Postoperative stoma complication identification method and system based on artificial intelligence

Abstract

The invention provides a postoperative stoma complication identification method and system based on artificial intelligence, and relates to the technical field of image processing. The method comprises the following steps: acquiring a stoma image through image acquisition equipment; determining whether the stoma patient suffers from postoperative stoma complications or not; if the stoma patient suffers from the postoperative stoma complications, acquiring nutritional index data of the stoma patient by extracting blood of the stoma patient; inputting the nutrition index data and the stoma image into a trained stoma peripheral skin complication probability prediction model to obtain predicted stoma peripheral skin complication probability data; obtaining a stoma position and a stoma type; inputting the stoma position, the stoma type and the stoma image into a trained stoma complication probability prediction model to obtain predicted stoma complication probability data; and determining complication types. According to the method and the device, the types of the postoperative stoma complications can be identified, and the accuracy and the comprehensiveness of identifying the types of the postoperative stoma complications are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method and system for identifying postoperative stoma complications based on artificial intelligence. Background Art

[0002] In the related art, CN117982115A discloses an intelligent memory function stoma belt and a method of use, including: obtaining physiological data of stoma patients to determine whether the stoma patients are at risk of complications; further analyzing the physiological characterization values; analyzing the number of complications when different physiological characterization values ​​occur, thereby determining whether the stoma patients are at risk of complications; analyzing the historical data of target physiological data before each complication occurs, thereby evaluating the correlation between the target physiological data and the occurrence of complications; when the target physiological data related to the occurrence of complications is obtained, further analyzing the target physiological data, thereby determining whether the stoma patients are at risk of complications; this solution can establish personalized management for each patient, make the early warning more accurate and reliable, reduce the risks brought by complications, and thus improve the safety and rehabilitation effects of patients.

[0003] CN118557275A discloses a stoma system. The stoma system includes a stoma device. The stoma device can cut tissue to form a stoma. The stoma device is also used to electrically connect to an external energy source to ablate tissue around the stoma during and / or after the stoma is formed. By reinforcing the tissue around the stoma after the stoma is formed, the stoma can be prevented from retracting or closing, thereby ensuring a better postoperative effect.

[0004] Therefore, although the relevant technology can determine whether a stoma patient is at risk of complications, the relevant technology does not consider the impact of the predicted peristomal skin complication probability data and the predicted stoma complication probability data on the identification of postoperative stoma complication types, that is, it is impossible to identify the postoperative stoma complication type based on the predicted peristomal skin complication probability data and the predicted stoma complication probability data.

[0005] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The present invention provides a postoperative stoma complication identification method and system based on artificial intelligence, which can solve the technical problem that the related technology cannot identify the type of postoperative stoma complication based on the predicted peristomal skin complication probability data and the predicted stoma complication probability data.

[0007] According to a first aspect of the present invention, there is provided a method for identifying postoperative stoma complications based on artificial intelligence, comprising: acquiring a stoma image through an image acquisition device; determining whether a stoma patient suffers from a postoperative stoma complication based on the stoma image; if the stoma patient suffers from a postoperative stoma complication, obtaining nutritional index data of the stoma patient by drawing blood from the stoma patient; inputting the nutritional index data and the stoma image into a trained probability prediction model for peri-stomal skin complications to obtain predicted probability data for peri-stomal skin complications; acquiring a stoma position and a stoma type; inputting the stoma position, the stoma type and the stoma image into a trained probability prediction model for stoma complications to obtain predicted probability data for stoma complications; and determining the type of complication based on the predicted probability data for peri-stomal skin complications and the predicted probability data for stoma complications.

[0008] Furthermore, in the stoma image, the image detection model is used to identify whether the stoma height is within a preset range to determine the stoma height recognition result; in the stoma image, the image detection model is used to identify whether the stoma color is beef red or pink to determine the stoma color recognition result; in the stoma image, the image detection model is used to identify whether the stoma shape is circular or elliptical to determine the stoma shape recognition result; in the stoma image, the image detection model is used to identify whether the skin around the stoma is intact or flat to determine the skin around the stoma recognition result; based on the stoma height recognition result, the stoma color recognition result, the stoma shape recognition result and the stoma skin recognition result, the first stoma assessment parameter is determined; if the first stoma assessment parameter is less than a set threshold, it is determined that the postoperative stoma patient suffers from postoperative stoma complications; if the first stoma assessment parameter is greater than or equal to the set threshold, it is determined that the postoperative stoma patient does not have postoperative stoma complications.

[0009] Further, according to the stoma height recognition result, the stoma color recognition result, the stoma shape recognition result and the stoma peristomal skin recognition result, determining the first stoma judgment parameter includes: according to the formula Determine the first evaluation parameter of stoma ,in, This is the stoma height recognition result. This is the stoma color recognition result. is the stoma shape recognition result, This is the peristomal skin identification result. , , , , , , and The preset weights.

[0010] Furthermore, the training steps of the peristomal skin complication probability prediction model include: extracting a plurality of sample stoma patients to obtain sample nutritional index data, sample stoma images and sample peristomal skin complication probability data; determining a plurality of abnormal color pixels whose corresponding chromaticity is higher than a first preset chromaticity threshold among a plurality of pixel points of the sample stoma image; performing a connected domain analysis on the plurality of abnormal color pixels to obtain a first sample abnormal color region and determine the area of ​​the first sample abnormal color region; performing feature extraction processing on the peristomal skin region through the skin feature encoding level of the trained image recognition neural network model to obtain a sample feature vector of the peristomal skin in the sample stoma image; and A peristomal skin complication probability prediction model is developed to process the sample nutritional index data and the sample stoma image to obtain sample predicted peristomal skin complication probability data; a loss function of the peristomal skin complication probability prediction model is determined based on the sample nutritional index data, the area of ​​the first sample abnormal color region, the sample feature vector of the peristomal skin, the sample peristomal skin complication probability data and the sample predicted peristomal skin complication probability data; the peristomal skin complication probability prediction model is trained based on the loss function of the peristomal skin complication probability prediction model to obtain the trained peristomal skin complication probability prediction model.

