A patient wound identification system for surgical care

Abstract

The present invention discloses a patient wound recognition system for surgical care, the system comprising: an analysis module, configured to obtain a real-time wound image, and perform similarity analysis on the wound image and each historical wound image in a first wound image sequence according to a preset image recognition model; a cropping module, configured to obtain a feature image corresponding to a historical wound image, and crop the wound area of ​​a wound image according to the ratio of the feature image to the historical wound image to obtain a target wound image containing a wound area to be recognized; a calculation module, configured to slide on the edge of the wound area to be recognized in a target wound image based on a preset dynamic sliding window, and calculate the area of ​​the real wound area in the wound area to be recognized according to the recognition result. The amount of pixel point data required for subsequent recognition is reduced as much as possible, thereby improving the recognition efficiency and accuracy of the patient's wound.

Description

Technical Field

[0001] The invention belongs to the technical field of wound identification, and in particular relates to a patient wound identification system for surgical care. Background Art

[0002] Clinically, the care and assessment of chronic wounds are based on the length, width, depth, area, angle and length of the wound / sinus tract / fistula, combined with the wound tissue composition, exudate properties, color and amount of exudate, odor, degree and frequency of wound pain, and surrounding skin condition, as reference indicators for nursing and treatment, in order to conduct wound assessment, treatment improvement, and recovery management and monitoring.

[0003] However, when photographing the wound surface, both the wound surface and the non-wound surface around the wound surface are coated with drugs. At this time, during the process of uploading and identifying the photographed image, the color of the drug or other factors may affect the area around the wound surface, making it difficult for the image recognition model to accurately obtain the true range of the wound surface, further increasing the recognition amount of the image recognition model, and thus affecting the efficiency and accuracy of wound surface recognition. Summary of the invention

[0004] The present invention provides a patient wound recognition system for surgical care, which is used to solve the technical problem that an interference area is generated around a wound surface, making it difficult for an image recognition model to accurately obtain the true wound surface range.

[0005] The present invention provides a patient wound identification system for surgical care, comprising a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the following steps:

[0006] Acquire an original wound image and at least one historical wound image within a preset historical time period, wherein the original wound image is a wound image that only contains an uncoated wound area before the preset historical time period, and any historical wound image contains at least one wound area;

[0007] Sorting the at least one historical wound image based on a time sequence to obtain a first wound image sequence corresponding to the preset historical time period, and comparing each historical wound image in the first wound image sequence with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each historical wound image;

[0008] Acquire a real-time wound image, perform similarity analysis on the wound image and each historical wound image in the first wound image sequence according to a preset image recognition model, and select a historical wound image associated with the wound image according to the analysis result, wherein the historical wound image is the historical wound image in the first wound image sequence that has the highest similarity with the wound image;

[0009] Acquire a characteristic image corresponding to the historical wound image, and cut the wound area of ​​the wound image according to the ratio of the characteristic image to the historical wound image to obtain a target wound image including the wound area to be identified;

[0010] Based on a preset dynamic sliding window, a slide is performed on the edge of the wound area to be identified in the target wound image, and the area of ​​the real wound area in the wound area to be identified is calculated according to the recognition result, wherein the dynamic sliding window dynamically adjusts the size of the dynamic sliding window according to the current proportion of target pixel points, and the target pixel point is a pixel point in the real wound area.

[0011] Furthermore, the comparing each historical wound image in the first wound image sequence with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each historical wound image includes:

[0012] aligning a first historical wound image in the first wound image sequence with the original wound image, and determining whether a wound area in the first historical wound image completely covers a wound area in the original wound image;

[0013] If it is completely covered, the wound area included in the original wound image is removed from the first historical wound image to obtain a first feature image corresponding to the first historical wound image.

[0014] Furthermore, after determining whether the wound area in the first historical wound image completely covers the wound area in the original wound image, the system further performs the following steps:

[0015] If it is not completely covered, the image formed by all the features in the first historical wound image is directly used as the first feature image corresponding to the first historical wound image.

