Wound monitoring system

Abstract

The present invention relates to the field of medical equipment, and in particular to a wound monitoring system. A wound data acquisition device collects current wound data of a target wound; an electronic device processes the current wound data, determines the current wound bacterial load and the current wound infection type corresponding to the target wound, and determines a monitoring result; a server stores the current wound data, the current wound bacterial load, the current wound infection type, the target wound, and the target wound patient corresponding to the target wound, and displays the current wound data, the current wound bacterial load, and the current wound infection type to a target user; the server performs statistical analysis on the current wound infection type corresponding to each target wound patient, determines the wound infection characteristics corresponding to each target wound patient group, and generates a corresponding population wound care method. This ensures the accuracy of the current wound data obtained, and enables the processing and tracing of the current wound data.

Description

Technical Field

[0001] The present invention relates to the field of medical equipment, and in particular to a wound monitoring system. Background Art

[0002] Infection is one of the main reasons for prolonged wound healing, which brings huge economic burden to patients and significantly affects their quality of life.

[0003] Currently, wound infection and the need for wound debridement and dressing changes are often determined based on subjective judgment by medical staff. However, empirical dressing changes interfere with the normal wound healing process and increase the risk of infection.

[0004] In addition, the above method cannot generate accurate wound surface data, so the wound surface data cannot be stored, and thus the wound surface data cannot be uniformly processed and traced. Summary of the Invention

[0005] In view of this, the present invention provides a wound monitoring system to solve the problem that accurate wound data cannot be generated, wound data cannot be stored, and wound data cannot be uniformly processed and traced.

[0006] In a first aspect, the present invention provides a wound monitoring system, comprising: a wound data acquisition device, an electronic device, and a server, wherein the wound data acquisition device is communicatively connected to the electronic device, and the electronic device is communicatively connected to the server, wherein:

[0007] Wound surface data acquisition equipment, used to collect current wound surface data of the target wound surface and transmit the current wound surface data to the electronic device;

[0008] An electronic device is used to receive current wound surface data, process the current wound surface data, determine the current wound surface bacterial load and the current wound surface infection type corresponding to the target wound surface; determine the monitoring result corresponding to the target wound surface based on the current wound surface bacterial load, and send the current wound surface data, current wound surface bacterial load and current wound surface infection type corresponding to the target wound surface to a server;

[0009] The server is used to store the current wound data, the current wound bacterial load, the current wound infection type, the target wound, and the target wound patient corresponding to the target wound, so as to facilitate the tracing of all wound data corresponding to the target wound, and to display the current wound data, the current wound bacterial load, and the current wound infection type to the target user;

[0010] The server is also used to perform statistical analysis on the current wound infection type corresponding to each target wound patient, determine the wound infection characteristics corresponding to each target wound patient group, and generate corresponding wound care methods for the population based on the wound infection characteristics.

[0011] The wound monitoring system provided in the embodiments of the present application includes a wound data acquisition device for collecting current wound data of a target wound and transmitting the current wound data to an electronic device, thereby enabling the electronic device to obtain the current wound data of the target wound. The electronic device is configured to receive the current wound data and process the current wound data to determine the current wound bacterial load and the current wound infection type corresponding to the target wound, thereby ensuring the accuracy of the determined current wound bacterial load and the current wound infection type. The device can then determine the monitoring result corresponding to the target wound based on the current wound bacterial load, thereby ensuring the accuracy of the determined monitoring result. The device also transmits the current wound data, the current wound bacterial load, and the current wound infection type corresponding to the target wound to a server, thereby enabling the server to store the current wound data, the current wound bacterial load, and the current wound infection type corresponding to the target wound. The server is configured to store the current wound data, the current wound bacterial load, and the current wound infection type corresponding to the target wound and the target wound patient corresponding to the target wound, thereby enabling the storage of the wound data of the target wound and facilitating the tracing of all wound data corresponding to the target wound. The current wound data, current wound bacterial load, and current wound infection type are displayed to the target user, allowing the target user to receive the current wound bacterial load and current wound infection type. The server is also used to perform statistical analysis on the current wound infection type corresponding to each target wound patient, determine the wound infection characteristics corresponding to each target wound patient group, and ensure the accuracy of the wound infection characteristics corresponding to each target wound patient group. Based on the wound infection characteristics, the corresponding population wound care method is generated, ensuring the accuracy of the generated population wound care method, and then care can be provided to specific populations based on the population wound care method.

[0012] In an optional embodiment, the wound surface data collection device includes: at least one of a temperature collection component, an infiltration humidity collection component, a pH value collection component, a lactate collection component, a uric acid collection component, a glucose collection component, and a cytokine and inflammation-related protein collection component;

[0013] A temperature collection component is used to collect the current wound temperature of the target wound surface;

[0014] The infiltration humidity collection component is used to collect the current infiltration humidity of the target wound surface;

[0015] A pH value collection component is used to collect the current pH value of the target wound surface;

[0016] A lactic acid collection component is used to collect the current wound lactic acid of the target wound;

[0017] A uric acid collection component is used to collect the current uric acid of the target wound surface;

[0018] A glucose collection component is used to collect the current wound surface glucose of the target wound surface;

[0019] The cytokine and inflammation-related protein collection component is used to collect the current wound inflammatory factors of the target wound.

[0020] The wound monitoring system provided in the embodiment of the present application, the wound data acquisition equipment includes: at least one of a temperature acquisition component, an infiltration humidity acquisition component, a pH value acquisition component, a lactate acquisition component, a uric acid acquisition component, a glucose acquisition component, and a cytokine and inflammation-related protein acquisition component; the temperature acquisition component is used to collect the current wound temperature of the target wound; the infiltration humidity acquisition component is used to collect the current infiltration humidity of the target wound; the pH value acquisition component is used to collect the current wound pH value of the target wound; the lactate acquisition component is used to collect the current wound lactic acid of the target wound; the uric acid acquisition component is used to collect the current wound uric acid of the target wound; the glucose acquisition component is used to collect the current wound glucose of the target wound; the cytokine and inflammation-related protein acquisition component is used to collect the current wound inflammatory factors of the target wound, so that one or more physiological parameters of the current wound temperature, current infiltration humidity, current wound pH value, current wound lactic acid, current wound uric acid, current wound glucose, and current wound inflammatory factors corresponding to the target wound can be collected, and the accuracy of the collected physiological parameters can be guaranteed.

[0021] In an optional embodiment, the cytokine and inflammation-related protein collection component includes at least one of a matrix metalloproteinase 9 sensor, a matrix metalloproteinase inhibitor 1 sensor, an interleukin 1β sensor, a tumor necrosis factor α sensor, and an interleukin 6 sensor.