[0011] Further, according to the sample nutrition index data, the area of ​​the first sample abnormal color region, the sample feature vector of the peristomal skin, the sample peristomal skin complication probability data and the sample predicted peristomal skin complication probability data, the loss function of the peristomal skin complication probability prediction model is determined, including: according to the formula Determining the loss function of the peristomal skin complication probability prediction model ,in, is the sample feature vector of the skin around the stoma in the sample stoma image of the jth sample stoma patient, is the feature vector of the skin around the stoma in the stoma image of the stoma patient, for The transposed vector of is the area of ​​the abnormal color region of the first sample of the jth sample stoma patient, is the area of ​​the first abnormal color region of the stoma patient, is the i-th sample nutritional index data of the j-th sample stoma patient, is the standard nutritional index data, is the probability data of peristomal skin complications of the jth sample stoma patient, Predict the probability data of peristomal skin complications for the jth sample of stoma patients, N is the number of nutritional indicator types, is the number of stoma patients in the xth batch of samples, U is the number of sample batches, i≤N, j≤ , x≤U, and i, j, x, N, and U are both positive integers.

[0012] Furthermore, the training step of the stoma complication probability prediction model includes: extracting multiple sample stoma patients to obtain sample stoma positions, sample stoma types, sample stoma images and sample stoma complication probability data; performing feature extraction processing on the stoma through the stoma feature encoding level of the trained image recognition neural network model to obtain the sample feature vector of the stoma in the sample stoma image; processing the sample stoma position, the sample stoma type and the sample stoma image through the stoma complication probability prediction model to obtain the sample predicted stoma complication probability data; determining the loss function of the stoma complication probability prediction model according to the sample stoma position, the sample stoma type, the sample feature vector of the stoma, the sample stoma complication probability data and the sample predicted stoma complication probability data; training the stoma complication probability prediction model according to the loss function of the stoma complication probability prediction model to obtain the trained stoma complication probability prediction model.

[0013] Further, according to the sample stoma location, the sample stoma type, the sample feature vector of the stoma, the sample stoma complication probability data and the sample predicted stoma complication probability data, the loss function of the stoma complication probability prediction model is determined, including: according to the formula Determining the loss function of the stoma complication probability prediction model ,in, is the sample stoma position of the jth sample stoma patient, The stoma position of stoma patients, for The transposed vector of is the sample stoma type of the jth sample stoma patient, For patients with a stoma, the type of stoma for The transposed vector of is the sample feature vector of the stoma in the sample stoma image of the jth sample stoma patient, is the feature vector of the stoma in the stoma image of the stoma patient, for The transposed vector of is the sample stoma complication probability data of the jth sample stoma patient, Predict the probability data of stoma complications for the jth sample of stoma patients, is the number of stoma patients in the xth batch of samples, U is the number of sample batches, j≤ , x≤U, and j, x, and U are both positive integers.

[0014] According to a second aspect of the present invention, there is provided a postoperative stoma complication identification system based on artificial intelligence, comprising: a stoma image module, for acquiring a stoma image by means of an image acquisition device; a judgment module, for determining whether a stoma patient suffers from a postoperative stoma complication based on the stoma image; a nutritional index data module, for acquiring nutritional index data of the stoma patient by drawing blood from the stoma patient if the stoma patient suffers from a postoperative stoma complication; a predicted peristomal skin complication probability data module, for inputting the nutritional index data and the stoma image into a trained peristomal skin complication probability prediction model to obtain predicted peristomal skin complication probability data; a stoma position and stoma type module, for acquiring the stoma position and the stoma type; a predicted stoma complication probability data module, for inputting the stoma position, the stoma type and the stoma image into a trained stoma complication probability prediction model to obtain predicted stoma complication probability data; a complication type module, for determining the complication type based on the predicted peristomal skin complication probability data and the predicted stoma complication probability data.

[0015] Technical effect: According to the present invention, by using stoma images to determine whether a stoma patient suffers from postoperative stoma complications, the inspection efficiency can be improved while reducing the probability of misidentification. Through information in multiple dimensions such as stoma images, nutritional index data, stoma location, and stoma type, the condition of stoma patients can be more comprehensively understood, thereby predicting the probability of peristomal skin complications and stoma complications. According to the predicted peristomal skin complication probability data and the predicted stoma complication probability data, the postoperative stoma complication type is identified to improve the accuracy and comprehensiveness of the postoperative stoma complication type identification. When determining the first stoma evaluation parameter, the stoma height recognition result, stoma color recognition result, stoma shape recognition result, and peristomal skin recognition result can be used to determine the first stoma evaluation parameter. The height, color, shape, and surrounding skin condition of the stoma can be detected by an artificial intelligence model, and whether the stoma patient suffers from postoperative stoma complications can be determined based on the detection results, which helps to more accurately reflect the overall health status of the stoma, thereby improving the comprehensiveness and accuracy of the first evaluation parameter. When determining the loss function of the peristomal skin complication probability prediction model, the influence of the sample nutritional index data on the probability of peristomal skin complications can be used to determine the influence of the above data on the error of the sample predicted peristomal skin complication probability data, thereby setting weights based on the influence and the relative error between the sample peristomal skin complication probability data and the sample predicted peristomal skin complication probability data, and based on the characteristic that the more similar the peristomal skin condition of the stoma patient is to the sample stoma patient, the greater the reference value, and the characteristic that the shorter the time interval with the first batch, the lower the accuracy, thereby setting weights for the errors output by the peristomal skin complication probability prediction model of multiple sample nutritional index data for each sample stoma patient in the xth batch to obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and improving the accuracy of the peristomal skin complication probability prediction model. When determining the loss function of the stoma complication probability prediction model, weights can be set based on the relative error between the sample stoma complication probability data and the sample predicted stoma complication probability data, and based on the characteristic that the more similar the stoma condition of the stoma patient is to the sample stoma patient, the greater the reference value, and the characteristic that the shorter the time interval with the first batch, the lower the accuracy. Thus, the errors output by the stoma complication probability prediction model for each sample stoma patient in the xth batch are weightedly summed to obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and improving the accuracy of the stoma complication probability prediction model.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic diagram of a process of a postoperative stoma complication identification method based on artificial intelligence according to an embodiment of the present invention is exemplarily shown;

[0019] Figure 2 A block diagram of a postoperative stoma complication identification system based on artificial intelligence according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0020] 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 only 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.