[0016] Furthermore, the cropping of the wound area of ​​the wound image according to the ratio of the feature image to the historical wound image to obtain a target wound image including the wound area to be identified includes:

[0017] Determining the number of predicted pixels in a predicted feature image corresponding to the wound image according to a ratio of the number of pixels in the feature image to the number of pixels in the historical wound image;

[0018] According to the predicted number of pixels, a preset cropping rule is used to perform pixel cropping in the wound area of ​​the wound image to obtain a target wound image containing the wound area to be identified, wherein the cropping rule is to remove pixels in a circumferential direction from the periphery to the inside of the wound area of ​​the wound image.

[0019] Furthermore, the sliding of the preset dynamic sliding window on the edge of the wound area to be identified in the target wound image includes:

[0020] Obtaining a first number of target pixel points in the dynamic sliding window at a current moment, and adjusting a size of the dynamic sliding window according to a ratio of the first number to the number of all pixel points in the dynamic sliding window, wherein a negatively correlated corresponding relationship exists between the size, the first number, and the ratio of the number of all pixel points in the dynamic sliding window;

[0021] The resized dynamic sliding window continues to slide on the edge of the wound area to be identified, and determines that a second number of target pixel points in the dynamic sliding window at a next moment is greater than a preset number threshold;

[0022] If it is not greater than the preset number threshold, the second number of target pixel points in the dynamic sliding window at the next moment is obtained, and the size of the dynamic sliding window is adjusted again according to the ratio of the second number to the number of all pixel points in the dynamic sliding window.

[0023] Furthermore, after determining that the second number of target pixel points in the dynamic sliding window at the next moment is greater than the preset number threshold, the system further performs the following steps:

[0024] If the number is greater than a preset threshold, the size of the dynamic sliding window at the next moment is not adjusted.

[0025] Further, calculating the area of ​​the real wound region in the to-be-identified wound region according to the identification result comprises:

[0026] Acquire a first target number of all target pixels in the sliding area covered by the dynamic sliding window when sliding, and acquire a second target number of all target pixels in other areas of the wound area to be identified, wherein the other areas are areas of the wound area to be identified excluding the sliding area;

[0027] The first target quantity and the second target quantity are superimposed to calculate the area of ​​the real wound region in the wound region to be identified.

[0028] The present application discloses a patient wound recognition system for surgical care, which obtains a real-time wound image, performs similarity analysis on the wound image and each historical wound image in the first wound image sequence according to a preset image recognition model, and selects a historical wound image associated with the wound image according to the analysis result. In this way, since the similarity is the largest, it means that although the time interval between the wound images is long, the degree of change in the wound area is not large. The reason for this phenomenon may be the existence of a large interference area, that is, a non-real wound area. Therefore, a feature image corresponding to the historical wound image is obtained, and the wound area of ​​the wound image is cropped according to the ratio of the feature image to the historical wound image to obtain a target wound image containing the wound area to be identified. The target wound image has more non-real wound areas removed than the feature image, thereby reducing the amount of pixel data that needs to be identified later as much as possible, and solving the problem that interference areas are generated around the wound surface, making it difficult for the image recognition model to accurately obtain the real wound surface range. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0030] Figure 1 A flowchart of the steps performed in a patient wound identification system for surgical care provided by one embodiment of the present invention;

[0031] Figure 2 A structural block diagram of a patient wound identification system for surgical care provided by an embodiment of the present invention;

[0032] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] See also Figure 1 , which shows a flowchart of the steps executed in a patient wound identification system for surgical care of the present application.

[0035] like Figure 1 As shown, step S101, obtains the original wound image and at least one historical wound image within a preset historical time period, wherein the original wound image is a wound image that only contains uncoated wound areas before the preset historical time period, and any historical wound image contains at least one wound area.

