An intelligent wound recognition and evaluation method for burn plastic surgery care

Abstract

The present invention relates to the field of patient burn image recognition, and particularly relates to an intelligent wound recognition and evaluation method for burn plastic surgery nursing. The present invention obtains normal color feature data through the color space characteristics of normal skin; combines the difference between each pixel point of the sliding window on the burn area and the pixel points of normal skin to obtain pixel point color parameters; obtains the second-order color difference according to the parameter differences between the pixel points within the sliding window and other pixel points; and further divides the image into different color regions; analyzes the overall region margin of different color regions, and further obtains a burn evaluation coefficient; and intelligently recognizes the wound according to the coefficient. The present invention can adapt to the complexity of burn wounds and timely intelligently recognize and evaluate the burn recovery situation of patients.

Description

Technical Field

[0001] The present invention relates to the field of patient burn image recognition, and particularly to an intelligent recognition and evaluation method for wounds in burn plastic surgery nursing. Background Art

[0002] With the continuous progress of artificial intelligence technology and the increasing demand for accurate and efficient diagnosis in the medical industry, the intelligent recognition market for burn plastic surgery nursing wounds has great development potential. More and more medical institutions begin to recognize the advantages of intelligent recognition technology in improving diagnostic accuracy and treatment efficiency and are willing to invest in relevant equipment and technologies. Although the market prospect is broad, the intelligent recognition industry for burn wounds still faces some challenges. For example, it is relatively difficult to obtain high-quality burn wound image data, and the data collected by different hospitals and different devices are different, which brings challenges to the training and generalization of the model. In addition, more research and practice are needed for the clinical verification and promotion of intelligent recognition technology.

[0003] In reality, the burn wounds of patients will have different color manifestations during the recovery process, and the burn wounds are of a certain complexity. During the period of burn wound recovery, patients usually need the judgment and analysis of relevant personnel to judge the recovery situation of the burn wounds, and it is difficult to achieve intelligent analysis and recognition of the burn wound care of patients. Summary of the Invention

[0004] In order to solve the technical problem that during the recovery process of a patient's burn wound, there will be different color manifestations, and the burn wound has a certain degree of complexity. Usually, relevant personnel need to conduct judgment and analysis during the recovery period of the patient's burn wound to determine the recovery situation of the burn wound, making it difficult to achieve intelligent analysis and recognition of the patient's burn wound care. The purpose of the present invention is to provide a method for intelligent recognition and evaluation of wounds for burn plastic surgery care. The specific technical solution adopted is as follows: A method for intelligent recognition and evaluation of wounds for burn plastic surgery care, the method comprising: collecting a patient's burn image; the patient's burn image being a Lab image; obtaining a normal skin image of the patient; obtaining the normal color feature data of the pixel points in the normal skin image according to the color space features of each pixel point in the normal skin image; presetting a region in the patient's burn image as a preset sliding window, and traversing the preset sliding window in the patient's burn image; obtaining the color parameters of each pixel point in the preset sliding window according to the color feature difference between each pixel point in the preset sliding window and the pixel points of the normal skin image, and the normal color feature data; obtaining the second-order color difference of each pixel point in the preset sliding window according to the color parameter difference between each pixel point in the preset sliding window and all other pixel points; obtaining the overall color difference of the preset sliding window according to the second-order color difference of each pixel point in the preset sliding window; dividing the patient's burn image into different color regions according to the difference between the second-order color difference of each pixel point in the preset sliding window and the overall color difference; obtaining the overall region margin of each color region according to the distance between the edge pixel points and the wound center pixel points of each color region; obtaining a burn judgment coefficient according to the color parameters of the pixel points in all color regions and the overall region margin; and performing intelligent recognition on the wounds for burn plastic surgery care according to the burn judgment coefficient.

[0005] Further, the method for obtaining the normal color feature data includes: obtaining the normal color feature data according to the normal color feature data calculation formula, and the normal color feature data calculation formula is as follows: In the formula, represents the normal color feature data of the pixel points in the patient's burn image; represents the number of pixel points in the patient's burn image; represents the lightness component of the th pixel point in the patient's burn image; represents the green - red axis component of the th pixel point in the patient's burn image; represents the blue - yellow axis component of the th pixel point in the patient's burn image.

