Image feature extraction method after hepatic arterial infusion

Abstract

This invention discloses a method for extracting image features after hepatic artery perfusion surgery, belonging to the field of image feature extraction technology, and includes the following steps: obtaining real-time change values ​​based on images before and after hepatic artery perfusion to be detected; obtaining historical perfusion change values ​​based on normal images before and after hepatic artery perfusion in a database; obtaining a perfusion change threshold based on historical perfusion change values; obtaining a real-time initial region based on real-time change values ​​and perfusion change threshold; obtaining historical qualified values ​​based on historical perfusion change values; obtaining a noise region threshold based on historical qualified values; and obtaining the contour of the perfusion region based on the real-time initial region and noise region threshold. This invention addresses the problem that existing image feature extraction technologies struggle to distinguish between noise points and small target regions in images, resulting in low accuracy in image feature extraction.

Description

Technical Field

[0001] This invention relates to the field of image feature extraction technology, specifically a method for extracting image features after hepatic artery perfusion surgery. Background Technology

[0002] Hepatic artery infusion chemotherapy (HAIC) is a core minimally invasive interventional treatment for intermediate and advanced liver cancer, liver cancer with portal vein tumor thrombus, etc. It has become an important clinical treatment for unresectable liver cancer due to its advantages such as high local drug concentration in the tumor, low systemic toxicity and side effects, and repeatability. The efficacy evaluation of HAIC and the determination of surgical infusion success rate directly determine the adjustment of subsequent treatment plans. Digital subtraction angiography is the core means of postoperative evaluation. The information it contains, such as tumor staining, drug concentration, and vascular visualization, is the key basis for reflecting the infusion effect and tumor efficacy.

[0003] Traditional feature extraction methods rely heavily on subjective human analysis, which is not only inefficient and highly subjective, but also greatly affected by differences in physician experience, making it difficult to achieve standardization and reproducibility of feature extraction. At the same time, postoperative images may contain respiratory motion artifacts, resulting in image discontinuity. This discontinuity can lead to small target areas, making it difficult to distinguish between noise points and small target areas in the image, thus affecting the accuracy of perfusion effect and efficacy assessment. Existing image feature extraction technologies are unable to distinguish between noise points and small target areas in the image, resulting in low accuracy of image feature extraction. Summary of the Invention

[0004] This invention aims to at least partially solve one of the technical problems in the prior art by obtaining real-time change values ​​based on images before and after hepatic artery perfusion surgery; obtaining historical perfusion change values ​​based on normal images before and after hepatic artery perfusion surgery in a database; obtaining a perfusion change threshold based on historical perfusion change values; obtaining a real-time initial region based on real-time change values ​​and perfusion change threshold; obtaining historical qualified values ​​based on historical perfusion change values; obtaining a noise region threshold based on historical qualified values; and obtaining the contour of the perfusion region based on the real-time initial region and noise region threshold. This addresses the problem that existing image feature extraction techniques struggle to distinguish between noise points and small target regions in images, resulting in low accuracy in image feature extraction.

[0005] To achieve the above objectives, this application provides a method for extracting imaging features after hepatic artery perfusion surgery, comprising the following steps:

[0006] Real-time change values ​​are obtained based on images of the hepatic artery perfusion procedure before and after the procedure.

[0007] Historical perfusion change values ​​were obtained from normal hepatic artery perfusion images before and after surgery in the database.

[0008] The perfusion change threshold is obtained based on historical perfusion change values;

[0009] The real-time initial region is obtained based on real-time change values ​​and perfusion change thresholds;

[0010] Historical qualified values ​​are obtained based on historical perfusion variation values;

[0011] The threshold for the noise region is obtained based on historical qualified values;

[0012] The contour of the infusion region is obtained based on the real-time initial region and noise region thresholds.

[0013] Furthermore, obtaining real-time change values ​​based on the images of the hepatic artery perfusion procedure before and after the procedure includes the following sub-steps:

[0014] The preoperative images of hepatic artery perfusion to be tested are marked as real-time pre-perfusion images;

[0015] The images of hepatic artery perfusion to be tested are labeled as real-time post-perfusion images;

[0016] Obtain the absolute value of the difference in grayscale values ​​of pixels at the same location in the image after real-time perfusion and the image before real-time perfusion, and mark it as the real-time change value.