[0022] The wound monitoring system provided in the embodiment of the present application, the cytokine and inflammation-related protein collection component includes at least one of a matrix metalloproteinase 9 sensor, a matrix metalloproteinase inhibitor 1 sensor, an interleukin 1β sensor, a tumor necrosis factor α sensor and an interleukin 6 sensor, thereby ensuring that at least one of matrix metalloproteinase 9, matrix metalloproteinase inhibitor 1, interleukin 1β, tumor necrosis factor α and interleukin 6 is collected.

[0023] In an optional embodiment, the current wound data includes one or more physiological parameters of the current wound temperature, current infiltration humidity, current wound pH value, current wound lactic acid, current wound uric acid, current wound glucose, and current wound inflammatory factors. The electronic device is used to obtain the target site corresponding to the target wound; obtain the target wound infection determination model corresponding to the target site based on the correspondence between the site and the wound infection determination model; input the current wound data into the target wound infection determination model, and the target wound infection determination model performs feature extraction on the current wound data, and outputs the current wound bacterial load and the current wound infection type based on the extracted features.

[0024] The wound monitoring system provided in the embodiments of the present application comprises an electronic device for obtaining a target site corresponding to a target wound; and obtaining a target wound infection determination model corresponding to the target site based on the correspondence between the site and the wound infection determination model, thereby ensuring the accuracy of the determined target wound infection determination model and its correspondence with the target wound. Current wound data is input into the target wound infection determination model, which performs feature extraction on the current wound data and, based on the extracted features, outputs the current wound bacterial load and the current wound infection type, thereby ensuring the accuracy of the output current wound bacterial load.

[0025] In an optional embodiment, the electronic device is further used to obtain multiple sets of historical wound surface data; the historical wound surface data includes one or more physiological parameters of historical wound surface temperature, historical infiltration moisture, historical wound surface pH value, historical wound surface lactic acid, historical wound surface uric acid, historical wound surface glucose, and historical wound surface inflammatory factors;

[0026] Based on the gold standard of historical wound bacterial load and historical wound infection type corresponding to each group of historical wound data, each group of historical wound data is labeled to generate the first wound data training set;

[0027] Performing data enhancement processing on each group of first wound surface training data in the first wound surface data training set to generate a second training data set;

[0028] performing high-dimensional feature learning on the second training data in the second training data set to generate multiple groups of high-dimensional features;

[0029] Divide the generated multiple sets of high-dimensional features into a high-dimensional feature training set and a high-dimensional feature verification set;

[0030] The high-dimensional feature training set is input into the initial wound infection determination network, and a preset number of nested cross-validations are performed. Each repeated nested cross-validation is performed on the high-dimensional feature training set. In the nested cross-validation, the outer-loop cross-validation is used to resample the high-dimensional feature training set, and the inner-loop cross-validation is used to perform hyperparameter tuning and feature selection on the high-dimensional feature training set of each outer-loop cross-validation. In the inner-loop cross-validation, the default parameters of the grid search function are used to try different hyperparameter combinations and select the optimal hyperparameters and stable features.

[0031] Create a candidate wound infection determination model based on the best hyperparameters and stable features from nested cross-validation training;

[0032] Using the high-dimensional feature validation set, the candidate wound infection determination model is validated based on the preset loss function to obtain the validation results;

[0033] According to the verification results, the optimal hyperparameters corresponding to the candidate wound infection determination model are adjusted to generate the target wound infection determination model.

[0034] The wound monitoring system provided in the embodiment of the present application is an electronic device that obtains multiple groups of historical wound data, and based on the historical wound bacterial load corresponding to each group of historical wound data and the gold standard of the historical wound infection type, labels each group of historical wound data to generate a first wound data training set, thereby ensuring the accuracy of the labeling of each group of historical wound data and the accuracy of the generated first wound data training set. Data enhancement processing is performed on each group of first wound training data in the first wound data training set to generate a second training data set, thereby ensuring the accuracy of the generated second training data set and ensuring that the data of the generated second training data set is comprehensive and has a large data volume. High-dimensional feature learning is performed on the second training data in the second training data set to generate multiple groups of high-dimensional features; the accuracy of the generated multiple groups of high-dimensional features is ensured, and the features are diversified. The generated multiple groups of high-dimensional features are divided into a high-dimensional feature training set and a high-dimensional feature verification set. The high-dimensional feature training set was input into the initial wound infection determination network, and nested cross-validation was performed with a preset number of iterations. Each iteration of nested cross-validation was performed on the high-dimensional feature training set. Within the nested cross-validation, the outer loop cross-validation was used to resample the high-dimensional feature training set, while the inner loop cross-validation was used to perform hyperparameter tuning and feature selection on each high-dimensional feature training set from the outer loop cross-validation. In the inner loop cross-validation, the default parameters of the grid search function were used to try different hyperparameter combinations and select the optimal hyperparameters and stable features, ensuring the accuracy of the selected optimal hyperparameters. Based on the optimal hyperparameters and stable features from the nested cross-validation training, a candidate wound infection determination model was created, ensuring the accuracy of the candidate wound infection determination model. The candidate wound infection determination model was then validated using the high-dimensional feature validation set based on a preset loss function to obtain validation results, ensuring the accuracy of the validation results. Based on the validation results, the optimal hyperparameters corresponding to the candidate wound infection determination model were adjusted to generate a target wound infection determination model, ensuring the accuracy of the generated target wound infection determination model.

[0035] In an optional embodiment, the electronic device is used to compare the current wound bacterial load with a first wound bacterial load threshold and a second wound bacterial load threshold, respectively; wherein the first wound bacterial load threshold is greater than the second wound bacterial load threshold;

[0036] When the bacterial load of the current wound surface is greater than the first wound surface bacterial load threshold, the monitoring result corresponding to the target wound surface is determined to be a first-level monitoring result;

[0037] When the current wound bacterial load is less than the first wound bacterial load threshold and greater than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a secondary monitoring result;

[0038] When the current wound bacterial load is less than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a third-level monitoring result; the wound infection degree corresponding to the first-level monitoring result is greater than the second-level monitoring result, and the wound infection degree corresponding to the second-level monitoring result is greater than the third-level monitoring result.

[0039] The wound monitoring system provided in the embodiment of the present application, the electronic device is used to compare the current wound bacterial load with the first wound bacterial load threshold and the second wound bacterial load threshold respectively; when the current wound bacterial load is greater than the first wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a first-level monitoring result, thereby ensuring the accuracy of the determined first-level monitoring result. When the current wound bacterial load is less than the first wound bacterial load threshold and greater than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a second-level monitoring result, thereby ensuring the accuracy of the determined second-level monitoring result. When the current wound bacterial load is less than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a third-level monitoring result, thereby ensuring the accuracy of the determined third-level monitoring result.

