image frame by frame, calculate the position difference of each pixel in adjacent frames, and obtain the boundary pixel position change data by accumulating the position change amounts of adjacent frames.

[0082] Based on the tumor region boundary data, extract the pixel coordinates of the tumor boundary in the image frame by frame. By reading the boundary data of each frame of the image, parse and extract the two-dimensional space coordinates of the boundary pixel points, calculate the position difference of each pixel in adjacent frames frame by frame, and use the difference calculation method to obtain the displacement of each pixel, including extracting the X and Y coordinates of the same pixel in adjacent frames, calculating the magnitude and direction of the displacement vector, and accumulating the difference values of each pixel to form a statistical result, so as to obtain the boundary pixel position change data, and the data is finally saved in the form of a list or matrix.

[0083] S302: Based on the boundary pixel position change data, statistically analyze the change amount of the tumor boundary in each frame, analyze the expansion rate of the boundary coordinate change, and analyze the boundary expansion trend by accumulating the expansion rates in the frame sequence to obtain the boundary expansion rate data.

[0084] Based on the boundary pixel position change data, statistically analyze the change amount of the tumor boundary in each frame. By classifying the pixel position change data within each frame sequence, integrate the boundary change amounts of adjacent frames into the statistical process of the inter-frame expansion rate. The specific method for analyzing the expansion rate of the boundary coordinate change is to perform time-step calculation on the cumulative position change value, and evaluate the overall change trend by accumulating the specific values of the expansion rates in the frame sequence to obtain the boundary expansion rate data, which represents the dynamic expansion of the boundary of the tumor region and is used for subsequent dynamic characteristic analysis.

[0085] S303: Based on the boundary expansion rate data, compare the complexity differences of the boundary coordinates of adjacent frames frame by frame, calculate the overall change trend by statistically analyzing the differences in the coordinate changes of each frame, and generate the dynamic change data of the tumor boundary.

[0086] Based on the boundary expansion rate data, compare the complexity differences of the boundary coordinates of adjacent frames frame by frame, extract the boundary coordinate data of each frame, analyze its geometric complexity, including the length change, curvature change of the boundary lines, and the uniformity of pixel distribution, count the coordinate change data of each frame, calculate the overall change trend, which is used to describe the overall change trend. Through the analysis of the complexity data, generate the dynamic change data of the tumor boundary. The data is the basis for completely recording the boundary changes and can intuitively display the dynamic characteristics of tumor expansion.

[0087] The formula for calculating the overall change trend is:

[0088]

[0089] where S is the cumulative trend value of the boundary expansion rate, w i represents the weighting coefficient of the i-th frame, C i and C i-1 represent the boundary coordinate sets of the i-th frame and the (i - 1)-th frame respectively, E i and E i-1 represent the boundary expansion rates of the i-th frame and the (i - 1)-th frame respectively, T i and T i-1 represent the timestamps of the i-th frame and the (i - 1)-th frame respectively, α is the expansion rate adjustment parameter, β is the time influence adjustment parameter, and n represents the total number of image frames.

[0090] Formula:

[0091]

[0092] Parameter meanings and acquisition methods:

[0093] w i : The weighting coefficient of the i-th frame, which reflects the importance of different frames in the overall analysis. For example, frames closer in time may be given higher weights because their data is more relevant. These weights can be obtained through expert input or automatically calculated based on previous analysis results.

[0094] C i and C i-1 : The boundary coordinate sets of the i-th frame and the (i - 1)-th frame, extracted from image processing algorithms (such as Canny edge detection).

[0095] E i and E i-1 : The boundary expansion rate, which represents the rate of boundary change between two consecutive frames and can be calculated by measuring the change in the boundary area (such as area or perimeter).

[0096] T i and T i-1: Timestamp, indicating the specific time when the image frame was recorded. Usually obtained directly from the metadata of the image file or the image acquisition system.

[0097] α and β: Adjustment parameters, fixed values obtained from experiments or actual observations, used to adjust the weights in the calculation of the expansion rate and time influence. Usually, the parameters are best estimated through an optimization algorithm in preliminary experiments.