[0017] Furthermore, obtaining historical perfusion change values ​​based on normal hepatic artery perfusion images before and after surgery from the database includes the following sub-steps:

[0018] Normal preoperative hepatic artery perfusion images in the database are marked as historical preoperative perfusion images;

[0019] The post-perfusion hepatic artery images corresponding to the historical pre-perfusion images are marked as historical post-perfusion images;

[0020] Obtain the absolute value of the difference in grayscale values ​​of pixels at the same location in the image after historical perfusion and the image before historical perfusion, and mark it as the historical change value;

[0021] Obtain the historical change values ​​of the completed irrigation area and mark them as historical irrigation change values.

[0022] Furthermore, obtaining the perfusion change threshold based on historical perfusion change values ​​includes the following sub-steps:

[0023] Obtain the first number of historical perfusion change values;

[0024] A Cartesian coordinate system is established with historical perfusion change values ​​as the horizontal axis data and the number of historical perfusion change values ​​as the vertical axis data, and this system is marked as the first data coordinate system.

[0025] Obtain all historical perfusion change values ​​and their corresponding quantities as coordinate points on the x and y axes, and mark them as the first data coordinate points;

[0026] Plot all the first data coordinate points in the first data coordinate system.

[0027] Furthermore, obtaining the perfusion change threshold based on historical perfusion change values ​​also includes the following sub-steps:

[0028] Get the maximum value of the ordinate among all the first data coordinate points and mark it as the first data height;

[0029] Set a length and mark it as the first set length;

[0030] Create a rectangle on the horizontal axis of the first data coordinate point with a height of the first data height and a width of the first set length, and be able to move left and right. Mark it as the first judgment rectangle.

[0031] The total number of historical perfusion change values ​​corresponding to the first data coordinate points included within the first judgment rectangle is marked as the first judgment quantity;

[0032] The range of historical perfusion variation values ​​is marked as the first data range;

[0033] Assuming that the historical perfusion change values ​​are uniformly distributed within the first data range, the first judgment quantity at this time is obtained and marked as the first average quantity;

[0034] Set a first percentage value, obtain the product of the first average quantity and the first percentage value, and mark it as the first quantity threshold;

[0035] In the first data coordinate system, the first judgment rectangle is shifted to the right starting from the leftmost historical perfusion change value. When the first judgment quantity is greater than or equal to the first quantity threshold, the movement of the first judgment rectangle is stopped. The historical perfusion change value corresponding to the smallest horizontal coordinate of the first judgment rectangle at this time is obtained and marked as the perfusion change threshold.

[0036] Furthermore, obtaining the real-time initial region based on real-time change values ​​and perfusion change thresholds includes the following sub-steps:

[0037] Obtain real-time change values ​​that are greater than or equal to the perfusion change threshold and mark them as real-time perfusion change values;

[0038] Obtain the pixels in the image after real-time perfusion change value and mark them as real-time perfusion pixels;

[0039] Obtain the independent region composed of real-time infused pixel groups and mark it as the real-time initial region.

[0040] Furthermore, obtaining historical qualified values ​​based on historical perfusion change values ​​includes the following sub-steps:

[0041] Obtain the historical perfusion change values ​​in the image after historical perfusion and mark them as historical perfusion pixels;

[0042] Obtain the independent region composed of historical infused pixels and mark it as the historical infused region;

[0043] Obtain the number of pixels within the historical infusion area and mark them as historical qualified values.

[0044] Furthermore, obtaining the noise region threshold based on historical qualified values ​​includes the following sub-steps:

[0045] Obtain the second number of historical qualified values;

[0046] Establish a Cartesian coordinate system with historical qualified values ​​as the horizontal axis data and the number of historical qualified values ​​as the vertical axis data, and mark it as the second data coordinate system;

[0047] Obtain all historical qualified values ​​and their corresponding quantities as coordinate points on the x and y axes, and mark them as the second data coordinate points;

[0048] Plot all the second data coordinate points in the second data coordinate system.

[0049] Furthermore, obtaining the noise region threshold based on historical qualified values ​​also includes the following sub-steps:

[0050] Get the maximum value of the ordinate among all the second data coordinate points and mark it as the second data height;

[0051] Set a length and mark it as the second set length;

[0052] Create a rectangle on the horizontal axis of the second data coordinate point with a height equal to the second data height and a width equal to the second set length, and allow it to move left and right. Mark this rectangle as the second judgment rectangle.