[0040] In an optional embodiment, the wound monitoring system further includes a camera device, which is communicatively connected to the electronic device, wherein:

[0041] A camera device is used to capture images of the target wound surface, generate a current wound surface image, and transmit the current wound surface image to an electronic device;

[0042] An electronic device is used to process the current wound surface image, determine the current wound surface area and current wound surface depth corresponding to the target wound surface, and store the current wound surface area and current wound surface depth in correspondence with the current wound surface bacterial load and current wound surface infection type corresponding to the target wound surface, so as to facilitate tracing the recovery status of the target wound surface;

[0043] The electronic device is also used to send the current wound image, current wound area and current wound depth corresponding to the target wound to the server, so that the server can store the current wound image, current wound area, current wound depth and the target wound in correspondence.

[0044] The wound monitoring system provided in an embodiment of the present application further includes a camera device, which is communicatively connected to the electronic device. The camera device is configured to capture an image of the target wound, generate a current wound image, and transmit the current wound image to the electronic device. This allows the electronic device to receive and process the current wound image. The electronic device is configured to process the current wound image to determine the current wound area and current wound depth corresponding to the target wound, ensuring the accuracy of the current wound area and current wound depth determined for the target wound. The current wound area and current wound depth are stored in correspondence with the current wound bacterial load and current wound infection type corresponding to the target wound, facilitating the tracing of the recovery of the target wound. The electronic device is further configured to transmit the current wound image, current wound area, and current wound depth corresponding to the target wound to a server, so that the server stores the current wound image, current wound area, and current wound depth in correspondence with the target wound, thereby enabling the server to store all wound data corresponding to the target wound.

[0045] In an optional embodiment, the electronic device is used to input the current wound image into a preset wound image recognition model, which performs feature extraction on the current wound image and outputs the current wound area and current wound depth corresponding to the target wound based on the extracted features.

[0046] The wound monitoring system and electronic device provided in the embodiments of the present application are also used to input the current wound image into a preset wound image recognition model. The preset wound image recognition model extracts features from the current wound image and outputs the current wound area and current wound depth corresponding to the target wound based on the extracted features, thereby ensuring the accuracy of the output current wound area and current wound depth corresponding to the target wound.

[0047] In an optional embodiment, the electronic device is further configured to receive historical wound area, historical wound depth, and historical wound bacterial load corresponding to the target wound from the server, compare the historical wound area with the current wound area, compare the historical wound depth with the current wound depth, and compare the historical wound bacterial load with the current wound bacterial load;

[0048] If the wound area difference between the historical wound area and the current wound area is greater than the preset area difference, and the wound depth difference between the historical wound depth and the current wound depth is greater than the preset depth difference, and the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is greater than the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be good recovery;

[0049] If the wound area difference between the historical wound area and the current wound area is less than or equal to the preset area difference, or the wound depth difference between the historical wound depth and the current wound depth is less than or equal to the preset depth difference, or the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is less than or equal to the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be poor recovery, and a prompt message for strengthening wound care is output.

[0050] In the wound monitoring system provided in the embodiment of the present application, the electronic device is further used to receive the historical wound area, historical wound depth, and historical wound bacterial load corresponding to the target wound sent by the server, compare the historical wound area with the current wound area, compare the historical wound depth with the current wound depth, and compare the historical wound bacterial load with the current wound bacterial load. If the wound area difference between the historical wound area and the current wound area is greater than the preset area difference, and the wound depth difference between the historical wound depth and the current wound depth is greater than the preset depth difference, and the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is greater than the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be well recovered, thereby ensuring the accuracy of the determined recovery result corresponding to the target wound being well recovered. If the wound area difference between the historical and current wound areas is less than or equal to a preset area difference, or the wound depth difference between the historical and current wound depths is less than or equal to a preset depth difference, or the wound bacterial load difference between the historical and current wound bacterial loads is less than or equal to a preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be poor recovery, and a prompt message is output to strengthen wound care. This ensures the accuracy of the determination of the poor recovery result for the target wound, and by outputting the prompt message, the user can be prompted to strengthen wound care to facilitate recovery of the target wound.

[0051] In an optional embodiment, the wound surface data collection device and the electronic device are detachable, and the electronic device is communicatively connected to at least one wound surface data collection device, wherein:

[0052] An electronic device is used to send current wound data, current wound bacterial load, current wound infection type, current wound image, current wound area, current wound depth and recovery results of the target wound to the terminal device corresponding to the target wound patient, and when the monitoring result is a first-level monitoring result, send the corresponding operation plan to the terminal device.

[0053] The wound monitoring system and electronic device provided in the embodiment of the present application are used to send current wound data, current wound bacterial load, current wound infection type, current wound image, current wound area, current wound depth and recovery result of the target wound to the terminal device corresponding to the target wound patient, and when the monitoring result is a first-level monitoring result, send the corresponding operation plan to the terminal device, so that the target patient can complete debridement according to the operation plan, without going to the hospital. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. 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 paying any creative work.

[0055] Figure 1 is a structural diagram of a wound monitoring system according to an embodiment of the present invention;

[0056] Figure 2 is a structural diagram of another wound monitoring system according to an embodiment of the present invention;

[0057] Figure 3 2 is a schematic structural diagram of another wound monitoring system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0058] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying 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 those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0059] In this embodiment, a wound monitoring system is provided. Figure 1 As shown, the wound monitoring system includes: a wound data acquisition device 1, an electronic device 2 and a server 3, wherein the wound data acquisition device 1 is communicatively connected to the electronic device 2, and the electronic device 2 is communicatively connected to the server 3, wherein:

[0060] Wound surface data acquisition device 1, used to collect current wound surface data of the target wound surface and transmit the current wound surface data to electronic device 2;

[0061] Electronic device 2 is configured to receive current wound surface data, process the current wound surface data, determine the current wound surface bacterial load and the current wound surface infection type corresponding to the target wound surface; determine the monitoring result corresponding to the target wound surface based on the current wound surface bacterial load, and send the current wound surface data, current wound surface bacterial load, and current wound surface infection type corresponding to the target wound surface to server 3;

[0062] Server 3 is used to store the current wound data, the current wound bacterial load, the current wound infection type, the target wound, and the target wound patient corresponding to the target wound, so as to facilitate the tracing of all wound data corresponding to the target wound, and to display the current wound data, the current wound bacterial load, and the current wound infection type to the target user;

[0063] Server 3 is further configured to perform statistical analysis on the current wound infection type corresponding to each target wound patient, determine the wound infection characteristics corresponding to each target wound patient group, and generate corresponding wound care methods for the group based on the wound infection characteristics.

[0064] Specifically, the wound data acquisition device 1 can be a wound dressing device, which includes a variety of physiological parameter acquisition devices for target wounds, so that the wound dressing device can be applied to the target wound to collect the current wound data corresponding to the target wound.

[0065] Then, the wound surface data acquisition device 1 transmits the collected current wound surface data of the target wound surface to the electronic device 2 based on the communication connection between the wound surface data acquisition device 1 and the electronic device 2 .

[0066] It should be noted that the electronic device 2 may be an independent terminal device, which may include a processor and a memory.