[0021] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0022] Figure 1 A flowchart of a method for identifying postoperative stoma complications based on artificial intelligence according to an embodiment of the present invention is exemplarily shown, wherein the method comprises:

[0023] Step S101, obtaining a stoma image through an image acquisition device;

[0024] Step S102, determining whether the stoma patient suffers from postoperative stoma complications according to the stoma image;

[0025] Step S103, if the stoma patient suffers from postoperative stoma complications, obtaining nutritional index data of the stoma patient by drawing blood from the stoma patient;

[0026] Step S104, inputting the nutrition index data and the stoma image into a trained peristomal skin complication probability prediction model to obtain predicted peristomal skin complication probability data;

[0027] Step S105, obtaining the stoma position and stoma type;

[0028] Step S106, inputting the stoma position, the stoma type and the stoma image into a trained stoma complication probability prediction model to obtain predicted stoma complication probability data;

[0029] Step S107, determining the type of complication according to the predicted peristomal skin complication probability data and the predicted stoma complication probability data.

[0030] According to the artificial intelligence-based postoperative stoma complication identification method of an embodiment of the present invention, whether a stoma patient suffers from postoperative stoma complications can be judged through stoma images, thereby improving the inspection efficiency and reducing the probability of misidentification. Through information in multiple dimensions such as stoma images, nutritional index data, stoma location, and stoma type, the condition of stoma patients can be more comprehensively understood, thereby predicting the probability of peristomal skin complications and stoma complications. Based on the predicted peristomal skin complication probability data and the predicted stoma complication probability data, the postoperative stoma complication type is identified, thereby improving the accuracy and comprehensiveness of the postoperative stoma complication type identification.

[0031] According to one embodiment of the present invention, in step S101, a high-definition camera can be used to aim at the stoma position of a stoma patient to take pictures, and the stoma image of the stoma patient can be collected, which can clearly capture the height, shape, color of the stoma and the condition of the surrounding skin.

[0032] According to an embodiment of the present invention, in step S102, it is determined whether the stoma patient suffers from postoperative stoma complications based on the stoma image.

[0033] According to one embodiment of the present invention, step S102 includes: in the stoma image, identifying whether the stoma height is within a preset range through an image detection model, and determining a stoma height recognition result; in the stoma image, identifying whether the stoma color is beef red or pink through an image detection model, and determining a stoma color recognition result; in the stoma image, identifying whether the stoma shape is circular or elliptical through an image detection model, and determining a stoma shape recognition result; in the stoma image, identifying whether the skin around the stoma is intact or flat through an image detection model, and determining a stoma skin recognition result; determining a first stoma assessment parameter based on the stoma height recognition result, the stoma color recognition result, the stoma shape recognition result, and the stoma skin recognition result; if the first stoma assessment parameter is less than a set threshold, determining that the postoperative stoma patient suffers from postoperative stoma complications; if the first stoma assessment parameter is greater than or equal to the set threshold, determining that the postoperative stoma patient has no postoperative stoma complications.

[0034] According to one embodiment of the present invention, the image detection model is a type of deep learning model. The image detection model is trained through historical data so that the image detection model can identify whether the stoma in the stoma image is a normal stoma. An abnormal stoma will cause peristomal skin complications and stoma complications. The image detection model is used to identify whether the stoma height is within the preset range. If the stoma height is within the preset range, the stoma height recognition result is 1; if the stoma height is not within the preset range, the stoma height recognition result is 0; the image detection model is used to identify whether the stoma color is beef red or pink; if the stoma color is beef red or pink, the stoma color recognition result is 1; if the stoma color is not beef red or pink, the stoma color recognition result is 0; the image detection model is used to identify whether the stoma shape is circular or elliptical; if the stoma shape is circular or elliptical, the stoma shape recognition result is 1; if the stoma shape is not circular or elliptical, the stoma shape recognition result is 0; the image detection model is used to identify whether the skin around the stoma is complete or flat; if the skin around the stoma is complete or flat, the stoma skin recognition result is 1; if the skin around the stoma is incomplete or uneven, the stoma skin recognition result is 0; the first stoma assessment parameter is determined based on the stoma height recognition result, stoma color recognition result, stoma shape recognition result and stoma skin recognition result. The first stoma judgment parameter is compared with the set threshold. If the first stoma judgment parameter is less than the set threshold, it is determined that the postoperative stoma patient suffers from postoperative stoma complications. If the first stoma judgment parameter is greater than or equal to the set threshold, it is determined that the postoperative stoma patient does not have postoperative stoma complications.

[0035] According to an embodiment of the present invention, determining the first stoma evaluation parameter according to the stoma height recognition result, the stoma color recognition result, the stoma shape recognition result and the stoma peristomal skin recognition result includes: determining the first stoma evaluation parameter according to formula (1): , (1),

[0036] in, This is the stoma height recognition result. This is the stoma color recognition result. is the stoma shape recognition result, This is the peristomal skin identification result. , , , , , , and The preset weights.