[0098] Calculation example

[0099] Suppose there are two frames of image data as follows:

[0100] Data of the first frame:

[0101] Weight w 1 = 1.0, boundary coordinates C 1 = {(10, 20), (15, 25)}, boundary expansion rate E 1 = 0.0 (initial frame), timestamp T 1 = 100 (seconds).

[0102] Data of the second frame:

[0103] Weight w 2 = 1.5, boundary coordinates C 2 = {(12, 22), (17, 27)}, boundary expansion rate E 2 = 0.4, timestamp T 2 = 101 (seconds),

[0104] Use the Euclidean distance to calculate the sum of the distances between all corresponding points in the two frames:

[0105]

[0106] Calculate the time difference T 2 -T 1 : 101 - 100 = 1 second;

[0107] Substitute the values into the formula:

[0108]

[0109] The calculation result S represents the change in the tumor boundary from the first frame to the second frame. Considering the combined effects of weight, expansion rate, and time difference, this quantitative analysis is very useful for understanding the growth rate and dynamic changes of the tumor.

[0110] Please refer to Figure 5, based on the dynamic change data of the tumor boundary, analyze the gray-scale distribution characteristics within the tumor region pixel by pixel, calculate the gradient direction and gradient magnitude change of each pixel point, statistically analyze the variance of the gradient distribution within the target region, and determine the uniformity of the internal structure of the region through the consistency of the gradient change direction of the pixel points. By combining the density gradient change range and magnitude, analyze the density difference of the target region. The specific steps for obtaining the density distribution characteristic data are as follows:

[0111] S401: Based on the dynamic change data of the tumor boundary, extract the gray-scale values within the tumor region pixel by pixel, calculate the gradient direction of each pixel point, and obtain the pixel gradient direction change data by comparing the change amount of the pixel gradient direction point by point.

[0112] Based on the dynamic change data of the tumor boundary, extract the gray-scale values within the tumor region pixel by pixel. Load the gray-scale matrix data of the tumor region using an image processing tool. By scanning each pixel point one by one, extract the gray-scale value of each pixel. Subsequently, calculate the gradient direction of the pixel point using a gradient direction calculation method, such as determining the gradient direction of each point through the gradient calculation formula of the pixel gray-scale value change. Compare the change amount of the pixel gradient direction point by point. By calculating the angular difference of the pixel gradient direction within the spatial neighborhood range, statistically obtain the pixel gradient direction change data, which can be saved as a gradient direction change matrix for subsequent analysis.

[0113] S402: Based on the pixel gradient direction change data, statistically analyze the local consistency and distribution difference of the gradient direction region by region, calculate the consistency ratio of the local gradient direction, and analyze the uniformity of the internal structure of the tumor region through the change of the cumulative consistency ratio to obtain the gradient consistency distribution data.

[0114] Based on the pixel gradient direction change data, statistically analyze the local consistency and distribution difference of the gradient direction region by region, including grouping the pixel gradient direction data within each region, calculating the local consistency ratio by statistically analyzing the number of pixel points in the same gradient direction, using the angular distribution analysis method to evaluate the distribution characteristics of the gradient direction within the target region, gradually accumulating the change of the consistency ratio, and performing spatio-temporal analysis on the gradient consistency within the tumor region. By quantitatively evaluating the dynamic change trend of the gradient consistency ratio, form the gradient consistency distribution data.

[0115] S403: Based on the gradient consistency distribution data, extract the density gradient change range and magnitude region by region, statistically analyze the distribution characteristics of the density difference, and analyze the overall density change trend by combining the uniformity of the gradient distribution within the target region to obtain the density distribution characteristic data.

[0116] Based on the data with gradient consistency distribution, extract the range and amplitude of density gradient change region by region. Analyze the range of density difference by extracting the gradient consistency distribution data. Adopt the statistical method of pixel gradient density within the region to gradually count the characteristics of density difference distribution. Combine the uniformity of gradient distribution within the target region, analyze the gradient difference of density change point by point, and count the density gradient change trend in different regions to complete the generation of the overall density distribution characteristic data. The data is finally used as the basis for the dynamic characteristic analysis of density distribution and for subsequent processing.