[0067] After receiving the current wound surface data corresponding to the target wound surface, the electronic device 2 may identify the current wound surface data and determine the current wound surface bacterial load and the current wound surface infection type corresponding to the target wound surface based on the identification result.

[0068] The current wound infection type may include, but is not limited to, at least one of bacterial infections such as Staphylococcus aureus infection, Pseudomonas aeruginosa infection, and Klebsiella pneumoniae infection. The present embodiment does not specifically limit the current wound infection type.

[0069] Electronic device 2 can compare the current wound bacterial load with a preset bacterial load threshold and, based on the comparison result, determine the monitoring result corresponding to the target wound. Electronic device 2 can then send the current wound data, current wound bacterial load, and current wound infection type corresponding to the target wound to server 3, thereby enabling corresponding storage of the current wound data, current wound bacterial load, and current wound infection type corresponding to the target wound.

[0070] After receiving the current wound data, current wound bacterial load, and current wound infection type corresponding to the target wound, server 3 can store the current wound data, current wound bacterial load, and current wound infection type in correspondence with the target wound and the target wound patient corresponding to the target wound, so as to facilitate the tracing of all wound data corresponding to the target wound. Then, server 3 can also display the current wound data, current wound bacterial load, and current wound infection type corresponding to the target wound to the target user. Among them, the target person can be a medical staff, so that the medical staff can provide remote medical guidance through the system, issue treatment and nursing plans to the nursing staff, explore the wound management model of telemedicine, avoid empirical dressing changes, and achieve the prediction and early diagnosis of wound infection.

[0071] Based on the above content, it can be seen that the server can store the current wound infection type corresponding to each target wound. Then, the server 3 can perform statistical analysis on the current wound infection type corresponding to each target wound patient to determine the wound infection characteristics corresponding to each target wound patient group. For example, the server 3 can perform classification statistics based on the age or gender of each target wound patient, count the number of various current wound infection types at each age stage, and then determine the wound infection characteristics corresponding to each target wound patient group at each age stage. For example, the wounds of young patients are mainly infected with Gram-positive bacteria, while the wounds of elderly patients are mainly infected with Gram-negative bacteria.

[0072] Then, the server 3 can generate corresponding wound care methods based on the wound infection characteristics of each target wound patient group. For example, for Gram-positive resistant bacteria infection, vancomycin can be recommended for debridement; for Gram-negative resistant bacteria infection, aminoglycosides can be recommended for debridement.

[0073] The wound monitoring system provided in the embodiment of the present application includes a wound data acquisition device 1, which is used to collect the current wound data of the target wound and transmit the current wound data to an electronic device 2, so that the electronic device 2 can obtain the current wound data of the target wound. The electronic device 2 is used to receive the current wound data and process the current wound data to determine the current wound bacterial load and the current wound infection type corresponding to the target wound, thereby ensuring the accuracy of the determined current wound bacterial load and the current wound infection type. The monitoring result corresponding to the target wound can then be determined based on the current wound bacterial load, thereby ensuring the accuracy of the determined monitoring result. The current wound data, current wound bacterial load, and current wound infection type corresponding to the target wound are sent to a server 3. Thus, the server 3 can store the current wound data, current wound bacterial load, and current wound infection type corresponding to the target wound.

[0074] Server 3 is used to store the current wound data, the current wound bacterial load, the current wound infection type, the target wound, and the target wound patient corresponding to the target wound, thereby realizing the storage of the wound data of the target wound, and facilitating the tracing of all wound data corresponding to the target wound. The current wound data, the current wound bacterial load, and the current wound infection type are displayed to the target user, so that the target user can receive the current wound bacterial load and the current wound infection type. Server 3 is also used to perform statistical analysis on the current wound infection type corresponding to each target wound patient, determine the wound infection characteristics corresponding to each target wound patient group, and ensure the accuracy of the wound infection characteristics corresponding to each target wound patient group. According to the wound infection characteristics, the corresponding population wound care method is generated, and the accuracy of the generated population wound care method is ensured, and then specific populations can be cared for according to the population wound care method.

[0075] In an optional embodiment of the present application, Figure 2 As shown, the wound surface data acquisition device 1 includes: at least one of a temperature acquisition component 11, an infiltration humidity acquisition component 12, a pH value acquisition component 13, a lactic acid acquisition component 14, a uric acid acquisition component 15, a glucose acquisition component 16, and a cytokine and inflammation-related protein component 17;

[0076] The temperature collection component 11 is used to collect the current wound temperature of the target wound surface;

[0077] The infiltration humidity collection component 12 is used to collect the current infiltration humidity of the target wound surface;

[0078] The pH value collection component 13 is used to collect the current pH value of the target wound surface;

[0079] Lactic acid collection component 14, used to collect the current wound lactic acid of the target wound;

[0080] The uric acid collection component 15 is used to collect the current uric acid of the target wound surface;

[0081] A glucose collection component 16 is used to collect glucose from the current wound surface of the target wound surface;

[0082] The cytokine and inflammation-related protein component 17 is used to collect the current wound inflammatory factors of the target wound.

[0083] Among them, the cytokine and inflammation-related protein component 17 includes at least one of a matrix metalloproteinase 9 sensor, a matrix metalloproteinase inhibitor 1 sensor, an interleukin 1β sensor, a tumor necrosis factor α sensor, and an interleukin 6 sensor.

[0084] Specifically, the current wound temperature can reflect information such as local blood flow, lymphocyte infiltration and infection, which is mainly determined by the periwound blood flow and environmental temperature, and the local blood flow is affected by vasoactive substances and environmental temperature.

[0085] The current wound humidity is highly correlated with the target wound healing. The current wound humidity refers to the infiltration humidity rather than the air humidity. A moist environment is not only conducive to maintaining cell vitality, promoting regeneration, enhancing immune cell function and promoting epidermal cell migration, but also reduces scars. However, increased exudation also indicates an increase in local inflammation and a high risk of infection. The current wound pH is highly correlated with biological processes such as infection, angiogenesis and enzyme activity. An acidic environment can promote fibroblast proliferation, angiogenesis and epithelialization, and inhibit bacterial reproduction. The pH value of chronic wounds is often higher than that of normal wounds, and infection can cause it to rise further.

[0086] Indicators such as current wound lactic acid, current wound uric acid, and current wound glucose concentration also reflect the metabolism and bacterial colonization of the target wound. However, the presence of bacteria in the wound does not necessarily indicate infection. Generally speaking, infection occurs only when the bacterial load in the wound exceeds 105 cFU / mL. However, this value varies due to individual differences in immunity and bacterial virulence.

[0087] Current wound inflammatory factors can reflect the inflammation level and healing status of the target wound.

[0088] Optionally, the current wound data includes one or more physiological parameters including current wound temperature, current infiltration humidity, current wound pH value, current wound lactic acid, current wound uric acid, current wound glucose, and current wound inflammatory factors. The electronic device 2 is used to obtain the target site corresponding to the target wound; obtain the target wound infection determination model corresponding to the target site based on the correspondence between the site and the wound infection determination model; input the current wound data into the target wound infection determination model, and the target wound infection determination model extracts features from the current wound data, and outputs the current wound bacterial load and the current wound infection type based on the extracted features.