[0037] According to one embodiment of the present invention, in formula (1), This is the stoma height recognition result. When the stoma height is within the preset range, it means that the stoma height is similar to a normal stoma. The value is 1, otherwise, it means that the stoma height is abnormal and there is a possibility of postoperative stoma complications. The value of is 0, This is the stoma color recognition result. When the stoma color is beef red or pink, it means that the stoma color is similar to a normal stoma. The value is 1, otherwise, it means that the stoma color is abnormal and there is a possibility of postoperative stoma complications. The value of is 0, is the stoma shape recognition result. When the stoma shape is circular or elliptical, it means that the stoma shape is similar to a normal stoma. The value is 1, otherwise, it means that the stoma shape is abnormal and there is a possibility of postoperative stoma complications. The value of is 0, This is the peristomal skin identification result. When the peristomal skin is intact or flat, it means that the peristomal skin is similar to a normal stoma. The value is 1, otherwise it means that the skin around the stoma is abnormal and there is a possibility of postoperative stoma complications. The value of is 0. , , and The weighted sum can obtain the first stoma judgment parameter. The larger the first stoma judgment parameter is, the lower the possibility that the patient with postoperative stoma will suffer from postoperative stoma complications. The smaller the first stoma judgment parameter is, the higher the possibility that the patient with postoperative stoma will suffer from postoperative stoma complications.

[0038] In this way, the first assessment parameter of the stoma can be determined through the stoma height recognition results, stoma color recognition results, stoma shape recognition results and peristomal skin recognition results. The height, color, shape of the stoma and the condition of the surrounding skin can be detected by the artificial intelligence model, and whether the stoma patient suffers from postoperative stoma complications can be determined based on the test results. This helps to more accurately reflect the overall health status of the stoma, thereby improving the comprehensiveness and accuracy of the first assessment parameter.

[0039] According to one embodiment of the present invention, in step S103, when it is confirmed that the stoma patient has postoperative stoma complications, blood sampling is prepared to accurately measure various nutrients (e.g., hemoglobin, albumin, electrolytes, etc.) in the stoma patient's blood, thereby obtaining nutritional index data of the stoma patient.

[0040] According to one embodiment of the present invention, in step S104, the nutritional indicator data and the stoma image are input into a trained peristomal skin complication probability prediction model. The peristomal skin complication probability prediction model may be a neural network model, which is trained based on a large amount of sample data through machine learning or deep learning technology to obtain predicted peristomal skin complication probability data.

[0041] According to one embodiment of the present invention, the training step of the peristomal skin complication probability prediction model includes: extracting a plurality of sample stoma patients to obtain sample nutritional index data, sample stoma images and sample peristomal skin complication probability data; determining a plurality of abnormal color pixels whose corresponding chromaticity is higher than a first preset chromaticity threshold among a plurality of pixel points of the sample stoma image; performing a connected domain analysis on the plurality of abnormal color pixels to obtain a first sample abnormal color region and determine the area of ​​the first sample abnormal color region; performing feature extraction processing on the peristomal skin region through the skin feature encoding level of the trained image recognition neural network model to obtain a sample feature vector of the peristomal skin in the sample stoma image. ; The sample nutritional index data and the sample stoma image are processed through the peristomal skin complication probability prediction model to obtain the sample predicted peristomal skin complication probability data; the loss function of the peristomal skin complication probability prediction model is determined according to the sample nutritional index data, the area of ​​the first sample abnormal color region, the sample feature vector of the peristomal skin, the sample peristomal skin complication probability data and the sample predicted peristomal skin complication probability data; according to the loss function of the peristomal skin complication probability prediction model, the peristomal skin complication probability prediction model is trained to obtain the trained peristomal skin complication probability prediction model.

[0042] According to one embodiment of the present invention, the extracted multiple sample stoma patients are divided into different batches of sample stoma patients, and the number of sample stoma patients in each batch is the same, so as to train the probability prediction model of peristomal skin complications of different batches, for example, , M is the number of stoma patients in the sample, , , …, are the number of sample stoma patients in the 1st, 2nd, ..., U batches respectively. For each sample stoma patient, the sample stoma skin complication probability data can be obtained from the recorded probability of occurrence of peristomal skin complications. For example, if the sample stoma patient suffers from peristomal skin complications, the sample stoma skin complication probability data of the sample stoma patient is 1, and if the sample stoma patient has no peristomal skin complications, the sample stoma skin complication probability data of the sample stoma patient is 0. The first preset chromaticity threshold can be set to be 20% or 30% higher than the chromaticity of normal skin color, etc., and the present invention is not limited to this. The connectivity relationship between multiple abnormal color pixels can be determined to form a connected domain, wherein the connected domain refers to a set of abnormal color pixels with mutual connection or adjacent relationship, and the boundary and range of the connected domain can be used as the boundary and range of the first sample abnormal color area, based on which, the area of ​​the first sample abnormal color area is determined. The image recognition neural network model can be a convolutional neural network model. The skin feature encoding layer of the image recognition neural network model is used to extract features from the peristomal skin area to obtain a sample feature vector of the peristomal skin in the sample stoma image. The sample feature vector of the peristomal skin usually includes various features of the peristomal skin of stoma patients, such as skin color, skin smoothness, etc., which can be used to analyze and identify the condition of the peristomal skin. The sample nutritional index data will affect the probability of peristomal skin complications. For example, when the sample nutritional index data is not up to standard, it may lead to poor healing of the peristomal skin and increase the incidence of complications such as fungal infection and dermatitis. The peristomal skin complication probability prediction model can predict the sample predicted peristomal skin complication probability data based on the relationship between the above-mentioned sample nutritional index data and the probability of peristomal skin complications, based on the area of ​​the first sample abnormal color area and the sample feature vector of the peristomal skin. The loss function is determined according to the relative difference between the sample predicted peristomal skin complication probability data and the sample peristomal skin complication probability data recorded by the sample. By feedback-adjusting the loss function, the trained peristomal skin complication probability prediction model is obtained.

