A negative pressure cryogenic plasma wound treatment system, device, and media

Abstract

The embodiment of the specification provides a negative pressure low-temperature plasma wound surface treatment system, device and medium, the system comprises: a space module, a plasma module, a negative pressure module and a control module, the space module comprises a closed space with adjustable space size, and the closed space is used for placing a wound surface; the plasma module is used for generating plasma gas, and the plasma module is connected with the space module through a flexible pipeline to transport the plasma gas into the closed space; the negative pressure module comprises a fan and an exhaust port, the exhaust port is arranged on at least one surface of the closed space, and the fan exhausts the gas in the closed space; and the control module is used for generating a control instruction and controlling the working parameters of the closed space, the plasma module and the fan through the control instruction.

Description

TECHNICAL FIELD

[0001] The present specification relates to the technical field of medical devices, in particular to a negative pressure low-temperature plasma wound treatment system, device and medium. BACKGROUND

[0002] In daily life, accidental incidents such as bumps, falls, and accidents often cause wounds, and if not treated in time, they often have the risk of infection. Traditional wound treatment methods mainly include cleaning, debridement, and drainage, but for some wounds, these methods do not have obvious promoting effects on their healing. With the continuous development and maturity of plasma technology, it has broad application prospects in the medical field.

[0003] Therefore, it is desirable to provide a negative pressure low-temperature plasma wound treatment system, device and medium, which can more accurately treat wounds through plasma technology, accelerate the healing process of wounds, and reduce the ulceration of wounds. SUMMARY

[0004] One or more embodiments of the present specification provide a negative pressure low-temperature plasma wound treatment system, which comprises a space module, a plasma module, a negative pressure module, and a control module. The space module comprises a closed space with adjustable size, and the closed space is used to place a wound. The plasma module is used to generate plasma gas, and the plasma module is connected to the space module through a flexible pipeline to deliver the plasma gas into the closed space. The negative pressure module comprises a fan and an exhaust port, and the exhaust port is arranged on at least one face of the closed space. The fan extracts the gas in the closed space. The control module is used to generate a control instruction and control the working parameters of the closed space, the plasma module, and the fan through the control instruction.

[0005] One or more embodiments of the present specification provide a negative pressure low-temperature plasma wound treatment device, which comprises a processor. The processor is used to generate a control instruction and control the working parameters of the closed space, the plasma module, and the fan of the negative pressure low-temperature plasma wound treatment system according to any one of the above embodiments through the control instruction.

[0006] One or more embodiments of the present specification provide a computer-readable storage medium. The storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes the generation of a control instruction and controls the working parameters of the closed space, the plasma module, and the fan of the negative pressure low-temperature plasma wound treatment system according to any one of the above embodiments through the control instruction. BRIEF DESCRIPTION OF DRAWINGS

[0007] The present specification will be further described in the manner of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same numbers represent the same structures, wherein:

[0008] Figure 1 is a schematic diagram of a negative pressure cryogenic plasma wound treatment system according to some embodiments of the present specification;

[0009] Figure 2 is an exemplary flow chart of the function of a control module of a negative pressure cryogenic plasma wound treatment system according to some embodiments of the present specification;

[0010] Figure 3 is a structure diagram of a first prediction model according to some embodiments of the present specification;

[0011] Figure 4 is an exemplary flow chart of a method of determining working time according to some embodiments of the present specification;

[0012] Figure 5 is a structure diagram of a judgment model according to some embodiments of the present specification. DETAILED DESCRIPTION

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can also be applied to other similar scenarios without creative labor on the basis of these drawings. Unless it is clear from the language context or otherwise indicated, the same reference numbers in the drawings represent the same structures or operations.

[0014] It should be understood that the "system", "device", "unit" and / or "module" used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

[0015] As shown in the present specification and claims, unless the context clearly indicates otherwise, "one", "a", "an" and / or "the" do not refer to the singular, but can also include the plural. Generally speaking, the terms "comprise" and "include" only indicate the inclusion of the steps and elements explicitly identified, and these steps and elements do not constitute an exclusive list, and the method or device can also include other steps or elements.

[0016] Flowcharts are used in the present specification to illustrate the operations performed by the system according to the embodiments of the present specification. It should be understood that the preceding or subsequent operations are not necessarily performed in sequence. On the contrary, each step can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or more steps of the operation can be removed from these processes.