[0117] Please refer to Figure 6 , based on the dynamic change data of the tumor boundary and the density distribution characteristic data, analyze the changes of the tumor region in the time series. Calculate the consistency of the tumor volume expansion rate, the change amplitude of the boundary complexity, and the density gradient change direction frame by frame. Combine the change trend in the time dimension, analyze the dynamic characteristics of the tumor region, extract the quantitative indicators of the regional morphology and structure changes, and predict the changes in the tumor treatment effect in the future period. The steps to obtain the dynamic characteristic indicators for efficacy prediction are specifically as follows:

[0118] S501: Based on the dynamic change data of the tumor boundary and the density distribution characteristic data, extract the pixel positions and density values within the tumor region frame by frame. Calculate the change rate of the tumor volume in each frame of the image, compare the volume change amounts of adjacent frames, and count the expansion trend in the time series to obtain the volume expansion rate data;

[0119] Based on the dynamic change data of the tumor boundary and the density distribution characteristic data, extract the pixel positions and density values within the tumor region frame by frame. After processing each frame of the image sequence, extract the tumor region in the image, read the gray value of each pixel point, and combine its position coordinates to construct a data matrix containing spatial positions and gray values. Subsequently, calculate the change rate of the tumor volume in each frame of the image, perform differential processing on the frame-by-frame volume data using the three-dimensional reconstruction method to obtain the volume change rate, compare the volume change amounts of adjacent frames, analyze the expansion trend in the time series by means of cumulative volume change, count and plot the volume expansion rate data to obtain the form of a time series curve.

[0120] S502: Based on the volume expansion rate data, extract the coordinate set of the tumor boundary frame by frame. Calculate the change amount of the complexity of each frame boundary, analyze the evolution characteristics of the tumor boundary complexity in the time dimension by counting the cumulative trend of complexity change in the time series to obtain the boundary complexity change data;

[0121] Based on the volume expansion rate data, extract the coordinate set of the tumor boundary frame by frame, analyze the complexity of each frame boundary. By statistically analyzing indicators such as the curvature change of the tumor boundary, the uniformity of point distribution, and the contour length, calculate the complexity change amount of each frame boundary. Perform time accumulation processing on the calculation results, draw the cumulative trend graph of the complexity change in the time series, analyze the evolution characteristics of the tumor boundary complexity in the time dimension, and quantify the characteristics into a numerical set to obtain the boundary complexity change data.

[0122] S503: Based on the boundary complexity change data and the volume expansion rate data, analyze the time series trend of the density gradient change direction pixel by pixel. Combine the volume expansion rate and the boundary complexity change trend, extract the quantitative indicators of the regional morphology and structure change, analyze the tumor change indicators, predict the change of the tumor treatment effect in the future period, and generate the dynamic characteristic indicators of the treatment effect prediction;

[0123] Based on the boundary complexity change data and the volume expansion rate data, analyze the time series trend of the density gradient change direction pixel by pixel. Analyze the density gradient direction of each pixel frame by frame. By comparing the density gradient change angles and amplitudes of adjacent frames in the time series, accumulate the dynamic changes of each pixel point to the global region. Combine the volume expansion rate and the boundary complexity change trend, analyze the tumor change indicators, analyze the tumor change trend through the statistical law of the density distribution within the region, and use the quantitative indicators to predict the tumor treatment effect in the future period and generate the dynamic characteristic indicators of the treatment effect prediction.

[0124] The formula for analyzing the tumor change indicators is:

[0125]

[0126] where I is the tumor change quantitative indicator, ΔD j represents the boundary complexity change amount at the j-th moment, ΔV j represents the volume expansion rate change amount at the j-th moment, w D is the weight coefficient of the boundary complexity change amount, w V is the weight coefficient of the volume expansion rate change amount, T j and T j-1 represent the time stamps at the j-th moment and the previous moment respectively, γ is the time smoothing parameter, and N is the total number of time points.