[0089] In an optional embodiment of the present application, the electronic device 2 may receive a target site corresponding to the target wound surface input by the user, and then determine the target wound surface infection determination model corresponding to the Mu Bai wound surface based on the correspondence between the site and the wound surface infection determination model. The electronic device 2 may then search the storage space for the target wound surface infection determination model corresponding to the target wound surface, or receive a target wound surface infection determination model sent by another device or a target wound surface infection determination model sent by another device.

[0090] In an optional embodiment of the present application, the electronic device 2 is further used to obtain multiple sets of historical wound surface data; the historical wound surface data includes one or more physiological parameters of historical wound surface temperature, historical infiltration humidity, historical wound surface pH value, historical wound surface lactic acid, historical wound surface uric acid, historical wound surface glucose, and historical wound surface inflammatory factors;

[0091] Based on the gold standard of historical wound bacterial load and historical wound infection type corresponding to each group of historical wound data, each group of historical wound data is labeled to generate the first wound data training set;

[0092] Performing data enhancement processing on each group of first wound surface training data in the first wound surface data training set to generate a second training data set;

[0093] performing high-dimensional feature learning on the second training data in the second training data set to generate multiple groups of high-dimensional features;

[0094] Divide the generated multiple sets of high-dimensional features into a high-dimensional feature training set and a high-dimensional feature verification set;

[0095] The high-dimensional feature training set is input into the initial wound infection determination network, and a preset number of nested cross-validations are performed. Each repeated nested cross-validation is performed on the high-dimensional feature training set. In the nested cross-validation, the outer-loop cross-validation is used to resample the high-dimensional feature training set, and the inner-loop cross-validation is used to perform hyperparameter tuning and feature selection on the high-dimensional feature training set of each outer-loop cross-validation. In the inner-loop cross-validation, the default parameters of the grid search function are used to try different hyperparameter combinations and select the optimal hyperparameters and stable features.

[0096] Create a candidate wound infection determination model based on the best hyperparameters and stable features from nested cross-validation training;

[0097] Using the high-dimensional feature validation set, the candidate wound infection determination model is validated based on the preset loss function to obtain the validation results;

[0098] According to the verification results, the optimal hyperparameters corresponding to the candidate wound infection determination model are adjusted to generate the target wound infection determination model.

[0099] Specifically, the electronic device 2 can obtain multiple sets of historical wound surface data of the target part corresponding to the target wound surface, wherein the historical wound surface data includes one or more physiological parameters of historical wound surface temperature, historical infiltration humidity, historical wound surface pH value, historical wound surface lactic acid, historical wound surface uric acid, historical wound surface glucose, and historical wound surface inflammatory factors.

[0100] Then, the electronic device 2 may label each set of historical wound data based on the historical wound bacterial load corresponding to each set of historical wound data and the gold standard of the historical wound infection type to generate a first wound data training set.

[0101] Optionally, the electronic device 2 may also clean the collected first wound surface data training set to remove outliers and noise, thereby ensuring the quality and accuracy of the first wound surface data training set.

[0102] The electronic device 2 may also perform data enhancement processing on each group of first wound surface training data in the first wound surface data training set by using a preset data enhancement processing method such as interpolation and noise addition to generate a second training data set.

[0103] Electronic device 2 can calculate statistical features, such as mean, variance, maximum, and minimum values, based on the numerical features of the second training data in the second training dataset. These statistical features can reflect the distribution and central tendency of the second training data in the second training dataset. For time series second training data, frequency domain features can be extracted using frequency domain analysis methods, such as Fourier transform and wavelet transform. Frequency domain features can reflect the periodicity and frequency distribution of the data. Furthermore, the second training data is typically time-series, and dynamic changes in the second training data can be captured by extracting time series features. Common time series features include autocorrelation coefficients, differential coefficients, and sliding window statistical features. Electronic device 2 can also use signal processing methods to pre-process the second training data, such as filtering, noise reduction, and smoothing, to improve feature quality and representativeness. During feature extraction, feature selection methods can be considered to select the most representative feature subset. Common feature selection methods include variance selection, correlation coefficient selection, and recursive feature elimination, thereby generating multiple sets of high-dimensional features.

[0104] The electronic device 2 may divide the generated multiple sets of high-dimensional features into a high-dimensional feature training set and a high-dimensional feature training set. For example, 70% is the high-dimensional feature training set and 30% is the high-dimensional feature training set.

[0105] Exemplarily, the electronic device 2 can input the high-dimensional feature training set into the initial wound infection determination network and perform 5-fold nested cross-validation, where each repeated nested cross-validation is performed on the high-dimensional feature training set; wherein, in the nested cross-validation, the outer loop cross-validation is used to resample the high-dimensional feature training set, and the inner loop cross-validation is used to perform hyperparameter tuning and feature selection on the high-dimensional feature training set of each outer loop cross-validation, wherein, in the inner loop cross-validation, the default parameters of the grid search function are used to try different hyperparameter combinations and select the best hyperparameters and stable features. This ensures that the selected stable features can maintain good performance under different models and hyperparameter settings.

[0106] The grid search function is a commonly used hyperparameter tuning method. It iterates over all combinations of given hyperparameter candidate values, performs model training and evaluation, and finds the optimal hyperparameter combination. By applying the grid search function for hyperparameter tuning, you can find the hyperparameters that best suit the dataset and model, thereby improving model performance.

[0107] Then, electronic device 2 uses the high-dimensional feature validation set to validate the candidate wound infection determination model based on a preset loss function to obtain a validation result. Based on the validation result, the optimal hyperparameters corresponding to the candidate wound infection determination model are adjusted to generate a target wound infection determination model.

[0108] Among them, the evaluation method uses the receiver operating characteristic (ROC) curve, and the evaluation criteria are area under the curve (AUC), accuracy, precision, and recall, etc. After multiple rounds of optimization, the prediction model with the best performance is selected.

[0109] Among them, the target wound infection determination model can be any one of SVC (Support Vector Classification), random forest, XG-Boost and recurrent neural network based on integration thinking, radial basis function (RBF) network, feedforward neural network (FFNN), convolutional neural network (CNN), deconvolutional neural network (DN), deep convolutional inverse graphics network (DCIGN), generative adversarial network (GAN), recurrent neural network (RNN), long / short term memory network (LSTM), deep residual network (DRN) and extreme learning machine (ELM). The embodiment of the present application does not specifically limit the target wound infection determination model.