[0017] Figure 1 is a schematic diagram of a negative pressure cryogenic plasma wound treatment system according to some embodiments of the present specification. The negative pressure cryogenic plasma wound treatment system 100 according to some embodiments of the present specification will be described in detail below. It should be noted that the following embodiments are only used to explain the present specification and do not constitute a limitation on the present specification. For example, the negative pressure cryogenic plasma wound treatment system can be used for flexible plasma, flexible cold plasma, low temperature plasma and other application scenarios.

[0018] As shown in Figure 1 , the negative pressure cryogenic plasma wound treatment system 100 can include a space module 110, a plasma module 120, a negative pressure module 130 and a control module 140.

[0019] Plasma refers to a highly ionized gas cloud generated by a gas under the action of a strong electromagnetic field, mainly composed of electrons, ions, atoms, molecules, free radicals and other active substances. Plasma is a new form of matter in addition to solid, liquid and gas. It can be understood that the total number of positive charges in the plasma is equal in value to the number of negative charges. With the development of plasma wound treatment, plasma can deeply clean and sterilize efficiently. Therefore, plasma wound treatment has rapidly developed in clinical medicine due to its small skin irritation, light patient pain and convenient use, and can be widely used in the medical field.

[0020] The space module 110 can refer to a sealed space for accommodating all or part of the body part where the wound is located.

[0021] In some embodiments, the space module 110 can include at least one camera for taking a wound image. The space module 110 can also include a sensor, and the camera can take a picture after the sensor detects the object to be processed when the object to be processed is inserted into the sealed space.

[0022] In some embodiments, the space module 110 can also include a user terminal outside the space module 110 for interacting with the user, for example, a touch screen, through which the user can input wound parameters. The details of the wound parameters can be referred to the description of other contents of the present specification, for example, Figure 2 .

[0023] In some embodiments, the space module 110 includes a sealed space with adjustable space size, and the sealed space is used to place the wound.

[0024] The closed space can refer to a space for treating a wound. In some embodiments, a plasma gas can be introduced into the closed space for treating the wound. The number of closed spaces can be at least one. The shape of the closed space can be various, and a suitable closed space can be selected according to the size of the body part where the wound is located, the shape of the wound and / or the body part where the wound is located, so that the closed space is relatively closed and the efficiency of wound treatment can be improved. For example, for a wound on the wrist, a closed space that fits the shape of the wrist can be selected.

[0025] The closed space can be a relatively closed space for accommodating a wound. The space module 110 can be divided into a closed type, a semi-open type, etc. One side of the semi-open space module 110 can be provided with an opening, which can be part or the entire side of the closed space, and the opening can form a relatively closed space in contact with the body part. Further, the side in contact with the body part is provided with a material having elasticity and strength, such as rubber, which can make the contact between the space module 110 and the skin tighter and form a relatively closed space. For example, when the wound is on the stomach, the space module 110 can be placed on the stomach where the wound is located, and the corresponding closed space can be formed by pressing or vacuuming.

[0026] The closed space module 110 can have an inlet and / or an outlet for the body part where the wound is located to extend in and / or out. The material of the inlet or outlet can have a certain elasticity and strength, and the body part where the wound is located can extend into the closed space through the inlet and can extend out from the outlet, for example, when the wound is at the elbow, the arm can extend into the closed space through the inlet and extend out from the outlet, and the wound can be located in the closed space.

[0027] In some embodiments, the closed space can include at least one movable side that changes the size of the closed space based on the movement of the lifting device. In some embodiments, the outside of the space module 110 can be provided with a lifting device for moving one movable side of the closed space. The part of the movable side in contact with other parts of the closed space can be provided with a material having elasticity and strength. The liftable device can be a push rod, a screw rod, etc.

[0028] In some embodiments of the present specification, by providing a movable side in the closed space, the size of the closed space can be flexibly adjusted according to the size and location of the wound, which has strong applicability and can improve the efficiency of wound treatment.

[0029] In some embodiments, the space module 110 can include an internal controller, which can be used to control the working mode of the closed space.

[0030] The plasma module 120 can refer to a device for generating plasma. For example, the plasma module 120 can be a plasma generator, such as an arc plasma generator, a power frequency arc plasma generator, a high-frequency induction plasma generator, a low-pressure plasma generator, a combustion plasma generator, etc. In some embodiments, the plasma module 120 is used to generate plasma gas, and the plasma module 120 is connected to the space module 110 through a flexible pipe to deliver the plasma gas to the sealed space.

[0031] The working parameters of the plasma module 120 can include power, working duration, etc. The working duration of the plasma module 120 can be determined by searching in the wound database. The details of the wound database can be found in the detailed description of other parts of the specification, for example, Figure 4 .