[0006] Further, the method for obtaining the color parameters includes: establishing a Cartesian coordinate system in the patient's burn image to obtain the position coordinates of each pixel point in the patient's burn image; obtaining the color parameters according to the color parameter calculation formula, and the color parameter calculation formula is as follows: In the formula, represents the color parameter of the pixel point located at the position coordinate within the preset sliding window; represents the lightness component of the pixel point located at the position coordinate within the preset sliding window; represents the green - red axis component of the pixel point located at the position coordinate within the preset sliding window; represents the blue - yellow axis component of the pixel point located at the position coordinate within the preset sliding window; represents the average value of the lightness components of all pixel points in the normal skin area; represents the average value of the green - red axis components of all pixel points in the normal skin area; represents the blue - yellow axis component of all pixel points in the normal skin area; represents the normal color feature data of the pixel points in the normal skin image; represents the arctangent function.

[0007] Further, the method for obtaining the second - order color difference of each pixel point within the preset sliding window includes: calculating the color parameter difference between each pixel point within the preset sliding window and other pixel points as the first color difference; calculating the average value of the squares of the first color differences between each pixel point within the preset sliding window and all other pixel points to obtain the second - order color difference of each pixel point within the preset sliding window.

[0008] Further, the method for obtaining the overall color difference of the preset sliding window includes: taking the average value of the second - order color differences of each pixel point within the preset sliding window as the overall color difference of the preset sliding window.

[0009] Further, dividing the patient's burn image into different color regions includes: starting from the upper - left corner of the patient's burn image, the preset sliding window traverses in the order from left to right and from top to bottom; during the process of the preset sliding window traversing the patient's burn image, classifying the pixel points with second - order color differences less than the overall color difference within the preset sliding window into the same type of pixel points, classifying the pixel points with second - order color differences greater than the overall color difference within the preset sliding window into another type of pixel points, traversing all pixel points, classifying all pixel points, and taking the region corresponding to each type of pixel points as each color region.

[0010] Further, the method for obtaining the overall region margin includes: calculating the position coordinates of the wound center pixel point in the patient's burn image by using double integral; obtaining the overall region margin according to the overall region margin calculation formula, and the overall region margin calculation formula is as follows: In the formula, represents the overall region margin of the th color region; represents the number of edge pixel points of the th color region; represents the abscissa of the wound center pixel point in the patient's burn image; represents the ordinate of the wound center pixel point in the patient's burn image; represents the abscissa of the th pixel point in the th color region; represents the ordinate of the th pixel point in the th color region.

[0011] Further, the method for obtaining the burn evaluation coefficient includes: obtaining the burn evaluation coefficient according to the burn evaluation coefficient calculation formula, and the burn evaluation coefficient calculation formula is as follows: In the formula, represents the burn evaluation coefficient; represents the number of color regions in the patient's burn image; represents the average color parameter of the pixel points in the th color region; represents the overall region margin of the th color region; represents the number of edge pixel points of the color region that is farthest from the wound center pixel point in the patient's burn image; represents the abscissa of the wound center pixel point in the patient's burn image; represents the ordinate of the wound center pixel point in the patient's burn image; represents the abscissa of the th edge pixel point of the color region that is farthest from the wound center pixel point; represents the ordinate of the th edge pixel point of the color region that is farthest from the wound center pixel point; represents the exponential function with the natural constant as the base.

[0012] Further, the wounds in burn plastic surgery nursing are intelligently identified according to the burn evaluation coefficient, including: recording the burn evaluation coefficient of the patient during each examination. When the ratio between the latest burn evaluation times and the burn evaluation coefficient of the previous examination is less than 1, it is considered that the patient's wound is recovering. When the ratio between the latest burn evaluation times and the burn evaluation coefficient of the previous examination is greater than 1, it is considered that the patient's wound is more serious.