[0036] For example, before the wound of a patient needs to be medicated, the wound is photographed to obtain a wound image containing only the uncoated wound area. Furthermore, multiple historical wound images are obtained within a continuous period of time after the medicated wound is applied, which facilitates subsequent analysis.

[0037] Step S102, sorting the at least one historical wound image based on time order to obtain a first wound image sequence corresponding to the preset historical time period, and comparing each historical wound image in the first wound image sequence with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each of the historical wound images.

[0038] In this step, each historical wound image is sorted based on the time sequence to obtain a first wound image sequence corresponding to a preset historical time period. Afterwards, the first historical wound image in the first wound image sequence is aligned with the original wound image, and it is determined whether the wound area in the first historical wound image completely covers the wound area in the original wound image; if it is completely covered, it means that the acquisition time of the first historical wound image and the acquisition time of the original wound image are not far apart, and the first situation in which the wound area in the first historical wound image completely covers the wound area in the original wound image may be caused by the presence of an interference area containing drugs. Therefore, the wound area in the original wound image is removed from the first historical wound image to obtain a first feature image corresponding to the first historical wound image, and the first feature image may be an interference area;

[0039] Furthermore, a second situation occurs in which the wound area in the first historical wound image completely covers the wound area in the original wound image. The acquisition time of the first historical wound image is far different from the acquisition time of the original wound image. However, the interference area containing the drug is particularly large, resulting in the above-mentioned second situation. Therefore, the wound area contained in the original wound image is removed from the first historical wound image to obtain a first feature image corresponding to the first historical wound image, and the first feature image is the partial interference area.

[0040] It should be noted that the method for aligning the first historical wound image in the first wound image sequence with the original wound image can be: obtaining multiple identical pixel points in the first historical wound image and the original wound image, and aligning the multiple identical pixel points, that is, aligning the first historical wound image in the first wound image sequence with the original wound image.

[0041] In a specific embodiment, after determining whether the wound area in the first historical wound image completely covers the wound area in the original wound image, if it does not completely cover, the image formed by all the features in the first historical wound image is directly used as the first feature image corresponding to the first historical wound image. In this way, the risk of removing the real wound area can be reduced, and it is not easy to affect the subsequent area calculation process of the real wound area.

[0042] Step S103, obtaining a real-time wound image, performing similarity analysis on the wound image and each historical wound image in the first wound image sequence according to a preset image recognition model, and screening out a historical wound image associated with the wound image according to the analysis result, wherein the historical wound image is the historical wound image in the first wound image sequence that has the highest similarity with the wound image.

[0043] In this step, the image recognition model is a VGGNet network model, and the VGGNet network model is used to calculate the similarity between a certain wound image and each historical wound image in the first wound image sequence. The VGGNet network model is obtained by machine learning using multiple sets of training images and test images.

[0044] Step S104, obtaining a certain characteristic image corresponding to the certain historical wound image, and cropping the wound area of ​​the certain wound image according to the ratio of the certain characteristic image to the certain historical wound image to obtain a certain target wound image containing the wound area to be identified.

[0045] In this step, the number of predicted pixels in a predicted feature image corresponding to a certain wound image is determined according to the ratio of the number of pixels in a certain feature image to the number of pixels in a certain historical wound image; according to the predicted number of pixels, a preset cropping rule is used to crop pixels in the wound area of ​​a certain wound image to obtain a certain target wound image containing the wound area to be identified, wherein the cropping rule is to remove pixels from the wound area of ​​a certain wound image in a direction from the periphery to the inside. By cropping, the interference area in a certain target wound image, such as caused by the color of the drug or other factors, can be reduced as much as possible, thereby reducing the amount of data required for subsequent identification.