[0053] The total number of historical qualified values ​​corresponding to the second data coordinate points included within the second judgment rectangle is marked as the second judgment quantity;

[0054] Mark the range of historical qualified values ​​as the second data range;

[0055] Assuming that the historical qualified values ​​are evenly distributed within the second data range, obtain the second judgment quantity at this time and mark it as the second average quantity;

[0056] Set a second percentage value, obtain the product of the second average quantity and the second percentage value, and mark it as the second quantity threshold;

[0057] In the second data coordinate system, the second judgment rectangle is shifted to the right starting from the leftmost historical qualified value. When the second judgment quantity is greater than or equal to the second quantity threshold, the movement of the second judgment rectangle is stopped. The historical qualified value corresponding to the minimum horizontal coordinate of the second judgment rectangle at this time is obtained and marked as the noise area threshold.

[0058] Furthermore, obtaining the contour of the perfusion region based on the real-time initial region and noise region thresholds includes the following sub-steps:

[0059] If the real-time initial quantity is less than the noise region threshold, the real-time initial region is identified as a noise region; if the real-time initial quantity is greater than or equal to the noise region threshold, the outline of the real-time initial region is obtained and regarded as the outline of the infusion region.

[0060] The beneficial effects of this invention are as follows: This invention obtains real-time change values ​​based on images before and after hepatic artery perfusion surgery; obtains historical perfusion change values ​​based on normal images before and after hepatic artery perfusion surgery in a database; obtains a perfusion change threshold based on historical perfusion change values; obtains a real-time initial region based on real-time change values ​​and perfusion change threshold; obtains historical qualified values ​​based on historical perfusion change values; obtains a noise region threshold based on historical qualified values; and obtains the contour of the perfusion region based on the real-time initial region and noise region threshold. The advantage lies in distinguishing noise points and small target regions in the image, thus improving the accuracy of image feature extraction.

[0061] This invention obtains historical qualified values ​​based on historical perfusion change values. Its advantage lies in distinguishing noise points and small target areas in the image by using historical qualified values, thereby improving the accuracy of image feature extraction. Attached Figure Description

[0062] Figure 1 This is a flowchart of the steps of the method of the present invention;

[0063] Figure 2 This is a schematic diagram of the historical perfusion variation values ​​of the present invention;

[0064] Figure 3 This is a schematic diagram of the noise region threshold of the present invention. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0066] Example 1, please refer to Figure 1 As shown, this application provides a method for extracting imaging features after hepatic artery perfusion surgery, including the following steps:

[0067] Step S1: Obtain real-time change values ​​based on the images of the hepatic artery perfusion procedure before and after the procedure; Step S1 includes the following sub-steps:

[0068] Step S101: Mark the preoperative image of hepatic artery perfusion to be detected as a real-time pre-perfusion image;

[0069] Step S102: Mark the post-hepatic artery perfusion images to be detected as real-time post-perfusion images; acquire real-time pre-perfusion images and real-time post-perfusion images through imaging.

[0070] Step S103: Obtain the absolute value of the difference in grayscale values ​​of pixels at the same position in the image after real-time perfusion and the image before real-time perfusion, and mark it as the real-time change value; based on angiography technology, the perfused area will appear, causing the grayscale value to change, so the real-time change value can be used to determine whether it is a perfused area.

[0071] Step S2 involves obtaining historical perfusion change values ​​based on normal pre- and post-operative images of hepatic artery perfusion from the database. Step S2 includes the following sub-steps:

[0072] Step S201: Mark normal preoperative images of hepatic artery perfusion in the database as historical preoperative images;

[0073] Step S202: Mark the post-perfusion image of the hepatic artery corresponding to the historical pre-perfusion image as the historical post-perfusion image;

[0074] Step S203: Obtain the absolute value of the difference in grayscale values ​​of pixels at the same position in the image after historical perfusion and the image before historical perfusion, and mark it as the historical change value;

[0075] Step S204: Obtain the historical change values ​​of the completed irrigation area and mark them as historical irrigation change values.