[0043] According to one embodiment of the present invention, the loss function of the peristomal skin complication probability prediction model is determined based on the sample nutritional index data, the area of ​​the first sample abnormal color region, the sample feature vector of the peristomal skin, the sample peristomal skin complication probability data and the sample predicted peristomal skin complication probability data, including: determining the loss function of the peristomal skin complication probability prediction model according to formula (2): , (2),

[0044] in, is the sample feature vector of the skin around the stoma in the sample stoma image of the jth sample stoma patient, is the feature vector of the skin around the stoma in the stoma image of the stoma patient, for The transposed vector of is the area of ​​the abnormal color region of the first sample of the jth sample stoma patient, is the area of ​​the first abnormal color region of the stoma patient, is the i-th sample nutritional index data of the j-th sample stoma patient, is the standard nutritional index data, is the probability data of peristomal skin complications of the jth sample stoma patient, Predict the probability data of peristomal skin complications for the jth sample of stoma patients, N is the number of nutritional indicator types, is the number of stoma patients in the xth batch of samples, U is the number of sample batches, i≤N, j≤ , x≤U, and i, j, x, N, and U are both positive integers.

[0045] According to one embodiment of the present invention, in formula (2), The relative error between the sample peristomal skin complication probability data of the j-th sample stoma patient and the sample predicted peristomal skin complication probability data of the j-th sample stoma patient. is the relative difference between the i-th sample nutritional index data and the standard nutritional index data of the j-th sample stoma patient. The larger the relative difference, the greater the probability of peristomal skin complications, that is, the sample nutritional index data is positively correlated with the sample peristomal skin complications probability data of the sample stoma patient. When the sample nutritional index data does not meet the standard, it may lead to poor healing of the peristomal skin and increase the probability of complications such as fungal infection and dermatitis. Therefore, the larger the sample nutritional index data is relative to the standard nutritional index data, that is, The larger the value is, the greater the impact on the error of the sample predicted peristomal skin complications probability data. is the similarity between the sample feature vector of the peristomal skin in the sample stoma image of the jth sample stoma patient and the feature vector of the peristomal skin in the stoma image of the stoma patient. It is the relative difference between the area of ​​the first abnormal color region of the stoma patient and the area of ​​the first sample abnormal color region of the jth sample stoma patient. is the similarity between the first abnormal color area of ​​the stoma patient and the first sample abnormal color area of ​​the jth sample stoma patient. In order to achieve a similar peristomal skin complication probability detection effect, if the feature vector and the first abnormal color area of ​​the peristomal skin in the stoma image of the stoma patient are closer to the sample feature vector and the first abnormal color area of ​​the peristomal skin in the sample stoma image of the sample stoma patient, that is, and The larger the value of is, the more similar the skin condition around the stoma in the stoma image of the stoma patient is to the skin condition around the stoma in the sample stoma image of the sample stoma patient, and the greater its reference value is, and therefore, the higher its weight is. It represents the sum of the relative errors of the probability of peristomal skin complications of multiple sample nutritional index data of each sample ostomy patient in the xth batch multiplied by the corresponding weight. is the weight of the xth batch, which is used to reasonably weight the relative errors of different training batches in the loss function. For the x+1th batch of sample stoma patients, the accuracy of the peristomal skin complication probability data output by the peristomal skin complication probability prediction model is usually higher than the accuracy of the peristomal skin complication probability data output by the x+1th batch, that is, the shorter the time interval between a batch and the first batch, the less accurate the prediction result. To improve the training efficiency, the higher the weight is set. Conversely, the more accurate the prediction result, the lower the weight is. Therefore, a higher weight can be given to items with lower accuracy, thereby improving the training intensity and training efficiency. and The product of the two items can represent the loss function of the peristomal skin complication probability prediction model.

[0046] According to one embodiment of the present invention, in the process of training a peristomal skin complication probability prediction model, the loss function is back-propagated and some parameters within the model are adjusted to reduce the value of the loss function of the peristomal skin complication probability prediction model, thereby improving the accuracy of the peristomal skin complication probability prediction model and obtaining a trained peristomal skin complication probability prediction model.

[0047] In this way, the influence of the sample nutritional index data on the probability of peristomal skin complications can be used to determine the influence of the above data on the error of the sample predicted peristomal skin complications probability data, thereby setting weights based on the influence and the relative error between the sample peristomal skin complications probability data and the sample predicted peristomal skin complications probability data, and based on the characteristics that the more similar the peristomal skin condition of the stoma patient is to the sample stoma patient, the greater the reference value, and the shorter the time interval with the first batch, the lower the accuracy, thereby setting weights for the errors output by the peristomal skin complications probability prediction model of multiple sample nutritional index data of each sample stoma patient in the xth batch, and obtaining a loss function to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and improving the accuracy of the peristomal skin complications probability prediction model.

[0048] According to one embodiment of the present invention, in step S105, the stoma position can be located in multiple parts of the body, such as the abdomen, perineum, etc., and the stoma type can be divided into multiple types according to its function and morphology, such as ileostomy, colostomy, urostomy, etc.

[0049] According to one embodiment of the present invention, in step S106, the stoma location, stoma type and stoma image are input into a trained stoma complication probability prediction model. The stoma complication probability prediction model may be a neural network model, which is trained based on a large amount of sample data through machine learning or deep learning technology to obtain predicted stoma complication probability data.