[0127] Formula:

[0128]

[0129] Parameter meanings and acquisition methods:

[0130] ΔD j:The change in boundary complexity at the j-th moment, representing the change in the complexity of the boundary structure from the previous moment to the current moment. The boundary complexity is extracted through an image analysis algorithm (such as fractal dimension calculation), and then the difference between adjacent time points is taken.

[0131] ΔV j :The change in the volume expansion rate at the j-th moment, representing the change in the tumor volume growth rate from the previous moment to the current moment. By measuring the tumor volume in consecutive frames, the difference in the volume change rate is calculated.

[0132] w D and w V :Weight coefficients, respectively used to adjust the influence of boundary complexity and volume expansion rate on the total index. The value range is usually determined through experimental optimization. Here, let w D = 0.6, w V = 0.4.

[0133] T j and T j-1 :The timestamps at the j-th and j-1 moments, directly obtained from the image data acquisition time, usually in days.

[0134] γ: The time smoothing parameter, used to prevent the denominator value from being too small due to too small a time interval. In this example, it is set to γ = 0.05.

[0135] Calculation example

[0136] Suppose the data for two time points are as follows:

[0137] The first time point: Timestamp: T 1 = 0 days, boundary complexity: D 1 = 1.2 (set value), volume: V 1 = 1000 cubic millimeters, the second time point: Timestamp: T 2 = 7 days, boundary complexity: D 2 = 1.5 (set value), volume: V 2 = 1200 cubic millimeters.

[0138] Calculate the change in boundary complexity ΔD 2 :

[0139] ΔD 2 = D 2 - D 1 = 1.5 - 1.2 = 0.3;

[0140] Calculate the change in volume expansion rate ΔV 2 :

[0141] Volume expansion rate:

[0142]

[0143] Change in volume expansion rate (relative to initial):

[0144] ΔV 2 = E 2 - 0 = 28.57 - 0 = 28.57 mm³ / day;

[0145] Calculation time interval:

[0146] |T 2 - T 1 | = |7 - 0| = 7 days;

[0147] Substitute into the formula to calculate the comprehensive morphological index I:

[0148] The formula is as follows:

[0149]

[0150] The obtained index I ≈ 4.37, representing the comprehensive quantification result combining the boundary complexity change and the volume expansion rate change, used to evaluate the dynamic change trend of the tumor. This value can be used as a dynamic characteristic index for predicting the treatment effect, compared with the index values at subsequent time points to form a trend analysis.

[0151] Please refer to Figure 7 , the image-based tumor treatment effect prediction system, which is used to execute the above-mentioned image-based tumor treatment effect prediction method. The system includes:

[0152] The image sequence correction module extracts the inter-frame pixel position coordinates of the CT image data based on the CT image data at multiple time points before and after treatment, calculates the offset vector of the inter-frame pixel positions, implements inter-frame position calibration, and generates time series image correction data;

[0153] The boundary recognition module selects the first frame image in the image sequence based on the time series image correction data, calibrates the tumor seed points in the first frame image, and extracts the outer contour of the tumor region point by point by calculating the intensity difference and gradient change between the seed points and the neighboring pixels, generating tumor region boundary data;

[0154] The boundary change analysis module calculates the position change amount of each pixel point in the time series based on the tumor region boundary data, compares the differences in the boundary coordinates between adjacent frames, combines the frame sequence change amount, and accumulatively analyzes the overall change trend to generate tumor boundary dynamic change data;

[0155] Based on the dynamic change data of the tumor boundary, the density distribution analysis module calculates the gradient direction and gradient magnitude change of each pixel point, statistically analyzes the variance of the gradient distribution within the target area, and determines the uniformity of the internal structure of the area through the consistency of the gradient change direction of the pixel points, analyzes the density difference of the target area, and obtains the density distribution characteristic data;

[0156] Based on the dynamic change data of the tumor boundary and the density distribution characteristic data, the effect prediction analysis module analyzes the changes of the tumor area in the time series, calculates the consistency of the tumor volume expansion rate, the change amplitude of the boundary complexity, and the gradient change direction of the density frame by frame, combines the change trend in the time dimension, extracts the quantitative indicators of the regional morphology and structure changes, predicts the changes in the tumor treatment effect in the future period, and obtains the dynamic characteristic indicators of the treatment effect prediction.