[0013] A wound intelligent identification and evaluation system for burn plastic surgery nursing, the system includes a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the computer program, it implements the steps of the above-mentioned method for intelligent identification and evaluation of wounds in burn plastic surgery nursing.

[0014] The present invention has the following beneficial effects: The present invention collects the Lab image of the patient's burn wound area as the patient's burn image. Because the burn wound has significant color characteristics, and the normal skin image of the patient is obtained and compared with the patient's burn image in terms of color characteristics. Since there is a serious color difference between the patient's burned part and the normal skin, and the color of the wound gradually fades from the wound edge to the wound center during the recovery process, the wound area is divided by the difference between the color areas corresponding to different recovery conditions of the burn and the normal skin area. Therefore, by presetting a sliding window to traverse each pixel point in the patient's burn image, the pixel points are classified by analyzing the color difference between the pixel point and other pixel points in the preset sliding window. First, the color parameters of the pixel points are defined, and then the color deviation between the color parameters of each pixel point and other pixel points in the preset sliding window is calculated through the second-order color difference, and then the patient's burn image is divided into different color areas. Since the characteristics of mild burn wounds are all closed figures, and the skin injury is more serious in the area closer to the center of the burn wound, and the recovery trend is also from the wound edge to the wound center, the un-recovered area can be found by finding the center point of the wound, and the wound recovery situation can be judged by the area margin of different color areas. Therefore, the burn evaluation coefficient is obtained according to the color parameters of the pixel points and the overall area margin. The wounds in burn plastic surgery nursing are intelligently identified according to the burn evaluation coefficient. The present invention can adapt to the complexity of burn wounds and timely intelligently identify and evaluate the burn recovery situation of patients. Description of the Drawings

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 A flowchart of a method for intelligent recognition and evaluation of wounds for burn plastic surgery care provided by an embodiment of the present invention. Detailed implementation manners

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following combines the accompanying drawings and preferred embodiments to detail the specific implementation manners, structures, features and effects of a method for intelligent recognition and evaluation of wounds for burn plastic surgery care proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of a method for intelligent recognition and evaluation of wounds for burn plastic surgery care provided by the present invention with reference to the accompanying drawings.

[0020] Please refer to Figure 1 , which shows a method for intelligent recognition and evaluation of wounds for burn plastic surgery care provided by an embodiment of the present invention. The method includes: Step S1: Collect the burn image of the patient; the burn image of the patient is a Lab image; obtain the normal skin image of the patient.

[0021] The embodiment of the present invention is mainly applied to the scenario of judging the recovery situation of the patient's burn wound. Since the burn wound has significant color characteristics, precise color matching is required as a judgment condition for relevant personnel to judge the wound recovery. The Lab color space is a device-independent color space, and its design goal is to be consistent with human visual perception. It consists of three components: L represents brightness, and a and b represent two axes of color. Among them, the negative value of the a axis represents green, and the positive value represents red; the negative value of the b axis represents blue, and the positive value represents yellow. The color characteristics of the image can be accurately obtained. Therefore, in the embodiment of the present invention, the Lab image of the patient's burn area is obtained as the burn image of the patient required for subsequent operations, and the normal skin area of the patient that is most similar to the color characteristics of the burn area is obtained to form the normal skin image to highlight the color difference of the burn area.

[0022] In one embodiment of the present invention, a high-definition camera is placed directly above the burned area of the patient. The distance between the patient's wound and the camera of the medical device is adjusted until the entire view of the patient's injured wound area and normal skin appear in the camera lens, and then the image is captured. And each time the image is captured later, the same distance as the first capture is maintained to eliminate the judgment error of burn wound plastic surgery care caused by the shooting distance factor. In subsequent evaluations, accurate image color information and wound characteristics are required. Therefore, sufficient light is maintained during shooting. After obtaining the patient's burn wound image, the image segmentation technology is used to segment the patient's body part and the background image in the image, and only the image of the patient's body part is retained. The retained image is denoised and then converted into an RGB image. The RGB image is converted into a Lab image as the patient's burn image required for subsequent operations. It should be noted that the normal skin image is obtained in the same way, and in other embodiments of the present invention, other implementation methods can also be used to obtain the patient's burn image and normal skin image, which are not limited here.