[0032] In some embodiments, the corresponding relationship between the working parameters of the plasma module 120 and the generation speed of the plasma gas can be preset based on historical data. For example, when the power of the plasma module 120 is 10kw, the corresponding generation speed of the plasma gas is 1L / min.

[0033] The negative pressure module 130 can refer to a device that can make the sealed space into negative pressure, such as a gas pump, etc. The negative pressure module 130 can extract gas from the sealed space. In some embodiments, the negative pressure module 130 can include a fan and an exhaust port, and the exhaust port can be arranged on at least one face of the sealed space, and the fan extracts the gas in the sealed space.

[0034] The working parameters of the fan can include the extraction power of the fan, the working duration, etc.

[0035] In some embodiments, the working duration of the fan can be determined by searching in the wound database. The details of the wound database can be found in the detailed description of other parts of the specification, for example, Figure 4 In some embodiments, the working duration of the fan can be the same as the working duration of the plasma module 120.

[0036] The control module 140 can be used to generate control instructions, and control the working parameters of the sealed space, the plasma module and the fan through the control instructions.

[0037] In some embodiments, the working parameters of the sealed space can include the working mode and / or the lifting parameters of the movable face. The details of the working parameters of the sealed space can be found in the description of other parts of the specification, for example, Figure 2In some embodiments, the control module 140 can directly control the working mode of the sealed space and / or the lifting parameter of the movable surface through control instructions. In some embodiments, the space module 110 can further comprise an internal controller, and the control module 140 can indirectly control the working parameter of the sealed space through the internal controller.

[0038] In some embodiments, the control module 140 can adjust the working parameter of the plasma gas generation module 120 to adjust the speed of the plasma gas generation, etc. For example, when the speed of the plasma gas generation is too slow, the power of the plasma module 120 can be increased. In some embodiments, after the control module 140 determines the speed of the plasma gas generation, based on the corresponding relationship between the power of the plasma module 120 and the speed of the plasma gas generation, the control module 140 can determine the corresponding working parameter of the plasma module 120 and control the plasma module 120 based on the working parameter.

[0039] In some embodiments, the control module 140 can determine the extraction power of the blower through a prediction model. The details of the prediction model can be found in the details of other parts of the specification, for example, Figure 3 The higher the extraction power of the blower is, the higher the efficiency of extracting the gas in the sealed space is, and the shorter the time is. When the plasma gas is used to treat the wound, ozone can be generated, so the control module 140 needs to control the blower to extract the generated ozone as quickly as possible to reduce the damage of the ozone to the normal skin or the wound. At the same time, after the ozone is extracted, a negative pressure and low oxygen environment is formed in the sealed space, which is beneficial to the healing of the wound.

[0040] It should be noted that the above description of the negative pressure and low temperature plasma wound treatment system 100 and its modules is for the convenience of description, and cannot limit the specification to the scope of the embodiments. It can be understood that, for those skilled in the art, after understanding the principle of the system, any combination of the modules or connection of the modules to form a subsystem can be made without departing from the principle. In some embodiments, Figure 1 The space module 110, the plasma module 120, the negative pressure module 130 and the control module 140 disclosed in the specification can be different modules in a system, or one module can realize the functions of two or more modules. For example, the modules can share a storage module, and the modules can also have their own storage modules. Variations such as this are within the scope of protection of the specification.

[0041] Figure 2 is an exemplary flowchart of the function of the control module of the negative pressure and low temperature plasma wound treatment system according to some embodiments of the specification. In some embodiments, the flowchart 200 can be executed by the control module. As Figure 2As shown, the flow 200 includes the following steps:

[0042] At step 210, an image of the wound surface is acquired by a camera arranged in the sealed space.

[0043] The wound surface image can refer to an image including the wound surface. The user can first extend the wound surface into the space module, or the camera can sense that an object has entered the space module, or the user of the negative pressure low-temperature plasma wound surface treatment system 100 can control the camera to acquire the wound surface image. In some embodiments, the control module can be connected to the space module to acquire the acquired wound surface image.

[0044] At step 220, a first control instruction is generated based on the wound surface image.

[0045] The first control instruction can refer to an instruction for controlling the space module. The first control instruction can include a command for controlling the space module to work according to the working parameters of the sealed space. For details of the working parameters of the sealed space, please refer to the description in other parts of the specification, for example, step 230.