[0110] The wound monitoring system provided in the embodiment of the present application is configured such that the electronic device 2 is used to obtain a target site corresponding to the target wound; based on the correspondence between the site and the wound infection determination model, the target wound infection determination model corresponding to the target site is obtained, thereby ensuring the accuracy of the determined target wound infection determination model and its correspondence with the target wound. The electronic device 2 obtains multiple sets of historical wound data, and based on the gold standard of the historical wound bacterial load and the historical wound infection type corresponding to each set of historical wound data, labels each set of historical wound data to generate a first wound data training set, thereby ensuring the accuracy of the labeling of each set of historical wound data and the accuracy of the generated first wound data training set. The electronic device 2 performs data enhancement processing on each set of first wound training data in the first wound data training set to generate a second training data set, thereby ensuring the accuracy of the generated second training data set, and ensuring the comprehensiveness of the data in the generated second training data set and a large data volume. The electronic device 2 performs high-dimensional feature learning on the second training data in the second training data set to generate multiple sets of high-dimensional features, thereby ensuring the accuracy of the generated multiple sets of high-dimensional features and the diversification of the features. The generated multiple sets of high-dimensional features were divided into a high-dimensional feature training set and a high-dimensional feature validation set. The high-dimensional feature training set was input into the initial wound infection determination network and subjected to a preset multiple nested cross-validation. Each repeated nested cross-validation was performed on the high-dimensional feature training set. In the nested cross-validation, the outer-loop cross-validation was used to resample the high-dimensional feature training set, and the inner-loop cross-validation was used to perform hyperparameter tuning and feature selection on each outer-loop cross-validation high-dimensional feature training set. In the inner-loop cross-validation, the default parameters of the grid search function were used to try different hyperparameter combinations and select the optimal hyperparameters and stable features, ensuring the accuracy of the selected optimal hyperparameters. Based on the optimal hyperparameters and stable features from the nested cross-validation training, a candidate wound infection determination model was created, ensuring the accuracy of the candidate wound infection determination model. The candidate wound infection determination model was validated using the high-dimensional feature validation set based on a preset loss function to obtain a validation result, ensuring the accuracy of the validation result. Based on the validation results, the optimal hyperparameters corresponding to the candidate wound infection determination model were adjusted to generate a target wound infection determination model, ensuring its accuracy. Current wound data was input into the target wound infection determination model, which extracted features from the data and, based on the extracted features, output the current wound bacterial load and infection type, ensuring the accuracy of the output current wound bacterial load.

[0111] In the wound monitoring system provided in the embodiment of the present application, the electronic device 2 is used to compare the current wound bacterial load with the first wound bacterial load threshold and the second wound bacterial load threshold respectively; when the current wound bacterial load is greater than the first wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a first-level monitoring result, thereby ensuring the accuracy of the determined first-level monitoring result. When the current wound bacterial load is less than the first wound bacterial load threshold and greater than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a second-level monitoring result, thereby ensuring the accuracy of the determined second-level monitoring result. When the current wound bacterial load is less than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a third-level monitoring result, thereby ensuring the accuracy of the determined third-level monitoring result.

[0112] In an optional embodiment of the present application, the electronic device 2 is used to compare the current wound bacterial load with a first wound bacterial load threshold and a second wound bacterial load threshold, respectively; wherein the first wound bacterial load threshold is greater than the second wound bacterial load threshold;

[0113] When the bacterial load of the current wound surface is greater than the first wound surface bacterial load threshold, the monitoring result corresponding to the target wound surface is determined to be a first-level monitoring result;

[0114] When the current wound bacterial load is less than the first wound bacterial load threshold and greater than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a secondary monitoring result;

[0115] When the bacterial load of the current wound is less than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a third-level monitoring result.

[0116] Among them, the degree of wound infection corresponding to the first-level monitoring result is greater than that of the second-level monitoring result, and the degree of wound infection corresponding to the second-level monitoring result is greater than that of the third-level monitoring result.

[0117] Specifically, the electronic device 2 can receive the first wound bacterial load threshold and the second wound bacterial load threshold input by the user, or can receive the first wound bacterial load threshold and the second wound bacterial load threshold sent by other devices. The electronic device 2 can also set the first wound bacterial load threshold and the second wound bacterial load threshold based on the target site corresponding to the target wound or other actual conditions. The embodiment of the present application does not specifically limit the manner in which the electronic device 2 obtains the first wound bacterial load threshold and the second wound bacterial load threshold.

[0118] Then, electronic device 2 compares the current wound bacterial load with the first wound bacterial load threshold and the second wound bacterial load threshold, respectively. When the current wound bacterial load is greater than the first wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a first-level monitoring result; when the current wound bacterial load is less than the first wound bacterial load threshold and greater than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a second-level monitoring result; when the current wound bacterial load is less than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a third-level monitoring result.

[0119] In an optional embodiment, as Figure 3 As shown, the wound monitoring system further includes a camera device 4, which is in communication with the electronic device 2, wherein:

[0120] The camera device 4 is used to capture images of the target wound surface, generate a current wound surface image, and transmit the current wound surface image to the electronic device 2;

[0121] Electronic device 2 is used to process the current wound image, determine the current wound area and current wound depth corresponding to the target wound, and store the current wound area and current wound depth in correspondence with the current wound bacterial load and current wound infection type corresponding to the target wound, so as to facilitate tracing the recovery of the target wound;

[0122] The electronic device 2 is also used to send the current wound image, current wound area and current wound depth corresponding to the target wound to the server 3, so that the server 3 stores the current wound image, current wound area, current wound depth and the target wound in correspondence.

[0123] Optionally, the electronic device 2 is used to input the current wound image into a preset wound image recognition model, which performs feature extraction on the current wound image and outputs the current wound area and current wound depth corresponding to the target wound based on the extracted features.

[0124] Specifically, the camera device 4 can capture images of the target wound surface, generate a current wound surface image corresponding to the target wound surface, and transmit the current wound surface image to the electronic device 2.

[0125] The electronic device 2 can input the current wound image into a preset wound image recognition model, which extracts features from the current wound image and outputs the current wound area and current wound depth corresponding to the target wound based on the extracted features.

[0126] It should be noted that the electronic device 2 can mark the historical wound depth and historical wound area corresponding to the historical wound image, generate a historical wound image training set, train the initial wound image recognition network based on the historical wound image training set, and generate a preset wound image recognition model.

[0127] Among them, the training process of the preset wound image recognition model can refer to the training process of the above-mentioned target wound infection determination model. The embodiment of the present application does not specifically limit the training process of the preset wound image recognition model.

[0128] Optionally, the electronic device 2 may also receive a preset wound surface image recognition model input by the user, or a preset wound surface image recognition model sent by other devices. The manner in which the electronic device 2 obtains the preset wound surface image recognition model in the embodiment of the present application is not specifically limited.

[0129] Then, the electronic device 2 may store the current wound area and the current wound depth in correspondence with the current wound bacterial load and the current wound infection type corresponding to the target wound, so as to trace the recovery of the target wound.