[0157] It should be understood that the term "and / or" in this article is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood by referring to the context before and after.

[0158] In the present invention, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following items" or its similar expressions refer to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0159] It should be understood that in various embodiments of the present invention, the magnitudes of the serial numbers of the above processes do not mean the sequence of execution. The execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0160] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0161] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the devices, apparatuses, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0162] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the couplings, direct couplings, or communication connections shown or discussed with each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0163] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0164] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0165] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention. The foregoing storage medium includes: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0166] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claimed rights.

Claims

1. A method for predicting tumor efficacy based on imaging, characterized in that: The following steps are involved: S1: Based on the CT image data at multiple time points before and after treatment, the inter-frame pixel position coordinates of the CT image data are extracted, the offset vector of the inter-frame pixel position is calculated, the inter-frame position calibration is performed, and the time series image correction data is generated; S2: Based on the time series image correction data, select the first frame image in the image sequence, calibrate the tumor seed point in the first frame image, and extract the outer contour of the tumor area point by point by calculating the intensity difference and gradient change between the seed point and the neighboring pixels to generate tumor area boundary data; S3: Based on the tumor region boundary data, calculate the position change of each pixel point in the time series, compare the difference of boundary coordinates in adjacent frames, combine the frame sequence change, cumulatively analyze the overall change trend, and generate tumor boundary dynamic change data; S4: Based on the tumor boundary dynamic change data, the gradient direction and gradient amplitude change of each pixel point are calculated, the gradient distribution variance in the target area is counted, and the uniformity of the internal structure of the region is judged by the consistency of the gradient change direction of the pixel point, and the density difference of the target area is analyzed to obtain density distribution characteristic data; S5: Based on the dynamic change data of the tumor boundary and the density distribution characteristic data, the changes in the tumor area in the time series are analyzed, and the consistency of the tumor volume expansion rate, the boundary complexity change amplitude and the density gradient change direction are calculated frame by frame. Combined with the change trend in the time dimension, quantitative indicators of regional morphology and structural changes are extracted to predict the changes in tumor treatment effects in future time periods, and obtain dynamic characteristic indicators for efficacy prediction.

2. The method for predicting tumor efficacy based on imaging according to claim 1, characterized in that: The time series image correction data includes the corrected pixel space coordinates, the grayscale intensity matching value between sequence frames and the overall corrected position consistency index; the tumor area boundary data includes the coordinate set of boundary pixel points, the intensity difference value inside and outside the boundary and the cumulative area of ​​regional expansion; the tumor boundary dynamic change data includes the boundary coordinate change amount, the expansion rate and complexity change value in the time dimension; the density distribution characteristic data includes the pixel gradient direction consistency value, the gradient amplitude distribution variance and the density change amplitude range; the dynamic characteristic indicators for efficacy prediction include the volume expansion rate, the density gradient change rate and the boundary complexity change trend.

3. The method for predicting tumor efficacy based on imaging according to claim 1, characterized in that: Based on the CT image data at multiple time points before and after treatment, the steps of extracting the inter-frame pixel position coordinates of the CT image data, calculating the offset vector of the inter-frame pixel position, and implementing inter-frame position calibration to generate time series image correction data are as follows: S101: Based on the CT image data at multiple time points before and after treatment, the pixel position coordinates in each frame of the image are extracted, the spatial position difference of each pixel point between frames is calculated, the corresponding displacement vector data is extracted, and the displacement vector of each pixel point is accumulated to obtain the pixel offset vector data between frames; S102: adjusting the position of the pixels in each frame of the image based on the inter-frame pixel offset vector data, correcting the actual position of the pixels by accumulating the offset vectors, and rearranging each frame of the image according to the pixel adjusted position to obtain a position adjusted image sequence; S103: Based on the position adjustment image sequence, the boundary feature points in each frame of the image are extracted, the position differences of the feature points of adjacent frames are compared point by point, the average displacement of the feature points is calculated, the inter-frame alignment error is adjusted, and the boundary feature points of each frame in the image sequence are globally aligned to obtain time series image correction data.