[0046] In some embodiments, in the wound surface database, the historical wound surface vectors correspond one-to-one to the historical first control instructions. The control module can determine the historical first control instruction based on the historical wound surface vector with the highest similarity to the current wound surface vector, and take the historical first control instruction as the first control instruction.

[0047] In some embodiments, the control module can send a reminder information to the user terminal. The reminder information is used to remind the user to input the wound surface parameters and to remind the user to confirm the wound surface placement position. The user inputs the wound surface parameters from the user terminal based on the reminder information, and the control module acquires the wound surface parameters input by the user through the user terminal, and generates the first control instruction based on the wound surface parameters and the wound surface image.

[0048] In some embodiments, when the user extends the body part where the wound surface is located into the sealed space, the sensor inside the space module detects the body part and uploads information to the built-in controller of the sealed space, and then the built-in controller controls the camera to acquire the wound surface image and sends it to the control module. The control module can determine whether the current wound surface placement position is standard by the position model based on the wound surface image.

[0049] The position model can be a machine learning model, for example, it can be a graph neural network model. The input of the position model is the wound surface image, and the output is whether the wound surface image is placed standardly. The position model can be acquired based on historical data training.

[0050] When the user stretches the body part where the wound is located into the closed space, the camera captures the wound image, and the control module inputs the wound image into the position model. The control module can determine whether the user needs to be reminded to reposition based on the output of the position model. When the user's position is identified to be in compliance with the standard, the user is sent a reminder message to confirm the position of the wound and input the wound parameters.

[0051] The reminder message can refer to information reminding the user to perform subsequent operations. For example, the reminder message can include "please confirm that the position has been placed" and "please input the wound parameters". The reminder message can be displayed to the user through an interactive touch screen outside the space module.

[0052] In some embodiments, after the space module receives the wound parameters input by the user, it sends them to the control module. After the control module obtains the wound parameters and the wound image, it generates the first control instruction.

[0053] The wound parameters input by the user and the wound image can be used to more accurately determine the first control instruction, thereby improving the treatment efficiency.

[0054] Step 230: The first control instruction is sent to the space module to control the working parameters of the closed space based on the first control instruction.

[0055] In some embodiments, the closed space can include at least one movable surface. For details about the movable surface, please refer to the description of other parts of this specification, for example, Figure 1 In some embodiments, the working parameters of the closed space include the working mode and / or the lifting parameters of the movable surface.

[0056] The working mode of the closed space can refer to the position state of the closed space relative to the body part where the wound is located when the space module is working. In some embodiments, the working mode of the closed space can include placing the body part where the wound is located into the closed space, and placing the opening of the space module on the body part around the wound, etc. The working mode of the closed space can be determined according to the area of the wound and / or the body part where the wound is located. For example, when the wound is on the back and the area is small, the working mode can be to place the opening of the space module on the body part around the wound.

[0057] The lifting parameter can refer to the parameter for controlling the lifting of the movable surface of the closed space. In some embodiments, the lifting parameter can include lifting or lowering, lifting or lowering speed, and lifting or lowering distance.

[0058] In some embodiments, the control module controls the closed space based on the first control instruction. For example, the first control instruction determined by the control module based on the wound parameter and the wound image comprises: tightly closing the opening of the space module on the body part around the wound, raising the movable surface, the raising speed being 12 cm / min, the raising distance being 1 cm, and then sending the first control instruction to the internal controller of the space module, and the internal controller controls the closed space based on the first control instruction.

[0059] Through the working parameters of the closed space controlled by some embodiments of the present specification, the space module can be more targeted when working, different working modes can be used according to different wounds, and the processing efficiency and processing quality can be improved.

[0060] In some embodiments, the control module can determine the working parameters of the closed space by searching in the wound database.

[0061] In some embodiments, the control module can preset the wound database, and the wound database comprises historical wound vectors corresponding to each wound constructed based on historical wound data. In the wound database, the historical wound vector corresponds to the historical working parameters and the historical volume parameters of the closed space one by one. The historical volume parameter can refer to the volume of the closed space when treating the wound.

[0062] The wound data can refer to data related to the current wound. The wound data can include wound parameters and wound images.

[0063] In some embodiments, the control module can obtain information related to the wound from the wound image, which can include the area of the wound, the current state of the wound, etc. The current state of the wound can include the infection period, the recovery period, the repair period, etc.

[0064] The wound parameter can refer to a parameter for characterizing the current wound. The wound parameter can include the wound type, the body part where the wound is located, the time when the wound is caused, the state of the current wound, the type and time of the last treatment received by the wound, etc. The wound type can include abrasions, contusions, burns, etc. The wound parameter can be input by the user through the touch screen outside the space module.