[0076] Step S3: Obtain the perfusion change threshold based on historical perfusion change values; Step S3 includes the following sub-steps:

[0077] Step S301: Obtain a first number of historical perfusion change values; in order to obtain the range of historical perfusion change values, the first number cannot be too small, for example, the first number is 1000;

[0078] Step S302: Establish a Cartesian coordinate system with historical perfusion change values ​​as the horizontal axis data and the number of historical perfusion change values ​​as the vertical axis data, and mark it as the first data coordinate system;

[0079] Step S303: Obtain all historical perfusion change values ​​and their corresponding quantities, and mark them as the first data coordinate points.

[0080] Step S304: Plot all the first data coordinate points in the first data coordinate system.

[0081] Step S305: Obtain the maximum value of the ordinate among all first data coordinate points and mark it as the first data height;

[0082] Step S306: Set a length, marked as the first set length; the first set length should not be too large, but should be smaller than the first data range, so as to facilitate the analysis of the range of historical perfusion change values, for example, the first set length is 1;

[0083] Step S307: Create a rectangle on the horizontal axis of the first data coordinate point with a height of the first data height and a width of the first set length, and be able to move left and right, and mark it as the first judgment rectangle;

[0084] For practical applications, please refer to Figure 2 As shown, the first data height is 60, the first set length is 1, and the first judgment rectangle is drawn.

[0085] Step S308: Mark the total number of historical perfusion change values ​​corresponding to the first data coordinate points included within the first judgment rectangle as the first judgment quantity;

[0086] Step S309: Mark the range of historical perfusion change values ​​as the first data range;

[0087] Step S310: Assuming that the historical perfusion change values ​​are uniformly distributed within the first data range, obtain the first judgment quantity at this time and mark it as the first average quantity;

[0088] For practical applications, please refer to Figure 2 As shown, the first data range is 40, and the first average quantity is 1000 × (1 ÷ 40) = 25; where 1000 is the first quantity, 1 is the first set length, and 40 is the first data range.

[0089] Step S311: Set a first proportion value, obtain the product of the first average quantity and the first proportion value, and mark it as the first quantity threshold; In order to observe the area with a small number of historical perfusion change values, the first quantity threshold should not be set too large, that is, the first proportion value should not be set too large, for example, the first proportion value is 0.2.

[0090] In practical applications, the first quantity threshold is: 25 × 0.2 = 5.

[0091] Step S312: In the first data coordinate system, the first judgment rectangle is shifted to the right from the leftmost historical perfusion change value. When the first judgment quantity is greater than or equal to the first quantity threshold, the movement of the first judgment rectangle is stopped. The historical perfusion change value corresponding to the smallest horizontal coordinate of the first judgment rectangle at this time is obtained and marked as the perfusion change threshold. In order to exclude excessively small historical perfusion change values ​​and obtain a more accurate minimum value of historical perfusion change value.

[0092] In practical applications, in the first data coordinate system, the first judgment rectangle is shifted to the right starting from the leftmost historical perfusion change value. The shift stops when the first judgment quantity is 6, which is greater than the first quantity threshold. (See also...) Figure 2 As shown, the position where the first judgment rectangle stops is obtained. The historical perfusion change value corresponding to the smallest horizontal coordinate of the first judgment rectangle at this time is 31. Then the perfusion change threshold is 31.

[0093] Step S4: Obtain the real-time initial region based on the real-time change value and the perfusion change threshold; Step S4 includes the following sub-steps:

[0094] Step S401: Obtain real-time change values ​​that are greater than or equal to the perfusion change threshold and mark them as real-time perfusion change values;

[0095] In practical applications, real-time change values ​​greater than or equal to 31 are recorded as real-time perfusion change values.

[0096] Step S402: Obtain the pixel points of the real-time perfusion change value in the image after real-time perfusion and mark them as real-time perfusion pixels;

[0097] Step S403: Obtain the independent region of the real-time injected pixel group and mark it as the real-time initial region; the real-time initial region is the region that conforms to the change of grayscale value after injection.

[0098] Step S5: Obtain historical qualified values ​​based on historical perfusion change values; Step S5 includes the following sub-steps:

[0099] Step S501: Obtain the pixels in the image after historical perfusion change value and mark them as historical perfusion pixels;

[0100] Step S502: Obtain an independent region composed of historical infused pixels and mark it as a historical infused region;

[0101] Step S503: Obtain the number of pixels in the historical infusion area and mark it as a historical qualified value.