[0023] Step S2: Obtain the normal color feature data of the pixel points in the patient's burn image according to the color space characteristics of each pixel point in the patient's burn image; preset a region in the patient's burn image as a preset sliding window, and traverse the preset sliding window in the patient's burn image; according to the color feature difference between each pixel point in the preset sliding window and the pixel points in the normal skin area, and the normal color feature data, obtain the color parameters of each pixel point in the preset sliding window; according to the color parameter difference between each pixel point in the preset sliding window and all other pixel points, obtain the second-order color difference of each pixel point in the preset sliding window; according to the second-order color difference of each pixel point in the preset sliding window, obtain the overall color difference of the preset sliding window; according to the difference between the second-order color difference of each pixel point in the preset sliding window and the overall color difference, divide the patient's burn image into different color regions; according to the distance between the edge pixel points and the wound center pixel points of each color region, obtain the overall region margin of each color region; according to the color parameters of the pixel points in all color regions and the overall region margin, obtain the burn evaluation coefficient.

[0024] Since the wound of the patient's burned area will turn red after mild burns, there is a serious color difference from the normal skin, and during the recovery process, the color of the wound will gradually fade from the wound edge to the wound center and get closer and closer to the normal skin color, that is, it slowly changes from the non-recovered area to the well-recovered area, and then the skin color of the well-recovered area will gradually approach the normal skin color. Therefore, the wound area is divided according to the color characteristics in the patient's burn image. Here, the color characteristics of the normal skin are first analyzed, and the wound area is divided according to the difference between the color regions corresponding to different recovery conditions of the burn and the normal skin area.

[0025] Preferably, in one embodiment of the present invention, the method for obtaining normal color feature data includes: obtaining normal color feature data according to the normal color feature data calculation formula, and the normal color feature data calculation formula is as follows: In the formula, represents the normal color feature data of the pixel points in the patient's burn image; represents the number of pixel points in the patient's burn image; represents the th pixel point's lightness component in the patient's burn image; represents the th pixel point's green - red axis component in the patient's burn image; represents the th pixel point's blue - yellow axis component in the patient's burn image.

[0026] Traverse each pixel point in the patient's burn image through a preset sliding window, analyze the color differences between the pixel point and other pixel points within the preset sliding window to classify the pixel points, and first analyze the color parameters of each pixel point within the preset sliding window for subsequent comparison processing.

[0027] Preferably, in one embodiment of the present invention, the method for obtaining color parameters includes: establishing a Cartesian coordinate system in the patient's burn image to obtain the position coordinates of each pixel point in the patient's burn image; it should be noted that in the embodiments of the present invention, subsequent operations are carried out in this Cartesian coordinate system.

[0028] Obtain color parameters according to the color parameter calculation formula, and the color parameter calculation formula is as follows: In the formula, represents the color parameter of the pixel point located at the position coordinate within the preset sliding window; represents the lightness component of the pixel point located at the position coordinate within the preset sliding window; represents the green - red axis component of the pixel point located at the position coordinate within the preset sliding window; represents the blue - yellow axis component of the pixel point located at the position coordinate within the preset sliding window; represents the average value of the lightness components of all pixel points in the normal skin area; represents the average value of the green - red axis components of all pixel points in the normal skin area; represents the blue - yellow axis component of all pixel points in the normal skin area; represents the normal color feature data of the pixel points in the normal skin image; represents the arctangent function.

[0029] In the color parameter calculation formula, during the recovery of burn wounds, the difference between the components of all color channels of a pixel point and the color channel components of the pixel point in the normal skin image gradually decreases. At this time , , will all become smaller, and the smaller the difference from the normal color feature data of the pixel points in the normal skin image, so the color parameter of the pixel point located at the position coordinate in the preset sliding window is smaller. Since the arctangent function is a positive correlation function and changes significantly near the coordinate origin, the embodiment of the present invention uses the arctangent function to reflect the change of the color parameter of the pixel points in the preset sliding window.