[0065] The wound vector can refer to data in the form of a vector reflecting the characteristics of the wound. For example, the determination rule of the wound vector can be that the numerical value of each dimension of the wound vector respectively represents the wound type, wound location, wound period, and wound size, etc. For example, the first element in the wound vector is the wound area, and the second element is the current state of the wound. Each element value can be represented by a code, and each element value can correspond to the content determined from the wound parameters and wound images. The element value can be a number, a letter, or other expressions. For example, for the wound area, 1 represents 0-1 square centimeters, 2 represents 1-10 square centimeters, 3 represents 10-30 square centimeters, etc.; for the current state of the wound, 1 represents the infection period, 2 represents the recovery period, 3 represents the repair period, etc.; for the wound type, 1 represents burn, 2 represents contusion, 3 represents abrasion, etc.; for the body part where the wound is located, 1 represents the face, 2 represents the abdomen, 3 represents the arm, 4 represents the hand, etc. Then the wound vector (1, 1, 2, 4) can represent that the wound area is between 0-1 square centimeters, the wound state is in the infection period, the wound type is contusion, and the wound is located in the hand. In some embodiments, the wound vector can be obtained through the embedding layer of the prediction model, and the details of the prediction model can be referred to the description in other parts of the specification, for example, Figure 3 .

[0066] In some embodiments, the wound vector can be obtained based on the wound data of the current wound and retrieved in the wound database to determine the working parameter reference vector.

[0067] In some embodiments, the control module can calculate the distance between the wound vector of the current wound and the historical wound vector in the wound database, and determine the working distance reference vector based on the vector distance. For example, the historical wound vector that meets the preset condition with the wound vector can be used as the working distance reference vector. The preset condition can be set according to the situation. For example, the preset condition can be that the vector distance is the smallest or the vector distance is less than the distance threshold, etc.

[0068] In some embodiments, the control module can determine the historical working parameter of the sealed space matched with the working distance reference vector through the working distance reference vector, and use the historical working parameter as the working parameter of the current sealed space.

[0069] In some embodiments, the control module can determine the historical volume parameter of the sealed space matched with the working distance reference vector through the working distance reference vector, and use the historical volume parameter as the volume parameter of the current sealed space.

[0070] By determining the working parameter of the current sealed space in the wound database, the accuracy of the determined working parameter can be improved, and it is more consistent with the actual situation.

[0071] The first control instruction can be more in line with actual needs by the wound image, and the treatment efficiency can be improved.

[0072] In some embodiments, the control module 140 can also be configured to generate a third control instruction.

[0073] The third control instruction can refer to an instruction for controlling the generation speed of the plasma gas and the extraction power of the fan.

[0074] The generation speed of the plasma gas can refer to the gas flow rate when the plasma gas enters the sealed space of the space module 110 from the plasma module 120. For example, the generation speed of the plasma gas can be 5 cm / s. 3 / s.

[0075] The extraction power of the fan can refer to the working power of the fan of the negative pressure module 130 when the gas in the sealed space of the space module 110 is extracted. For example, the extraction power of the fan can be 10 W.

[0076] In some embodiments, the third control instruction can be determined in various ways based on the relevant information of the wound, for example, it can be determined by looking up a preset instruction table, or the relevant information of the wound can be processed based on a machine learning model to determine the third control instruction.

[0077] In some embodiments, the third control instruction can be determined by a first prediction model. The first prediction model can be a machine learning model.

[0078] Figure 3 is a structural diagram of the first prediction model according to some embodiments of the present specification.

[0079] As shown in Figure 3 , the first prediction model 300 can include an embedding layer 320, a feature extraction layer 340, a first prediction layer 350, and a second prediction layer 380. The first prediction model 300 can determine the plasma gas generation speed 370 and the fan power 390 based on the processing of the wound data 310-1, the environmental parameters 310-2, and the sealed space parameters 310-3, and then determine the third control instruction based on this. The relevant description of the wound data 310-1, the environmental parameters 310-2, and the sealed space parameters 310-3 can be referred to in the following description.

[0080] In some embodiments, the embedding layer 320 of the first prediction model 300 can determine the wound vector 330 based on the wound data 310-1. The wound data 310-1 can include wound images, user-input wound parameters, etc. The relevant description of the wound parameters can be referred to in the following description. Figure 2and the related description. The wound vector 330 can refer to data in the form of a vector generated based on the wound image and the wound parameters, which can reflect the wound feature information. For the content and determination method of the wound vector, please refer to Figure 2 and the related description.