[0050] According to one embodiment of the present invention, the training step of the stoma complication probability prediction model includes: extracting a plurality of sample stoma patients to obtain sample stoma positions, sample stoma types, sample stoma images and sample stoma complication probability data; performing feature extraction processing on the stoma through the stoma feature encoding level of the trained image recognition neural network model to obtain a sample feature vector of the stoma in the sample stoma image; processing the sample stoma position, the sample stoma type and the sample stoma image through the stoma complication probability prediction model to obtain sample predicted stoma complication probability data; determining the loss function of the stoma complication probability prediction model according to the sample stoma position, the sample stoma type, the sample feature vector of the stoma, the sample stoma complication probability data and the sample predicted stoma complication probability data; training the stoma complication probability prediction model according to the loss function of the stoma complication probability prediction model to obtain the trained stoma complication probability prediction model.

[0051] According to one embodiment of the present invention, the multiple sample stoma patients are divided into different batches of sample stoma patients, and the number of sample stoma patients in each batch is the same, so as to train stoma complication probability prediction models for different batches, for example, , M is the number of stoma patients in the sample, , , …, are the number of sample stoma patients in the 1st, 2nd, …, U batches respectively. For each sample stoma patient, the sample stoma complication probability data can be obtained from the recorded probability of occurrence of stoma complications. For example, if the sample stoma patient suffers from stoma complications, the sample stoma complication probability data of the sample stoma patient is 1, and if the sample stoma patient has no stoma complications, the sample stoma complication probability data of the sample stoma patient is 0. The image recognition neural network model can be a convolutional neural network model. The stoma feature encoding layer of the image recognition neural network model is used to perform feature extraction processing on the stoma to obtain a sample feature vector of the stoma in the sample stoma image. The sample feature vector of the stoma generally includes stoma color, stoma shape, secretion properties, etc., which can be used to analyze and identify stoma health conditions. The stoma complication probability prediction model can predict the sample predicted stoma complication probability data based on the sample stoma location, sample stoma type and the sample feature vector of the stoma. The loss function is determined according to the relative difference between the sample predicted stoma complication probability data and the sample stoma complication probability data of the sample record. By feedback adjusting the loss function, the trained stoma complication probability prediction model is obtained.

[0052] According to one embodiment of the present invention, the loss function of the stoma complication probability prediction model is determined according to the sample stoma location, the sample stoma type, the sample feature vector of the stoma, the sample stoma complication probability data and the sample predicted stoma complication probability data, including: determining the loss function of the stoma complication probability prediction model according to formula (3): , (3),

[0053] in, is the sample stoma position of the jth sample stoma patient, The stoma position of stoma patients, for The transposed vector of is the sample stoma type of the jth sample stoma patient, For patients with a stoma, the type of stoma for The transposed vector of is the sample feature vector of the stoma in the sample stoma image of the jth sample stoma patient, is the feature vector of the stoma in the stoma image of the stoma patient, for The transposed vector of is the sample stoma complication probability data of the jth sample stoma patient, Predict the probability data of stoma complications for the jth sample of stoma patients, is the number of stoma patients in the xth batch of samples, U is the number of sample batches, j≤ , x≤U, and j, x, and U are both positive integers.

[0054] According to one embodiment of the present invention, in formula (3), The relative error between the sample stoma complication probability data of the j-th sample stoma patient and the sample predicted stoma complication probability data of the j-th sample stoma patient. is the similarity between the sample stoma position of the jth sample stoma patient and the stoma position of the stoma patient. is the similarity between the sample stoma type of the jth sample stoma patient and the stoma type of the stoma patient. is the similarity between the sample feature vector of the stoma in the sample stoma image of the jth sample stoma patient and the feature vector of the stoma in the stoma image of the stoma patient. In order to achieve a similar stoma complication probability detection effect, if the stoma position, stoma type, and stoma feature vector of the stoma patient are closer to the sample stoma position, sample stoma type, and stoma sample feature vector of the sample stoma patient, that is, , and The larger the value of , the more similar the stoma condition of the stoma patient is to that of the sample stoma patient, the greater its reference value, and therefore, the higher its weight. is the weight of the x-th batch, which is used to reasonably weight the relative errors of different training batches in the loss function. For the x+1-th batch of sample ostomy patients, the accuracy of the peristomal skin complication probability data output by the peristomal skin complication probability prediction model is usually higher than the accuracy of the peristomal skin complication probability data output by the x+1-th batch, that is, the shorter the time interval between a batch and the first batch, the less accurate the prediction result. In order to improve the training efficiency, the higher the weight is set. Conversely, the more accurate the prediction result, the lower the weight is. Therefore, a higher weight can be given to items with lower accuracy, thereby improving the training intensity and training efficiency.

[0055] According to one embodiment of the present invention, using , , and , the relative error of the probability of stoma complications of each sample stoma patient in the xth batch is weighted and summed to obtain the training loss function. In the process of training the stoma complication probability prediction model, the loss function is back-propagated and some parameters inside the model are adjusted to reduce the value of the loss function of the stoma complication probability prediction model, thereby improving the accuracy of the stoma complication probability prediction model and obtaining the trained stoma complication probability prediction model.

[0056] In this way, weights can be set based on the relative error between the sample stoma complication probability data and the sample predicted stoma complication probability data, and based on the characteristic that the more similar the stoma condition of the stoma patient is to the sample stoma patient, the greater the reference value, and the characteristic that the shorter the time interval with the first batch, the lower the accuracy. In this way, the errors output by the stoma complication probability prediction model for each sample stoma patient in the xth batch are weightedly summed to obtain a loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and improving the accuracy of the stoma complication probability prediction model.

[0057] According to one embodiment of the present invention, in step S107, when the predicted peristomal skin complication probability data is greater than the preset peristomal skin complication probability data, and the predicted peristomal complication probability data is less than or equal to the preset peristomal complication probability data, the complication type is peristomal skin complication. When the predicted peristomal skin complication probability data is less than or equal to the preset peristomal skin complication probability data, and the predicted peristomal complication probability data is greater than the preset peristomal complication probability data, the complication type is stoma complication. When the predicted peristomal skin complication probability data is greater than the preset peristomal skin complication probability data, and the predicted peristomal complication probability data is greater than the preset peristomal complication probability data, the complication types are peristomal skin complication and stoma complication.