[0030] In an embodiment of the present invention, the preset sliding window is set to a rectangular area of , and the preset sliding window can be set by itself, which is not limited here.

[0031] Calculate the color deviation between the color parameters of each pixel point in the preset sliding window and other pixel points through second-order color difference. Preferably, in an embodiment of the present invention, the method for obtaining the second-order color difference of each pixel point in the preset sliding window includes: calculating the color parameter difference between each pixel point in the preset sliding window and other pixel points as the first color difference; calculating the mean value of the squares of the first color differences between each pixel point in the preset sliding window and all other pixel points to obtain the second-order color difference of each pixel point in the preset sliding window. In an embodiment of the present invention, the second-order color difference calculation formula is as follows: In the formula,[[]] represents the second-order color difference of the pixel point located at the position coordinate in the preset sliding window; represents the number of pixel points other than the pixel point located at the position coordinate in the preset sliding window; represents the color parameter of the th other pixel point in the preset sliding window; represents the color parameter of the pixel point located at the position coordinate in the preset sliding window.

[0032] The second-order color difference reflects the color deviation between each pixel in the preset sliding window and other pixels. Since in the initial stage of a burn wound, inflammatory reactions such as redness, swelling, heat, and pain occur in the wound surface and the surrounding tissues, the local skin color changes, the skin of superficial burns turns red, and there is a huge color gap with the normal skin. Therefore, the second-order color difference of the obtained pixels will be very large. After the skin has recovered for a period of time, granulation tissue hyperplasia and epithelial cell regeneration begin to appear in the wound, the wound surface gradually shrinks, and the wound color will gradually approach the normal skin color. At this time, the obtained second-order color difference will be smaller, while the second-order color difference of the pixels in the normal skin image will be close to zero. Therefore, if there are skin pixels in different regions of the wound in the preset sliding window, there will be a huge fluctuation in the second-order color difference of the pixels. Therefore, in the embodiments of the present invention, the overall color difference of the pixels in the preset sliding window is calculated, and different color regions are classified based on the difference between the second-order color difference of each pixel in the preset sliding window and the overall color difference.

[0033] Preferably, in one embodiment of the present invention, the mean value of the second-order color differences of each pixel in the preset sliding window is used as the overall color difference of the preset sliding window.

[0034] Preferably, in one embodiment of the present invention, the patient's burn image is divided into different color regions, including: the preset sliding window starts from the upper left corner of the patient's burn image and traverses in the order from left to right and from top to bottom.

[0035] In reality, the color of the burn wound fades gradually from the wound edge to the wound center during the recovery process, and there may be completely recovered regions, regions under recovery, and unrecovered regions. Therefore, during the process of the preset sliding window traversing the pixels, there may be 2 to 3 color regions. During the process of the preset sliding window traversing the patient's burn image, the pixels with a second-order color difference less than the overall color difference in the preset sliding window are grouped into the same class of pixels, and the pixels with a second-order color difference greater than the overall color difference in the preset sliding window are grouped into another class of pixels. After traversing all the pixels, all the pixels are classified, and the region corresponding to each class of pixels is used as each color region. Among them, because the skin regions of the burn wound are continuous, that is, each skin region is complete and does not mix with small regions. Therefore, if the classification results of some pixels are different, then taking such pixels as the central pixels, in the preset neighborhood, the above operations are performed to finally classify the pixels with different classification results, and finally all color regions are obtained.

[0036] Based on the particularity of the shape characteristics of burn wounds, it can be known that the characteristics of light burn wounds are all closed figures, and the closer to the center of the burn wound, the more serious the skin injury, and the recovery trend is also from the edge of the wound to the center of the wound. Therefore, the unrecovered area can be found by finding the center point of the wound, and the wound recovery status can be judged by the regional margins of different color areas.