[0046] Step S105, based on a preset dynamic sliding window, slide on the edge of the wound area to be identified in the target wound image, and calculate the area of ​​the real wound area in the wound area to be identified according to the recognition result, wherein the dynamic sliding window dynamically adjusts the size of the dynamic sliding window according to the current proportion of target pixel points, and the target pixel point is a pixel point in the real wound area.

[0047] In this step, a first number of target pixel points in the dynamic sliding window at the current moment is obtained, and the size of the dynamic sliding window is adjusted according to the ratio of the first number to the number of all pixel points in the dynamic sliding window, wherein there is a negatively correlated correspondence between the size, the first number and the ratio of the number of all pixel points in the dynamic sliding window; the dynamic sliding window after the size adjustment continues to slide on the edge of the wound area to be identified, and it is determined whether the second number of target pixel points in the dynamic sliding window at the next moment is greater than a preset number threshold; if it is not greater than the preset number threshold, the second number of target pixel points in the dynamic sliding window at the next moment is obtained, and the size of the dynamic sliding window is adjusted again according to the ratio of the second number to the number of all pixel points in the dynamic sliding window; if it is greater than the preset number threshold, the size of the dynamic sliding window at the next moment is not adjusted.

[0048] It should be noted that calculating the area of ​​the real wound area in the wound area to be identified based on the recognition result includes: obtaining a first target number of all target pixel points in the sliding area covered by the dynamic sliding window when sliding, and obtaining a second target number of all target pixel points in other areas of the wound area to be identified, wherein the other areas are areas of the wound area to be identified excluding the sliding area; superimposing the first target number and the second target number to calculate the area of ​​the real wound area in the wound area to be identified. The above process of obtaining the first target number and the second target number can also be implemented by a trained existing neural network.

[0049] In summary, the method of the present application obtains a real-time wound image, performs similarity analysis on the wound image and each historical wound image in the first wound image sequence according to a preset image recognition model, and selects a historical wound image associated with the wound image according to the analysis result. In this way, since the similarity is the largest, it means that although the time interval between a certain wound image is long, the degree of change in the wound area is not large. The reason for this phenomenon may be the existence of a large interference area, that is, a non-real wound area. Therefore, a feature image corresponding to the historical wound image is obtained, and the wound area of ​​the wound image is cropped according to the ratio of the feature image to the historical wound image to obtain a target wound image containing the wound area to be identified. The target wound image removes more non-real wound areas than the feature image, thereby reducing the amount of pixel data that needs to be identified later as much as possible, and solves the problem that interference areas are generated around the wound surface, making it difficult for the image recognition model to accurately obtain the real wound surface range.

[0050] See also Figure 2 , which shows a structural block diagram of a patient wound identification system for surgical care of the present application.

[0051] like Figure 2 As shown, the patient wound identification system 200 includes an acquisition module 210 , a comparison module 220 , an analysis module 230 , a cutting module 240 and a calculation module 250 .

[0052] Among them, the acquisition module 210 is configured to acquire an original wound image and at least one historical wound image within a preset historical time period, wherein the original wound image is a wound image that only contains an uncoated wound area before the preset historical time period, and any historical wound image contains at least one wound area; the comparison module 220 is configured to sort the at least one historical wound image based on a chronological order to obtain a first wound image sequence corresponding to the preset historical time period, and compare each historical wound image in the first wound image sequence with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each historical wound image; the analysis module 230 is configured to acquire a real-time wound image, perform similarity analysis on the wound image and each historical wound image in the first wound image sequence according to a preset image recognition model, and compare the wound image with each historical wound image in the first wound image sequence according to the classification rule. The analysis result screens out a historical wound image associated with the certain wound image, wherein the certain historical wound image is the historical wound image with the highest similarity to the certain wound image in the first wound image sequence; a cropping module 240 is configured to obtain a certain feature image corresponding to the certain historical wound image, and crop the wound area of ​​the certain wound image according to the ratio of the certain feature image to the certain historical wound image to obtain a certain target wound image containing the wound area to be identified; a calculation module 250 is configured to slide on the edge of the wound area to be identified in the certain target wound image based on a preset dynamic sliding window, and calculate the area of ​​the real wound area in the wound area to be identified according to the recognition result, wherein the dynamic sliding window dynamically adjusts the size of the dynamic sliding window according to the current proportion of target pixels, and the target pixels are pixels in the real wound area.