[0058] According to the artificial intelligence-based postoperative stoma complication identification method of the embodiment of the present invention, whether the stoma patient suffers from postoperative stoma complications can be judged through the stoma image, which can improve the inspection efficiency and reduce the probability of misidentification. Through the information of multiple dimensions of stoma image, nutrition index data, stoma location and stoma type, the condition of the stoma patient can be more comprehensively understood, so as to predict the probability of peristomal skin complications and stoma complications. According to the predicted peristomal skin complication probability data and the predicted stoma complication probability data, the postoperative stoma complication type is identified to improve the accuracy and comprehensiveness of the postoperative stoma complication type identification. When determining the first stoma judgment parameter, the stoma height recognition result, the stoma color recognition result, the stoma shape recognition result and the peristomal skin recognition result can be used to determine the first stoma judgment parameter, and the stoma height, color, shape and surrounding skin condition of the stoma can be detected by the artificial intelligence model, and whether the stoma patient suffers from postoperative stoma complications can be determined according to the detection results, which helps to more accurately reflect the overall health status of the stoma, thereby improving the comprehensiveness and accuracy of the first judgment parameter. When determining the loss function of the peristomal skin complication probability prediction model, the influence of the sample nutritional index data on the probability of peristomal skin complications can be used to determine the influence of the above data on the error of the sample predicted peristomal skin complication probability data, thereby setting weights based on the influence and the relative error between the sample peristomal skin complication probability data and the sample predicted peristomal skin complication probability data, and based on the characteristic that the more similar the peristomal skin condition of the stoma patient is to the sample stoma patient, the greater the reference value, and the characteristic that the shorter the time interval with the first batch, the lower the accuracy, thereby setting weights for the errors output by the peristomal skin complication probability prediction model of multiple sample nutritional index data for each sample stoma patient in the xth batch to obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and improving the accuracy of the peristomal skin complication probability prediction model. When determining the loss function of the stoma complication probability prediction model, weights can be set based on the relative error between the sample stoma complication probability data and the sample predicted stoma complication probability data, and based on the characteristic that the more similar the stoma condition of the stoma patient is to the sample stoma patient, the greater the reference value, and the characteristic that the shorter the time interval with the first batch, the lower the accuracy. Thus, the errors output by the stoma complication probability prediction model for each sample stoma patient in the xth batch are weightedly summed to obtain the loss function, so as to improve the design accuracy and objectivity of the loss function, thereby improving the training efficiency during the training process and improving the accuracy of the stoma complication probability prediction model.

[0059] Figure 2 A block diagram of a postoperative stoma complication identification system based on artificial intelligence according to an embodiment of the present invention is exemplarily shown, wherein the system comprises:

[0060] A stoma image module, used for acquiring a stoma image through an image acquisition device;

[0061] A judgment module, used for determining whether the stoma patient suffers from postoperative stoma complications according to the stoma image;

[0062] A nutrition index data module is used to obtain nutrition index data of a stoma patient by drawing blood from the stoma patient if the stoma patient suffers from postoperative stoma complications;

[0063] A module for predicting the probability data of peristomal skin complications, used for inputting the nutrition index data and the stoma image into a trained peristomal skin complication probability prediction model to obtain predicted peristomal skin complication probability data;

[0064] A stoma location and stoma type module, used to obtain the stoma location and stoma type;

[0065] A stoma complication probability prediction data module is used to input the stoma position, the stoma type and the stoma image into a trained stoma complication probability prediction model to obtain stoma complication probability prediction data;

[0066] The complication type module is used to determine the complication type based on the predicted peristomal skin complication probability data and the predicted stoma complication probability data.

[0067] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0068] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

Claims

1. A method for identifying postoperative stoma complications based on artificial intelligence, characterized in that: include: Obtain stoma images through image acquisition equipment; determining whether the stoma patient suffers from postoperative stoma complications according to the stoma image; If the stoma patient suffers from postoperative stoma complications, the nutritional index data of the stoma patient will be obtained by drawing blood from the stoma patient; Inputting the nutrition index data and the stoma image into a trained peristomal skin complication probability prediction model to obtain predicted peristomal skin complication probability data; Obtain stoma location and stoma type; Inputting the stoma position, the stoma type and the stoma image into a trained stoma complication probability prediction model to obtain predicted stoma complication probability data; The type of complication is determined according to the predicted peristomal skin complication probability data and the predicted stoma complication probability data.

2. The method for identifying postoperative stoma complications based on artificial intelligence according to claim 1, characterized in that: According to the stoma image, determining whether the postoperative stoma patient suffers from postoperative stoma complications, including: In the stoma image, identifying whether the stoma height is within a preset range through an image detection model, and determining a stoma height recognition result; In the stoma image, identifying whether the stoma color is beef red or pink by an image detection model, and determining a stoma color recognition result; In the stoma image, identifying whether the stoma shape is circular or elliptical through an image detection model, and determining a stoma shape recognition result; In the stoma image, identifying whether the peristomal skin is intact or flat by using an image detection model, and determining a peristomal skin recognition result; Determining a first stoma assessment parameter according to the stoma height recognition result, the stoma color recognition result, the stoma shape recognition result, and the stoma peristomal skin recognition result; If the stoma first evaluation parameter is less than a set threshold, it is determined that the postoperative stoma patient suffers from postoperative stoma complications; If the first stoma assessment parameter is greater than or equal to a set threshold, it is determined that the postoperative stoma patient has no postoperative stoma complications.