[0037] Preferably, in one embodiment of the present invention, the method for obtaining the overall area margin of each color area includes: if the method of determining the center pixel of the wound is only based on the number of pixels in the horizontal and vertical directions, a large error will occur when facing the complexity and diversity of burn wounds, and the correct center pixel of the wound cannot be found. By fitting the edge pixel position information of the most central color area into a curve, and then finding the center point of the most central area through the curve relationship, it will be more accurate and have better anti-interference when facing the diversity of wound shapes.

[0038] Therefore, the double integral is used to calculate the position coordinates of the central pixel point of the wound in the patient's burn image. The embodiment of the present invention provides a method for calculating the position coordinates of the central pixel point of the wound, including: obtaining all edge pixel points of the most central color area in the patient's burn image, evenly dividing the edge pixel points into two parts, performing nonlinear fitting on the edge pixel points of each part, and obtaining the curve function corresponding to the upper and lower parts. , .

[0039] right , To perform double integral, the specific steps are as follows: ; ; In the above double integral formula, Indicates the horizontal coordinate of the pixel at the center of the wound; Indicates the vertical coordinate of the pixel at the center of the wound; Represents the extent of the most central color region in the patient's burn image; Represents the horizontal coordinate of the leftmost pixel in the most central color area of the patient's burn image; Represents the abscissa of the rightmost pixel in the most central color area of the patient's burn image; A curve function representing the lower half of the centermost color region in the patient's burn image; A curve function representing the upper half of the centermost color region in a patient burn image.

[0040] In this way, the central pixel of the wound in the patient's burn image is obtained .

[0041] Obtain the overall region margin according to the overall region margin calculation formula, and the overall region margin calculation formula is as follows: In the formula, represents the overall region margin of the th color region; represents the number of edge pixel points of the th color region; represents the abscissa of the wound center pixel point in the patient's burn image; represents the ordinate of the wound center pixel point in the patient's burn image; represents the abscissa of the th pixel point in the th color region; represents the ordinate of the th pixel point in the th color region.

[0042] In the overall region margin calculation formula, calculate the average distance between the edge pixel points of each color region and the wound center pixel point , and use it as the overall region margin of each color region.

[0043] Use the color parameters of each pixel point in the patient's burn image and the overall region margin to judge the burn recovery situation.

[0044] Preferably, in an embodiment of the present invention, the method for obtaining the burn evaluation coefficient includes: obtaining the burn evaluation coefficient according to the burn evaluation coefficient calculation formula, and the burn evaluation coefficient calculation formula is as follows: In the formula, represents the burn evaluation coefficient; represents the number of color regions in the patient's burn image; represents the average color parameter of the pixel points in the th color region; represents the overall region margin of the th color region; represents the number of edge pixel points of the color region farthest from the wound center pixel point in the patient's burn image; represents the abscissa of the wound center pixel point in the patient's burn image; represents the ordinate of the wound center pixel point in the patient's burn image; represents the abscissa of the th edge pixel point of the color region farthest from the wound center pixel point; represents the ordinate of the th edge pixel point of the color region farthest from the wound center pixel point; represents the exponential function with the natural constant as the base.

[0045] In the burn evaluation coefficient calculation formula, the overall area margin of each color region is larger, and the average distance of the outermost color region from the pixel point at the center of the wound is smaller, indicating that the proportion of the unhealed area of the wound in the patient's burn image is larger, the area of the wound that has not fully healed is larger, and at this time the burn evaluation coefficient is larger; and the larger the average color parameter of the pixel points in each color region, the more severe the burn condition, and at this time the burn evaluation coefficient is larger; among them, 3 is the adjustment coefficient of the function. According to its function image, the value range of is limited between 0.05 and 1. It should be noted that the adjustment coefficient can be set by oneself and is not limited here.

[0046] Step S3: Intelligently identify the wound of burn plastic surgery care according to the burn evaluation coefficient.

[0047] Preferably, in an embodiment of the present invention, intelligently identifying the wound of burn plastic surgery care according to the burn evaluation coefficient includes: recording the burn evaluation coefficient of the patient each time during the examination. When the ratio between the latest burn evaluation number and the burn evaluation coefficient of the previous examination is less than 1, it is considered that the patient's wound is recovering. When the ratio between the latest burn evaluation number and the burn evaluation coefficient of the previous examination is greater than 1, it is considered that the patient's wound is more severe.