[0130] The electronic device 2 can also send the current wound image, current wound area and current wound depth corresponding to the target wound to the server 3, so that the server 3 can store the current wound image, current wound area, current wound depth and the target wound in correspondence.

[0131] The wound monitoring system provided in the embodiment of the present application also includes a camera device 4, which is communicatively connected to the electronic device 2. The camera device 4 is used to capture images of the target wound, generate a current wound image, and transmit the current wound image to the electronic device 2. This allows the electronic device 2 to receive the current wound image and process the current wound image. The electronic device 2 is used to input the current wound image into a preset wound image recognition model. The preset wound image recognition model extracts features from the current wound image and outputs the current wound area and current wound depth corresponding to the target wound based on the extracted features, thereby ensuring the accuracy of the output current wound area and current wound depth corresponding to the target wound. The current wound area and current wound depth are stored in correspondence with the current wound bacterial load and current wound infection type corresponding to the target wound, so as to facilitate tracing the recovery of the target wound. The electronic device 2 is also used to send the current wound image, current wound area and current wound depth corresponding to the target wound to the server 3, so that the server 3 can store the current wound image, current wound area, current wound depth and the target wound in correspondence, so that the server 3 can store all wound data corresponding to the target wound in correspondence.

[0132] In an optional embodiment of the present application, the electronic device 2 is further configured to receive the historical wound area, historical wound depth, and historical wound bacterial load corresponding to the target wound sent by the server 3, compare the historical wound area with the current wound area, compare the historical wound depth with the current wound depth, and compare the historical wound bacterial load with the current wound bacterial load;

[0133] If the wound area difference between the historical wound area and the current wound area is greater than the preset area difference, and the wound depth difference between the historical wound depth and the current wound depth is greater than the preset depth difference, and the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is greater than the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be good recovery;

[0134] If the wound area difference between the historical wound area and the current wound area is less than or equal to the preset area difference, or the wound depth difference between the historical wound depth and the current wound depth is less than or equal to the preset depth difference, or the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is less than or equal to the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be poor recovery, and a prompt message for strengthening wound care is output.

[0135] Specifically, after obtaining the current wound area, current wound depth, and current wound load corresponding to the target wound, electronic device 2 may send a data request to server 3 to obtain the historical wound area, historical wound depth, and historical wound bacterial load corresponding to the target wound. After receiving the data request sent by electronic device 2, server 3 sends the historical wound area, historical wound depth, and historical wound bacterial load corresponding to the target wound to electronic device 2.

[0136] Then, the electronic device 2 may compare the historical wound surface area with the current wound surface area, compare the historical wound surface depth with the current wound surface depth, and compare the historical wound surface bacterial load with the current wound surface bacterial load.

[0137] If the wound area difference between the historical wound area and the current wound area is greater than the preset area difference, the wound depth difference between the historical wound depth and the current wound depth is greater than the preset depth difference, and the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is greater than the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be good recovery.

[0138] If the wound area difference between the historical wound area and the current wound area is less than or equal to the preset area difference, or the wound depth difference between the historical wound depth and the current wound depth is less than or equal to the preset depth difference, or the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is less than or equal to the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be poor recovery, and a prompt message for strengthening wound care is output.

[0139] In the wound monitoring system provided in the embodiment of the present application, the electronic device 2 is further used to receive the historical wound area, historical wound depth, and historical wound bacterial load corresponding to the target wound sent by the server 3, and compare the historical wound area with the current wound area, the historical wound depth with the current wound depth, and the historical wound bacterial load with the current wound bacterial load. If the wound area difference between the historical wound area and the current wound area is greater than the preset area difference, and the wound depth difference between the historical wound depth and the current wound depth is greater than the preset depth difference, and the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is greater than the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be well recovered, thereby ensuring the accuracy of the determined recovery result corresponding to the target wound being well recovered. If the wound area difference between the historical and current wound areas is less than or equal to a preset area difference, or the wound depth difference between the historical and current wound depths is less than or equal to a preset depth difference, or the wound bacterial load difference between the historical and current wound bacterial loads is less than or equal to a preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be poor recovery, and a prompt message is output to strengthen wound care. This ensures the accuracy of the determination of the poor recovery result for the target wound, and by outputting the prompt message, the user can be prompted to strengthen wound care to facilitate recovery of the target wound.

[0140] In an optional embodiment of the present application, the wound surface data acquisition device 1 and the electronic device 2 are detachable, and the electronic device 2 is communicatively connected to at least one wound surface data acquisition device 1, wherein:

[0141] Electronic device 2 is used to send the current wound data, the current wound bacterial load, the current wound infection type, the current wound image, the current wound area, the current wound depth and the recovery result of the target wound to the terminal device corresponding to the target wound patient, and when the monitoring result is a first-level monitoring result, send the corresponding operation plan to the terminal device.

[0142] Specifically, the wound surface data collection device 1 and the electronic device 2 are detachable, and the electronic device 2 is communicatively connected to at least one wound surface data collection device 1. For example, the wound surface data collection device 1 can be applied to the target wound surface of the target wound patient, while the electronic device 2 can be at the medical staff end, and the electronic device 2 can receive the current wound surface data collected by multiple wound surface data collection devices 1.

[0143] After processing the current wound data and obtaining the current wound bacterial load, current wound infection type, current wound image, current wound area, current wound depth, and target wound recovery result corresponding to the target wound, the electronic device 2 can send the current wound data, current wound bacterial load, current wound infection type, current wound image, current wound area, current wound depth, and target wound recovery result to the terminal device corresponding to the target wound patient. This allows the target wound patient to receive all the target wound data. The electronic device 2 can also send the corresponding operation plan to the terminal device when the monitoring result is a first-level monitoring result. This reduces the need for the target wound patient to travel.

[0144] The wound monitoring system provided in the embodiment of the present application, the electronic device 2, is used to send the current wound data, the current wound bacterial load, the current wound infection type, the current wound image, the current wound area, the current wound depth and the recovery result of the target wound to the terminal device corresponding to the target wound patient, and when the monitoring result is a first-level monitoring result, send the corresponding operation plan to the terminal device, so that the target patient can complete the debridement according to the operation plan, without going to the hospital.