3. The artificial intelligence-based postoperative stoma complication identification method according to claim 2, characterized in that: Determining a first stoma assessment parameter according to the stoma height recognition result, the stoma color recognition result, the stoma shape recognition result, and the stoma peristomal skin recognition result includes: According to the formula ， Determine the first evaluation parameter of stoma ,in, This is the stoma height recognition result. This is the stoma color recognition result. is the stoma shape recognition result, This is the peristomal skin identification result. , , , , , , and The preset weights.

4. The method for identifying postoperative stoma complications based on artificial intelligence according to claim 1, characterized in that: The training steps of the peristomal skin complication probability prediction model include: By sampling multiple sample stoma patients, sample nutritional index data, sample stoma images and sample peristomal skin complication probability data are obtained; Determine, among the plurality of pixels of the sample stoma image, a plurality of abnormal color pixels whose corresponding chromaticity is higher than a first preset chromaticity threshold; Performing a connected domain analysis on the multiple abnormal color pixels to obtain a first sample abnormal color region and determine the area of ​​the first sample abnormal color region; Through the skin feature encoding layer of the trained image recognition neural network model, feature extraction processing is performed on the peristomal skin area to obtain a sample feature vector of the peristomal skin in the sample stoma image; The sample nutrition index data and the sample stoma image are processed by a peristomal skin complication probability prediction model to obtain sample predicted peristomal skin complication probability data; Determine the loss function of the peristomal skin complication probability prediction model according to the sample nutrition index data, the area of ​​the first sample abnormal color region, the sample feature vector of the peristomal skin, the sample peristomal skin complication probability data and the sample predicted peristomal skin complication probability data; According to the loss function of the peristomal skin complications probability prediction model, the peristomal skin complications probability prediction model is trained to obtain the trained peristomal skin complications probability prediction model.

5. The artificial intelligence-based postoperative stoma complication identification method according to claim 4, characterized in that: Determining a loss function of the peristomal skin complication probability prediction model according to the sample nutritional index data, the area of ​​the first sample abnormal color region, the sample feature vector of the peristomal skin, the sample peristomal skin complication probability data, and the sample predicted peristomal skin complication probability data includes: According to the formula ， Determining the loss function of the peristomal skin complication probability prediction model ,in, is the sample feature vector of the skin around the stoma in the sample stoma image of the jth sample stoma patient, is the feature vector of the skin around the stoma in the stoma image of the stoma patient, for The transposed vector of is the area of ​​the abnormal color region of the first sample of the jth sample stoma patient, is the area of ​​the first abnormal color region of the stoma patient, is the i-th sample nutritional index data of the j-th sample stoma patient, is the standard nutritional index data, is the probability data of peristomal skin complications of the jth sample stoma patient, Predict the probability data of peristomal skin complications for the jth sample of stoma patients, N is the number of nutritional indicator types, is the number of stoma patients in the xth batch of samples, U is the number of sample batches, i≤N, j≤ , x≤U, and i, j, x, N, and U are both positive integers.

6. The method for identifying postoperative stoma complications based on artificial intelligence according to claim 1, characterized in that: The training steps of the stoma complication probability prediction model include: By sampling multiple sample stoma patients, sample stoma locations, sample stoma types, sample stoma images and sample stoma complication probability data are obtained; Through the stoma feature encoding layer of the trained image recognition neural network model, feature extraction processing is performed on the stoma to obtain a sample feature vector of the stoma in the sample stoma image; The sample stoma position, the sample stoma type and the sample stoma image are processed by a stoma complication probability prediction model to obtain sample predicted stoma complication probability data; Determining a loss function of the stoma complication probability prediction model according to the sample stoma location, the sample stoma type, the sample feature vector of the stoma, the sample stoma complication probability data, and the sample predicted stoma complication probability data; The stoma complication probability prediction model is trained according to the loss function of the stoma complication probability prediction model to obtain the trained stoma complication probability prediction model.

7. The artificial intelligence-based postoperative stoma complication identification method according to claim 6, characterized in that: Determining a loss function of the stoma complication probability prediction model according to the sample stoma position, the sample stoma type, the sample feature vector of the stoma, the sample stoma complication probability data, and the sample predicted stoma complication probability data includes: According to the formula ， Determining the loss function of the stoma complication probability prediction model ,in, is the sample stoma position of the jth sample stoma patient, The stoma position of stoma patients, for The transposed vector of is the sample stoma type of the jth sample stoma patient, For patients with a stoma, the type of stoma for The transposed vector of is the sample feature vector of the stoma in the sample stoma image of the jth sample stoma patient, is the feature vector of the stoma in the stoma image of the stoma patient, for The transposed vector of is the sample stoma complication probability data of the jth sample stoma patient, Predict the probability data of stoma complications for the jth sample of stoma patients, is the number of stoma patients in the xth batch of samples, U is the number of sample batches, j≤ , x≤U, and j, x, and U are both positive integers.

8. An artificial intelligence-based postoperative stoma complication identification system for executing the artificial intelligence-based postoperative stoma complication identification method according to any one of claims 1 to 7, characterized in that: include: A stoma image module, used for acquiring a stoma image through an image acquisition device; A judgment module, used for determining whether the stoma patient suffers from postoperative stoma complications according to the stoma image; A nutrition index data module is used to obtain nutrition index data of a stoma patient by drawing blood from the stoma patient if the stoma patient suffers from postoperative stoma complications; A module for predicting the probability data of peristomal skin complications, used for inputting the nutrition index data and the stoma image into a trained peristomal skin complication probability prediction model to obtain predicted peristomal skin complication probability data; A stoma location and stoma type module, used to obtain the stoma location and stoma type; A stoma complication probability prediction data module is used to input the stoma position, the stoma type and the stoma image into a trained stoma complication probability prediction model to obtain stoma complication probability prediction data; The complication type module is used to determine the complication type based on the predicted peristomal skin complication probability data and the predicted stoma complication probability data.

Images