[0102] Step S6: Obtain the noise region threshold based on historical qualified values; Step S6 includes the following sub-steps:

[0103] Step S601: Obtain a second number of historical qualified values; in order to obtain the range of historical qualified values, for example, the second number is 2000;

[0104] Step S602: Establish a Cartesian coordinate system with historical qualified values ​​as the horizontal axis data and the number of historical qualified values ​​as the vertical axis data, and mark it as the second data coordinate system;

[0105] Step S603: Obtain all historical qualified values ​​and their corresponding quantities, with the coordinates of the x-axis and y-axis respectively, and mark them as the second data coordinate points; the second data coordinate points are used to observe the range of historical qualified values.

[0106] Step S604: Plot all the second data coordinate points in the second data coordinate system;

[0107] Step S605: Obtain the maximum value of the ordinate among all second data coordinate points and mark it as the second data height; please refer to [link to relevant documentation]. Figure 3 As shown, the height of the second data obtained is 53;

[0108] Step S606: Set a length, marked as the second set length; the second set length is set in order to establish the second judgment rectangle, so the second set length should not be too large, so as to facilitate the analysis of the distribution of historical qualified values;

[0109] Step S607: Create a rectangle on the horizontal axis of the second data coordinate point with a height of the second data height and a width of the second set length, and mark it as the second judgment rectangle.

[0110] For practical applications, please refer to Figure 3 As shown, a second judgment rectangle with a height of 53 and a second set length of 1 is created on the horizontal axis of the second data coordinate point, and can be moved left and right.

[0111] Step S608: Mark the total number of historical qualified values ​​corresponding to the second data coordinate points included within the second judgment rectangle as the second judgment quantity;

[0112] Step S609: Mark the range of historical qualified values ​​as the second data range;

[0113] Step S610: Assuming that the historical qualified values ​​are evenly distributed within the second data range, obtain the second judgment quantity at this time and mark it as the second average quantity;

[0114] For practical applications, please refer to Figure 3 As shown, the second data range is 40, and the first average quantity is 2000 × (1 ÷ 40) = 5; where 2000 is the second quantity, 2 is the second set length, and 40 is the second data range.

[0115] Step S611: Set a second percentage value, obtain the product of the second average quantity and the second percentage value, and mark it as the second quantity threshold; in order to observe the area with a small number of historical qualified values, the second quantity threshold should not be set too large, that is, the second percentage value should not be set too large, for example, the first percentage value is 0.1;

[0116] In practical applications, the first quantity threshold is: 5 × 0.1 = 5.

[0117] Step S612: In the second data coordinate system, the second judgment rectangle is moved to the right from the leftmost historical qualified value. When the second judgment quantity is greater than or equal to the second quantity threshold, the movement of the second judgment rectangle is stopped. The historical qualified value corresponding to the smallest horizontal coordinate of the second judgment rectangle at this time is obtained and marked as the noise area threshold. In order to exclude excessively small historical qualified values ​​and obtain a more accurate minimum value of historical qualified values.

[0118] In practical applications, in the second data coordinate system, the second judgment rectangle is shifted to the right starting from the leftmost historical qualified value. The movement of the second judgment rectangle stops when the second judgment quantity is 7, which is greater than the second quantity threshold of 5. (See also...) Figure 3 As shown, the stopping position of the second judgment rectangle is obtained. The historical qualified value corresponding to the minimum horizontal coordinate of the second judgment rectangle at this time is 51, so the noise area threshold is 51.

[0119] Step S7: Obtain the contour of the perfusion region based on the real-time initial region and noise region thresholds; Step S7 includes the following sub-steps:

[0120] Step S701: If the real-time initial number is less than the noise region threshold, the real-time initial region is identified as a noise region; if the real-time initial number is greater than or equal to the noise region threshold, the contour of the real-time initial region is obtained and regarded as the contour of the infusion region; the number of pixels in the noise point is generally small, that is, if the real-time initial number is less than the noise region threshold, the real-time initial region is identified as a noise region.

[0121] In practical applications, for example, if the initial real-time quantity is 30, and the initial real-time quantity of 30 is less than the noise region threshold of 51, the initial real-time region is identified as a noise region, and the noise region does not need to display its outline.