[0048] So far, the intelligent identification and evaluation of the burn wound are completed.

[0049] In summary, collect the patient's burn image; the patient's burn image is a Lab image; obtain the patient's normal skin image; obtain the normal color feature data of the pixel points in the normal skin image according to the color space characteristics of each pixel point in the normal skin image; preset a region in the patient's burn image as a preset sliding window, and traverse the preset sliding window in the patient's burn image; obtain the color parameter of each pixel point in the preset sliding window according to the color feature difference between each pixel point in the preset sliding window and the pixel points in the normal skin image, and the normal color feature data; obtain the second-order color difference of each pixel point in the preset sliding window according to the color parameter difference between each pixel point in the preset sliding window and all other pixel points; obtain the overall color difference of the preset sliding window according to the second-order color difference of each pixel point in the preset sliding window; divide the patient's burn image into different color regions according to the difference between the second-order color difference of each pixel point in the preset sliding window and the overall color difference; obtain the overall area margin of each color region according to the distance between the edge pixel points of each color region and the pixel point at the center of the wound; obtain the burn evaluation coefficient according to the color parameter and the overall area margin of the pixel points in all color regions; intelligently identify the wound of burn plastic surgery care according to the burn evaluation coefficient.

[0050] An embodiment of the present invention provides a wound intelligent recognition and evaluation system for burn plastic surgery care. The system includes a memory, a processor, and a computer program, where the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, it can implement the method described in steps S1 - S3.

[0051] It should be noted that: the above - mentioned sequence of embodiments of the present invention is only for description and does not represent the advantages or disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0052] Each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An intelligent wound recognition and evaluation method for burn plastic surgery care, characterized in that, The method includes: collecting a burn image of a patient; the burn image of the patient being a Lab image; obtaining a normal skin image of the patient; obtaining normal color feature data of pixel points in the normal skin image according to the color space features of each pixel point in the normal skin image; presetting an area in the burn image of the patient as a preset sliding window, and traversing the preset sliding window in the burn image of the patient; obtaining color parameters of each pixel point in the preset sliding window according to the color feature difference between each pixel point in the preset sliding window and the pixel points in the normal skin image, and the normal color feature data; obtaining the second-order color difference of each pixel point in the preset sliding window according to the color parameter difference between each pixel point in the preset sliding window and all other pixel points; obtaining the overall color difference of the preset sliding window according to the second-order color difference of each pixel point in the preset sliding window; dividing the burn image of the patient into different color regions according to the difference between the second-order color difference of each pixel point in the preset sliding window and the overall color difference; obtaining the overall regional margin of each color region according to the distance between the edge pixel points and the wound center pixel points of each color region; obtaining a burn evaluation coefficient according to the color parameters and the overall regional margin of pixel points in all color regions; intelligently identifying the wound for burn plastic surgery nursing according to the burn evaluation coefficient; the method for obtaining the color parameters includes: establishing a Cartesian coordinate system in the burn image of the patient, and obtaining the position coordinates of each pixel point in the burn image of the patient; obtaining the color parameters according to the color parameter calculation formula, and the color parameter calculation formula is as follows: In the formula, represents the color parameter of the pixel point located at the position coordinate in the preset sliding window; represents the lightness component of the pixel point located at the position coordinate in the preset sliding window; represents the green - red axis component of the pixel point located at the position coordinate in the preset sliding window; represents the blue - yellow axis component of the pixel point located at the position coordinate in the preset sliding window; represents the average value of the lightness components of all pixel points in the normal skin area; represents the average value of the green - red axis components of all pixel points in the normal skin area; represents the blue - yellow axis component of all pixel points in the normal skin area; represents the normal color feature data of the pixel points in the normal skin image; represents the arctangent function; the method for obtaining the burn evaluation coefficient includes: obtaining the burn evaluation coefficient according to the burn evaluation coefficient calculation formula, and the burn evaluation coefficient calculation formula is as follows: In the formula, represents the burn evaluation coefficient; represents the number of color regions in the patient's burn image; Indicates The mean value of the color parameters of the pixels in the color area; Indicates The overall area margin of the color area; Indicates the number of edge pixels in the color area farthest from the wound center pixel in the patient's burn image; Represents the horizontal coordinate of the wound center pixel in the patient's burn image; Represents the vertical coordinate of the central pixel of the wound in the patient's burn image; The color area farthest from the wound center pixel The horizontal coordinate of the edge pixel; The color area farthest from the wound center pixel The vertical coordinate of the edge pixel; Represents an exponential function with a natural constant as base.