[0053] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0054] In other embodiments, embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the patient wound identification method for surgical care in any of the above method embodiments;

[0055] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0056] Acquire an original wound image and at least one historical wound image within a preset historical time period, wherein the original wound image is a wound image that only contains an uncoated wound area before the preset historical time period, and any historical wound image contains at least one wound area;

[0057] Sorting the at least one historical wound image based on a time sequence to obtain a first wound image sequence corresponding to the preset historical time period, and comparing each historical wound image in the first wound image sequence with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each historical wound image;

[0058] Acquire a real-time wound image, perform similarity analysis on the wound image and each historical wound image in the first wound image sequence according to a preset image recognition model, and select a historical wound image associated with the wound image according to the analysis result, wherein the historical wound image is the historical wound image in the first wound image sequence that has the highest similarity with the wound image;

[0059] Acquire a characteristic image corresponding to the historical wound image, and cut the wound area of ​​the wound image according to the ratio of the characteristic image to the historical wound image to obtain a target wound image including the wound area to be identified;

[0060] Based on a preset dynamic sliding window, a slide is performed on the edge of the wound area to be identified in the target wound image, and the area of ​​the real wound area in the wound area to be identified is calculated according to the recognition result, wherein the dynamic sliding window dynamically adjusts the size of the dynamic sliding window according to the current proportion of target pixel points, and the target pixel point is a pixel point in the real wound area.

[0061] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the patient wound identification system for surgical care, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the patient wound identification system for surgical care via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0062] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example, the connection via bus is taken as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the patient wound identification method for surgical care of the above-mentioned method embodiment is implemented. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the patient wound identification system for surgical care. The output device 340 may include a display device such as a display screen.

[0063] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0064] As an embodiment, the electronic device is applied to a patient wound identification system for surgical care, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:

[0065] Acquire an original wound image and at least one historical wound image within a preset historical time period, wherein the original wound image is a wound image that only contains an uncoated wound area before the preset historical time period, and any historical wound image contains at least one wound area;

[0066] Sorting the at least one historical wound image based on a time sequence to obtain a first wound image sequence corresponding to the preset historical time period, and comparing each historical wound image in the first wound image sequence with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each historical wound image;

[0067] Obtain a real-time image of a certain wound, perform similarity analysis on the certain wound image with each historical wound image in the first wound image sequence according to a preset image recognition model, and screen out a certain historical wound image associated with the certain wound image according to the analysis result, where the certain historical wound image is the historical wound image with the highest similarity to the certain wound image in the first wound image sequence;

[0068] Obtain a certain feature image corresponding to the certain historical wound image, and crop the wound area of the certain wound image according to the ratio of the certain feature image to the certain historical wound image to obtain a certain target wound image containing the wound area to be recognized;

[0069] Based on a preset dynamic sliding window, slide on the edge of the wound area to be recognized in the certain target wound image, and calculate the area of the real wound area in the wound area to be recognized according to the recognition result, where the dynamic sliding window dynamically adjusts the size of the dynamic sliding window according to the proportion of the current target pixel point, and the target pixel point is the pixel point in the real wound area.

[0070] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution, in essence, or the part that contributes to the prior art can be embodied in the form of a software product, and this computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiments.

[0071] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present invention.