4. The method for predicting tumor efficacy based on imaging according to claim 1, characterized in that: Based on the time series image correction data, the first frame image in the image sequence is selected, the tumor seed point in the first frame image is calibrated, and the outer contour of the tumor area is extracted point by point by calculating the intensity difference and gradient change between the seed point and the neighboring pixels. The steps of generating the tumor area boundary data are specifically as follows: S201: Based on the time series image correction data, select the first frame image in the image sequence, mark the center point of the tumor area in the image as a seed point, scan the neighborhood pixels around the seed point point by point, and generate neighborhood intensity difference data by calculating the gray intensity difference between the seed point and the neighborhood pixels; S202: based on the neighborhood intensity difference data, gradually expand the neighborhood scanning range, calculate the grayscale gradient change value in the expanded area, and use the set gradient change threshold as the expansion stop condition, adjust the expansion range, match the critical value of the grayscale gradient change, and obtain the expanded area gradient data; S203: Based on the extended area gradient data, extract boundary pixels of the extended area, perform fitting analysis on the boundary pixel points, and construct the outer contour of the tumor area point by point by determining the attribution relationship of the boundary pixel points to obtain tumor area boundary data.

5. The method for predicting tumor efficacy based on imaging according to claim 1, characterized in that: Based on the tumor region boundary data, the position change of each pixel point in the time series is calculated, the difference of boundary coordinates in adjacent frames is compared, and the overall change trend is cumulatively analyzed in combination with the frame sequence change to generate the tumor boundary dynamic change data in the following steps: S301: Based on the tumor region boundary data, extract the pixel coordinates of the tumor boundary in the image frame by frame, calculate the position difference of each pixel in adjacent frames, accumulate the position changes of adjacent frames, count the cumulative changes of the pixel points, and obtain the boundary pixel position change data; S302: Based on the boundary pixel position change data, the change amount of the tumor boundary in each frame is counted, the expansion rate of the boundary coordinate change is analyzed, and the boundary expansion trend is analyzed by accumulating the expansion rate in the frame sequence to obtain the boundary expansion rate data; S303: Based on the boundary expansion rate data, the complexity difference of the boundary coordinates of adjacent frames is compared frame by frame, and the overall change trend is calculated by counting the difference of the coordinate changes of each frame to generate tumor boundary dynamic change data.

6. The method for predicting tumor efficacy based on imaging according to claim 5, characterized in that: The formula for calculating the overall change trend is: Among them, S is the cumulative trend value of the boundary expansion rate, w i represents the weighting coefficient of the i-th frame, C i and C i-1 Represent the boundary coordinate sets of the i-th frame and the i-1-th frame respectively, E i and E i-1 Represent the boundary expansion rate of the i-th frame and the i-1-th frame, T i and T i-1 They represent the timestamps of the i-th frame and the i-1-th frame respectively, α is the expansion rate adjustment parameter, β is the time impact adjustment parameter, and n represents the total number of image frames.

7. The method for predicting tumor efficacy based on imaging according to claim 1, characterized in that: Based on the tumor boundary dynamic change data, the gradient direction and gradient amplitude change of each pixel point are calculated, the gradient distribution variance in the target area is counted, and the uniformity of the internal structure of the region is judged by the consistency of the gradient change direction of the pixel point, and the density difference of the target area is analyzed to obtain the density distribution characteristic data. Specifically, the steps are as follows: S401: Based on the tumor boundary dynamic change data, the grayscale value in the tumor area is extracted pixel by pixel, the gradient direction of each pixel is calculated, and the pixel gradient direction change data is obtained by comparing the change amount of the pixel gradient direction point by point; S402: Based on the pixel gradient direction change data, local consistency and distribution difference of gradient direction are counted region by region, and the consistency ratio of local gradient direction is calculated. By accumulating the change of consistency ratio, the uniformity of the internal structure of the tumor region is analyzed to obtain gradient consistency distribution data; S403: Based on the gradient consistency distribution data, the density gradient variation range and amplitude are extracted region by region, the distribution characteristics of the density difference are statistically analyzed, and the overall density variation trend is analyzed by combining the uniformity of the gradient distribution in the target area to obtain the density distribution characteristic data.