2. The intelligent wound recognition and evaluation method for burn plastic surgery nursing according to claim 1, wherein The method for obtaining the normal color feature data includes: obtaining the normal color feature data according to the normal color feature data calculation formula, and the normal color feature data calculation formula is as follows: In the formula, represents the normal color feature data of the pixel points in the normal skin image; represents the number of pixel points in the normal skin image; represents the lightness component of the th pixel point in the normal skin image; represents the green - red axis component of the th pixel point in the normal skin image; represents the blue - yellow axis component of the th pixel point in the normal skin image.

3. The intelligent wound recognition and evaluation method for burn plastic surgery nursing according to claim 1, characterized in that, The method for obtaining the second-order color difference of each pixel point within the preset sliding window includes: calculating the color parameter difference between each pixel point within the preset sliding window and other pixel points as the first color difference; calculating the mean value of the squares of the first color differences between each pixel point within the preset sliding window and all other pixel points to obtain the second-order color difference of each pixel point within the preset sliding window.

4. A method for intelligent recognition and evaluation of wounds for burn plastic surgery care according to claim 1, characterized in that, The method for obtaining the overall color difference of the preset sliding window includes: taking the mean value of the second-order color differences of each pixel point within the preset sliding window as the overall color difference of the preset sliding window.

5. A method for intelligent recognition and evaluation of wounds for burn plastic surgery care according to claim 1, characterized in that, Dividing the patient's burn image into different color regions includes: starting from the upper left corner of the patient's burn image, the preset sliding window traverses in the order from left to right and from top to bottom; during the process of the preset sliding window traversing the patient's burn image, the pixel points with a second-order color difference less than the overall color difference within the preset sliding window are classified into the same type of pixel points, and the pixel points with a second-order color difference greater than the overall color difference within the preset sliding window are classified into another type of pixel points. After traversing all pixel points and classifying all pixel points, the region corresponding to the pixel points of each category is used as each color region.

6. The method for intelligent recognition and evaluation of wounds for burn plastic surgery nursing according to claim 1, wherein, The method for obtaining the overall region margin includes: calculating the position coordinates of the wound center pixel in the patient's burn image by using double integral; obtaining the overall region margin according to the overall region margin calculation formula, and the overall region margin calculation formula is as follows: In the formula, represents the overall region margin of the th color region; represents the number of edge pixels of the th color region; represents the abscissa of the wound center pixel in the patient's burn image; represents the ordinate of the wound center pixel in the patient's burn image; represents the abscissa of the th pixel in the th color region; represents the ordinate of the th pixel in the th color region.

7. A method for intelligent recognition and evaluation of wounds for burn plastic surgery care according to claim 1, characterized in that, Intelligently identifying the wound of burn plastic surgery nursing according to the burn evaluation coefficient includes: recording the burn evaluation coefficient during each examination of the patient. When the ratio between the latest burn evaluation times and the burn evaluation coefficient of the previous examination is less than 1, it is considered that the patient's wound is recovering. When the ratio between the latest burn evaluation times and the burn evaluation coefficient of the previous examination is greater than 1, it is considered that the patient's wound is more serious.

8. An intelligent wound recognition and evaluation system for burn plastic surgery care, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for intelligent identification and evaluation of the wound for burn plastic surgery nursing according to any one of claims 1 to 7.

Images