Claims

1. A patient wound identification system for surgical care, characterized in that: The invention comprises a memory and a processor, wherein the processor executes a computer program stored in the memory to implement the following steps: Acquire an original wound image and at least one historical wound image within a preset historical time period, wherein the original wound image is a wound image that only contains an uncoated wound area before the preset historical time period, and any historical wound image contains at least one wound area; The at least one historical wound image is sorted based on a time sequence to obtain a first wound image sequence corresponding to the preset historical time period, and each historical wound image in the first wound image sequence is respectively compared with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each historical wound image, wherein the comparison of each historical wound image in the first wound image sequence with the original wound image based on a preset comparison rule to obtain a feature image corresponding to each historical wound image includes: aligning a first historical wound image in the first wound image sequence with the original wound image, and determining whether a wound area in the first historical wound image completely covers a wound area in the original wound image; If it is completely covered, removing the wound area in the original wound image from the first historical wound image to obtain a first feature image corresponding to the first historical wound image; If it is not completely covered, directly using the image formed by all the features in the first historical wound image as the first feature image corresponding to the first historical wound image; Acquire a real-time wound image, perform similarity analysis on the wound image and each historical wound image in the first wound image sequence according to a preset image recognition model, and select a historical wound image associated with the wound image according to the analysis result, wherein the associated historical wound image is the historical wound image in the first wound image sequence that has the highest similarity with the wound image; Acquire a characteristic image corresponding to the historical wound image, and crop a wound area of ​​the wound image according to a ratio of the characteristic image to the historical wound image to obtain a target wound image including a wound area to be identified, wherein cropping the wound area of ​​the wound image according to a ratio of the characteristic image to the historical wound image to obtain a target wound image including a wound area to be identified includes: Determining the number of predicted pixels in a predicted feature image corresponding to the wound image according to a ratio of the number of pixels in the feature image to the number of pixels in the historical wound image; According to the predicted number of pixels, a preset cutting rule is used to cut pixels in the wound area of ​​the wound image to obtain a target wound image including the wound area to be identified, wherein the cutting rule is to remove pixels in a direction from the periphery to the inside of the wound area of ​​the wound image; Based on a preset dynamic sliding window, a real wound area in the to-be-identified wound image is slid on the edge of the to-be-identified wound area, and the area of ​​the real wound area in the to-be-identified wound area is calculated according to the recognition result, wherein the dynamic sliding window dynamically adjusts the size of the dynamic sliding window according to the current proportion of target pixel points, the target pixel points are pixel points in the real wound area, and the area of ​​the real wound area in the to-be-identified wound area is calculated according to the recognition result, including: Acquire a first target number of all target pixels in the sliding area covered by the dynamic sliding window when sliding, and acquire a second target number of all target pixels in other areas of the wound area to be identified, wherein the other areas are areas of the wound area to be identified excluding the sliding area; The first target quantity and the second target quantity are superimposed to calculate the area of ​​the real wound region in the wound region to be identified.

2. A patient wound identification system for surgical care according to claim 1, characterized in that: The sliding of the preset dynamic sliding window on the edge of the wound area to be identified in the target wound image includes: Acquire a first number of target pixel points in the dynamic sliding window at a current moment, and adjust the size of the dynamic sliding window according to a ratio of the first number to the number of all pixel points in the dynamic sliding window, wherein there is a negatively correlated corresponding relationship between the size, the first number, and the ratio of the number of all pixel points in the dynamic sliding window; The resized dynamic sliding window continues to slide on the edge of the wound area to be identified, and determines that a second number of target pixel points in the dynamic sliding window at a next moment is greater than a preset number threshold; If it is not greater than the preset number threshold, the second number of target pixel points in the dynamic sliding window at the next moment is obtained, and the size of the dynamic sliding window is adjusted again according to the ratio of the second number to the number of all pixel points in the dynamic sliding window.

3. A patient wound identification system for surgical care according to claim 2, characterized in that: After determining that the second number of target pixel points in the dynamic sliding window at the next moment is greater than the preset number threshold, the system further performs the following steps: If the number is greater than a preset threshold, the size of the dynamic sliding window at the next moment is not adjusted.

Images