[0122] Example 2: This application also provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, steps such as those in the method for extracting image features after hepatic artery perfusion are performed to achieve the following functions: obtaining real-time change values ​​based on images before and after hepatic artery perfusion to be detected; obtaining historical perfusion change values ​​based on normal images before and after hepatic artery perfusion in the database; obtaining a perfusion change threshold based on historical perfusion change values; obtaining a real-time initial region based on real-time change values ​​and perfusion change threshold; obtaining historical qualified values ​​based on historical perfusion change values; obtaining a noise region threshold based on historical qualified values; and obtaining the contour of the perfusion region based on the real-time initial region and noise region threshold.

[0123] Furthermore, when the logical instructions in the aforementioned memory can be 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 application, in essence, or the part that contributes to the prior art, or a portion 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 application. 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.

[0124] Example 3: This application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the postoperative image feature extraction method for hepatic artery perfusion provided by the above methods. The method includes: obtaining real-time change values ​​based on images of hepatic artery perfusion before and after the procedure to be detected; obtaining historical perfusion change values ​​based on normal images of hepatic artery perfusion before and after the procedure in a database; obtaining a perfusion change threshold based on the historical perfusion change values; obtaining a real-time initial region based on the real-time change values ​​and the perfusion change threshold; obtaining historical qualified values ​​based on the historical perfusion change values; obtaining a noise region threshold based on the historical qualified values; and obtaining the contour of the perfusion region based on the real-time initial region and the noise region threshold.

[0125] Example 4: This application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it performs the steps of the above-described method for extracting postoperative image features of hepatic artery perfusion to achieve the following functions: obtaining real-time change values ​​based on images of hepatic artery perfusion before and after the procedure to be detected; obtaining historical perfusion change values ​​based on normal images of hepatic artery perfusion before and after the procedure in a database; obtaining a perfusion change threshold based on historical perfusion change values; obtaining a real-time initial region based on the real-time change values ​​and the perfusion change threshold; obtaining historical qualified values ​​based on historical perfusion change values; obtaining a noise region threshold based on historical qualified values; and obtaining the contour of the perfusion region based on the real-time initial region and the noise region threshold.

[0126] Based on the above description of the embodiments, the embodiments of the present invention can be provided as methods, systems, or computer program products. Based on this understanding, the above technical solutions, in essence or in terms of their contribution to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or certain parts of the embodiments.

[0127] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces. The indirect coupling or communication connection between systems, modules, and units may be electrical, mechanical, or other forms.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for extracting image features after hepatic arterial infusion, characterized by, Includes the following steps: Real-time change values ​​are obtained based on images of the hepatic artery perfusion procedure before and after the procedure. Historical perfusion change values ​​were obtained from normal hepatic artery perfusion images before and after surgery in the database. The perfusion change threshold is obtained based on historical perfusion change values; The real-time initial region is obtained based on real-time change values ​​and perfusion change thresholds; Historical qualified values ​​are obtained based on historical perfusion variation values; The threshold for the noise region is obtained based on historical qualified values; The contour of the infusion region is obtained based on the real-time initial region and noise region thresholds. 2.The method of claim 1, wherein, Obtaining real-time changes based on images before and after hepatic artery perfusion includes the following sub-steps: The preoperative images of hepatic artery perfusion to be tested are marked as real-time pre-perfusion images; The images of hepatic artery perfusion to be tested are labeled as real-time post-perfusion images; Obtain the absolute value of the difference in grayscale values ​​of pixels at the same location in the image after real-time perfusion and the image before real-time perfusion, and mark it as the real-time change value. 3.The method of claim 2, wherein, Obtaining historical perfusion change values ​​based on normal hepatic artery perfusion images before and after surgery from a database includes the following sub-steps: Normal preoperative hepatic artery perfusion images in the database are marked as historical preoperative perfusion images; The post-perfusion hepatic artery images corresponding to the historical pre-perfusion images are marked as historical post-perfusion images; Obtain the absolute value of the difference in grayscale values ​​of pixels at the same location in the image after historical perfusion and the image before historical perfusion, and mark it as the historical change value; Obtain the historical change values ​​of the completed irrigation area and mark them as historical irrigation change values. 4.The method of claim 3, wherein, Obtaining the perfusion change threshold based on historical perfusion change values ​​includes the following sub-steps: Obtain the first number of historical perfusion change values; A Cartesian coordinate system is established with historical perfusion change values ​​as the horizontal axis data and the number of historical perfusion change values ​​as the vertical axis data, and this system is marked as the first data coordinate system. Obtain all historical perfusion change values ​​and their corresponding quantities as coordinate points on the x and y axes, and mark them as the first data coordinate points; Plot all the first data coordinate points in the first data coordinate system. 5.The method of claim 4, wherein, Obtaining the perfusion change threshold based on historical perfusion change values ​​also includes the following sub-steps: Get the maximum value of the ordinate among all the first data coordinate points and mark it as the first data height; Set a length and mark it as the first set length; Create a rectangle on the horizontal axis of the first data coordinate point with a height of the first data height and a width of the first set length, and be able to move left and right. Mark it as the first judgment rectangle. The total number of historical perfusion change values ​​corresponding to the first data coordinate points included within the first judgment rectangle is marked as the first judgment quantity; The range of historical perfusion variation values ​​is marked as the first data range; Assuming that the historical perfusion change values ​​are uniformly distributed within the first data range, the first judgment quantity at this time is obtained and marked as the first average quantity; Set a first percentage value, obtain the product of the first average quantity and the first percentage value, and mark it as the first quantity threshold; In the first data coordinate system, the first judgment rectangle is shifted to the right starting from the leftmost historical perfusion change value. When the first judgment quantity is greater than or equal to the first quantity threshold, the movement of the first judgment rectangle is stopped. The historical perfusion change value corresponding to the smallest horizontal coordinate of the first judgment rectangle at this time is obtained and marked as the perfusion change threshold. 6.The method of claim 5, wherein, Obtaining the real-time initial region based on real-time change values ​​and perfusion change thresholds includes the following sub-steps: Obtain real-time change values ​​that are greater than or equal to the perfusion change threshold and mark them as real-time perfusion change values; Obtain the pixels in the image after real-time perfusion change value and mark them as real-time perfusion pixels; Obtain the independent region composed of real-time infused pixel groups and mark it as the real-time initial region.