[0145] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A wound monitoring system, characterized in that: The wound monitoring system includes: wound data acquisition equipment, electronic equipment and servers, including: A wound surface data acquisition device, configured to acquire current wound surface data of a target wound surface and transmit the current wound surface data to an electronic device; the wound surface data acquisition device comprises at least one of a temperature acquisition component, an infiltration humidity acquisition component, a pH value acquisition component, a lactate acquisition component, a uric acid acquisition component, a glucose acquisition component, and a cytokine and inflammation-related protein acquisition component; An electronic device is configured to receive current wound surface data and obtain a target site corresponding to a target wound surface; obtain a target wound surface infection determination model corresponding to the target site based on a correspondence between the site and the wound surface infection determination model; input the current wound surface data into the target wound surface infection determination model, which performs feature extraction on the current wound surface data and outputs a current wound surface bacterial load and a current wound surface infection type based on the extracted features; determine a monitoring result corresponding to the target wound surface based on the current wound surface bacterial load, and transmit the current wound surface data, the current wound surface bacterial load, and the current wound surface infection type corresponding to the target wound surface to a server; The server is used to store the current wound data, the current wound bacterial load, the current wound infection type, the target wound, and the target wound patient corresponding to the target wound, so as to facilitate the tracing of all wound data corresponding to the target wound, and to display the current wound data, the current wound bacterial load, and the current wound infection type to the target user; The server is also used to perform statistical analysis on the current wound infection type corresponding to each target wound patient, determine the wound infection characteristics corresponding to each target wound patient group, and generate corresponding wound care methods for the group based on the wound infection characteristics; The electronic device is further used to obtain multiple sets of historical wound surface data; the historical wound surface data includes one or more physiological parameters of historical wound surface temperature, historical infiltration moisture, historical wound surface pH value, historical wound surface lactic acid, historical wound surface uric acid, historical wound surface glucose, and historical wound surface inflammatory factors; Based on the gold standard of historical wound bacterial load and historical wound infection type corresponding to each group of historical wound data, each group of historical wound data is labeled to generate the first wound data training set; Performing data enhancement processing on each group of first wound surface training data in the first wound surface data training set to generate a second training data set; Calculating statistical features based on numerical features of second training data in the second training data set; For the second training data of the time series, frequency domain features are extracted using the frequency domain analysis method; Extracting time series features of second training data having time series characteristics; Use feature selection methods to select the most representative feature subsets and generate multiple sets of high-dimensional features; Divide the generated multiple sets of high-dimensional features into a high-dimensional feature training set and a high-dimensional feature verification set; The high-dimensional feature training set is input into the initial wound infection determination network, and a preset number of nested cross-validations are performed. Each repeated nested cross-validation is performed on the high-dimensional feature training set. In the nested cross-validation, the outer-loop cross-validation is used to resample the high-dimensional feature training set, and the inner-loop cross-validation is used to perform hyperparameter tuning and feature selection on the high-dimensional feature training set of each outer-loop cross-validation. In the inner-loop cross-validation, the default parameters of the grid search function are used to try different hyperparameter combinations and select the optimal hyperparameters and stable features. Create a candidate wound infection determination model based on the best hyperparameters and stable features from nested cross-validation training; Using the high-dimensional feature validation set, the candidate wound infection determination model is validated based on the preset loss function to obtain the validation results; According to the verification results, the optimal hyperparameters corresponding to the candidate wound infection determination model are adjusted to generate the target wound infection determination model.

2. The wound monitoring system according to claim 1, characterized in that: A temperature collection component is used to collect the current wound temperature of the target wound surface; an infiltration humidity collection component is used to collect the current infiltration humidity of the target wound surface; a pH value collection component is used to collect the current wound pH value of the target wound surface; A lactic acid collection component is used to collect the current wound lactic acid of the target wound; A uric acid collection component is used to collect the current uric acid of the target wound surface; A glucose collection component is used to collect the current wound surface glucose of the target wound surface; The cytokine and inflammation-related protein collection component is used to collect the current wound inflammatory factors of the target wound.

3. The wound monitoring system according to claim 1, characterized in that: The cytokine and inflammation-related protein collection component includes at least one of a matrix metalloproteinase 9 sensor, a matrix metalloproteinase inhibitor 1 sensor, an interleukin 1β sensor, a tumor necrosis factor α sensor, and an interleukin 6 sensor.

4. The wound monitoring system according to claim 1, characterized in that: The electronic device is used to compare the current wound bacterial load with a first wound bacterial load threshold and a second wound bacterial load threshold, respectively; wherein the first wound bacterial load threshold is greater than the second wound bacterial load threshold; When the bacterial load of the current wound surface is greater than the first wound surface bacterial load threshold, determining that the monitoring result corresponding to the target wound surface is a first-level monitoring result; When the current wound bacterial load is less than the first wound bacterial load threshold and greater than the second wound bacterial load threshold, determining that the monitoring result corresponding to the target wound is a secondary monitoring result; When the current wound bacterial load is less than the second wound bacterial load threshold, the monitoring result corresponding to the target wound is determined to be a third-level monitoring result; the wound infection degree corresponding to the first-level monitoring result is greater than the second-level monitoring result, and the wound infection degree corresponding to the second-level monitoring result is greater than the third-level monitoring result.

5. The wound monitoring system according to claim 4, characterized in that: The wound monitoring system further includes a camera device, which is communicatively connected to the electronic device, wherein: The camera device is used to capture images of the target wound surface, generate a current wound surface image, and transmit the current wound surface image to the electronic device; The electronic device is configured to process the current wound surface image, determine the current wound surface area and the current wound surface depth corresponding to the target wound surface, and store the current wound surface area and the current wound surface depth in correspondence with the current wound surface bacterial load and the current wound surface infection type corresponding to the target wound surface, so as to facilitate tracing the recovery of the target wound surface; The electronic device is further used to send the current wound image, the current wound area and the current wound depth corresponding to the target wound to the server, so that the server stores the current wound image, the current wound area, the current wound depth and the target wound in correspondence.

6. The wound monitoring system according to claim 5, characterized in that: The electronic device is used to input the current wound image into a preset wound image recognition model, and the preset wound image recognition model extracts features from the current wound image, and outputs the current wound area and the current wound depth corresponding to the target wound based on the extracted features.

7. The wound monitoring system according to claim 5, characterized in that: The electronic device is further configured to receive the historical wound area, historical wound depth, and historical wound bacterial load corresponding to the target wound sent by the server, compare the historical wound area with the current wound area, compare the historical wound depth with the current wound depth, and compare the historical wound bacterial load with the current wound bacterial load; If the wound area difference between the historical wound area and the current wound area is greater than the preset area difference, and the wound depth difference between the historical wound depth and the current wound depth is greater than the preset depth difference, and the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is greater than the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be good recovery; If the wound area difference between the historical wound area and the current wound area is less than or equal to the preset area difference, or the wound depth difference between the historical wound depth and the current wound depth is less than or equal to the preset depth difference, or the wound bacterial load difference between the historical wound bacterial load and the current wound bacterial load is less than or equal to the preset bacterial load difference, then the recovery result corresponding to the target wound is determined to be poor recovery, and a prompt message for strengthening wound care is output.

8. The wound monitoring system according to claim 7, characterized in that: The wound surface data collection device and the electronic device are detachable, and the electronic device is communicatively connected with at least one of the wound surface data collection devices, wherein: The electronic device is used to send the current wound data, the current wound bacterial load, the current wound infection type, the current wound image, the current wound area, the current wound depth and the recovery result of the target wound to the terminal device corresponding to the target wound patient, and when the monitoring result is a first-level monitoring result, send a corresponding operation plan to the terminal device.

Images