8. The method for predicting tumor efficacy based on imaging according to claim 1, characterized in that: Based on the tumor boundary dynamic change data and density distribution characteristic data, the changes in the tumor area in the time series are analyzed, the tumor volume expansion rate, the consistency of the boundary complexity change amplitude and the density gradient change direction are calculated frame by frame, and the quantitative indicators of regional morphology and structural changes are extracted in combination with the change trend in the time dimension to predict the changes in tumor treatment effects in future periods. The specific steps of obtaining the dynamic characteristic indicators for efficacy prediction are as follows: S501: Based on the tumor boundary dynamic change data and density distribution characteristic data, pixel positions and density values ​​in the tumor area are extracted frame by frame, the change rate of the tumor volume in each frame of the image is calculated, the volume changes of adjacent frames are compared, and the expansion trend in the time series is counted to obtain volume expansion rate data; S502: extracting a coordinate set of the tumor boundary frame by frame based on the volume expansion rate data, calculating the complexity change of the boundary of each frame, analyzing the evolution characteristics of the complexity of the tumor boundary in the time dimension by statistically analyzing the cumulative trend of the complexity change in the time series, and obtaining boundary complexity change data; S503: Based on the boundary complexity change data and volume expansion rate data, analyze the time series trend of the density gradient change direction pixel by pixel, combine the volume expansion rate and boundary complexity change trend, extract quantitative indicators of regional morphology and structure changes, analyze tumor change indicators, predict changes in tumor treatment effects in future time periods, and generate dynamic characteristic indicators for efficacy prediction.

9. The method for predicting tumor efficacy based on imaging according to claim 8, characterized in that: The formula for analyzing tumor change indicators is: Among them, I is the quantitative index of tumor change, ΔD j Represents the change in boundary complexity at the jth moment, ΔV j represents the volume expansion rate change at the jth moment, w D is the weight coefficient of the boundary complexity change, w V is the weight coefficient of the volume expansion rate change, T h and T j-1 Represent the timestamp of the jth moment and the previous moment respectively, γ is the time smoothing parameter, and N is the total number of time points.

10. An imaging-based tumor efficacy prediction system, characterized in that: According to any one of claims 1 to 9, the method for predicting tumor efficacy based on imaging comprises: The image sequence correction module extracts the inter-frame pixel position coordinates of the CT image data based on the CT image data at multiple time points before and after treatment, calculates the offset vector of the inter-frame pixel position, implements inter-frame position calibration, and generates time series image correction data; The boundary recognition module selects the first frame image in the image sequence based on the time series image correction data, calibrates the tumor seed point in the first frame image, and extracts the outer contour of the tumor area point by point by calculating the intensity difference and gradient change between the seed point and the neighboring pixels to generate tumor area boundary data; The boundary change analysis module calculates the position change of each pixel point in the time series based on the tumor area boundary data, compares the difference of boundary coordinates in adjacent frames, combines the frame sequence change, accumulates and analyzes the overall change trend, and generates tumor boundary dynamic change data; The density distribution analysis module calculates the gradient direction and gradient amplitude changes of each pixel point based on the dynamic change data of the tumor boundary, and counts the gradient distribution variance in the target area. It also determines the uniformity of the internal structure of the region through the consistency of the gradient change direction of the pixel points, analyzes the density difference of the target area, and obtains the density distribution characteristic data; The effect prediction and analysis module analyzes the changes in the tumor area in the time series based on the dynamic change data of the tumor boundary and the density distribution characteristic data, calculates the consistency of the tumor volume expansion rate, the boundary complexity change amplitude and the density gradient change direction frame by frame, and extracts quantitative indicators of regional morphology and structural changes in combination with the change trend in the time dimension, predicts the changes in tumor treatment effects in future time periods, and obtains dynamic characteristic indicators for efficacy prediction.

Images