7. The method of claim 6, wherein the method further comprises: Obtaining historical qualified values ​​based on historical perfusion change values ​​includes the following sub-steps: Obtain the historical perfusion change values ​​in the image after historical perfusion and mark them as historical perfusion pixels; Obtain the independent region composed of historical infused pixels and mark it as the historical infused region; Obtain the number of pixels within the historical infusion area and mark them as historical qualified values. 8.The method of claim 7, wherein, Obtaining the noise region threshold based on historical acceptable values ​​includes the following sub-steps: Obtain the second number of historical qualified values; Establish a Cartesian coordinate system with historical qualified values ​​as the horizontal axis data and the number of historical qualified values ​​as the vertical axis data, and mark it as the second data coordinate system; Obtain all historical qualified values ​​and their corresponding quantities as coordinate points on the x and y axes, and mark them as the second data coordinate points; Plot all the second data coordinate points in the second data coordinate system. 9.The method of claim 8, wherein, Obtaining the noise region threshold based on historical qualified values ​​also includes the following sub-steps: Get the maximum value of the ordinate among all the second data coordinate points and mark it as the second data height; Set a length and mark it as the second set length; Create a rectangle on the horizontal axis of the second data coordinate point with a height equal to the second data height and a width equal to the second set length, and allow it to move left and right. Mark this rectangle as the second judgment rectangle. The total number of historical qualified values ​​corresponding to the second data coordinate points included within the second judgment rectangle is marked as the second judgment quantity; Mark the range of historical qualified values ​​as the second data range; Assuming that the historical qualified values ​​are evenly distributed within the second data range, obtain the second judgment quantity at this time and mark it as the second average quantity; Set a second percentage value, obtain the product of the second average quantity and the second percentage value, and mark it as the second quantity threshold; In the second data coordinate system, the second judgment rectangle is shifted to the right starting from the leftmost historical qualified value. When the second judgment quantity is greater than or equal to the second quantity threshold, the movement of the second judgment rectangle is stopped. The historical qualified value corresponding to the minimum horizontal coordinate of the second judgment rectangle at this time is obtained and marked as the noise area threshold. 10.The method of claim 9, wherein, Obtaining the contour of the perfusion region based on real-time initial region and noise region thresholds includes the following sub-steps: If the real-time initial quantity is less than the noise region threshold, the real-time initial region is identified as a noise region; if the real-time initial quantity is greater than or equal to the noise region threshold, the outline of the real-time initial region is obtained and regarded as the outline of the infusion region.

Images