[0081] The first prediction layer 350 of the first prediction model 300 can determine the plasma gas generation speed 370 based on the wound vector 330.

[0082] The input of the first prediction layer 350 can be the wound vector 330, and the output can be the plasma gas generation speed 370. For the related description of the plasma gas generation speed, please refer to the foregoing description.

[0083] The feature extraction layer 340 of the first prediction model 300 can determine the environmental feature 360 based on the processing of the environmental parameters 310-2. The environmental parameters can refer to feature data parameters of the environment outside the wound, which can include environmental temperature, air pressure value, humidity, etc. The environmental feature can refer to data generated based on the environmental parameters, which can reflect the environmental feature information outside the wound. The environmental feature can be characterized by data in the form of a vector. For example, the content of a certain environmental feature can be (28, 101.3, 50%), representing that the temperature of the surrounding environment outside the wound is 28°C, the air pressure is 101.3 kPa, and the air humidity is 50%.

[0084] The second prediction layer 380 of the first prediction model 300 can determine the fan power 390 based on the processing of the environmental feature 360 and the plasma gas generation speed 370, and the sealed space parameters 310-3. For the related description of the fan extraction power, please refer to the foregoing description. The sealed space parameters can refer to the working parameters of the sealed space. For the related description of the working parameters of the sealed space, please refer to Figure 2 and the related description.

[0085] In some embodiments, the embedding layer can be trained first, and then the embedding layer can be jointly trained with other layers.

[0086] In some embodiments, the embedding layer 320 can be trained by a plurality of labeled training samples. For example, a plurality of labeled training samples can be input into the initial embedding layer, a loss function can be constructed by the label and the result of the initial embedding layer, and the parameters of the initial embedding layer can be iteratively updated based on the loss function. When the loss function of the initial embedding layer meets a preset condition, the training is completed, and the trained embedding layer is obtained. The preset condition can be that the loss function converges, the number of iterations reaches a threshold, etc.

[0087] In some embodiments, the training samples can include a plurality of groups of wound images and wound parameters. The label can be the wound vector corresponding to the sample, which can be determined based on manual annotation.

[0088] In some embodiments, the embedding layer 320 can be synchronously trained by training the judgment model. The structure of the judgment model and the related description of its training can be referred to Figure 5 and the related description thereof.

[0089] In some embodiments, the sample data for joint training of the embedding layer and the first prediction layer, the feature extraction layer, and the second prediction layer includes sample wound surface data, sample environment parameters, and sample closed space parameters, and the label includes sample fan extraction power. The sample wound surface data is input into the trained embedding layer to obtain a wound surface vector output by the embedding layer; the wound surface vector is input into the first prediction layer as training sample data to obtain a plasma gas generation speed output by the first prediction layer; the sample environment parameters are input into the feature extraction layer to obtain an environment feature output by the feature extraction layer; the environment feature, the plasma gas generation speed, and the sample closed space parameters are input into the second prediction layer as training sample data to obtain a fan extraction power output by the second prediction layer; a loss function is constructed based on the sample fan extraction power and the fan extraction power output by the second prediction layer, and the parameters of the feature extraction layer, the first prediction layer, and the second prediction layer are synchronously updated. Through parameter updating, the trained feature extraction layer, the first prediction layer, and the second prediction layer are obtained.

[0090] In some embodiments, the loss function involved in the joint training of each layer of the first prediction model 300 can include multiple loss terms. The multiple loss terms are respectively constructed based on the outputs of the feature extraction layer, the first prediction layer, and the second prediction layer. Exemplarily, the form of the loss function can be shown in the following formula (1):

[0091]

[0092] wherein, is a loss function symbol; , and respectively represent loss terms constructed based on the outputs of the feature extraction layer, the first prediction layer, and the second prediction layer; , and are constants, respectively representing , and weight values of is a regular term, and its value can also be preset.

[0093] In some embodiments, the weight value of the loss term in the loss function constructed based on the output of the first prediction layer and the weight value of the loss term in the loss function constructed based on the output of the second prediction layer can be determined based on the wound severity. For example, the weight value of the loss term corresponding to the first prediction layer in the loss function can be positively correlated with the wound severity, i.e., the greater the wound severity, the greater the weight value of the loss term corresponding to the first prediction layer in the loss function.

[0094] The wound severity can be comprehensively evaluated based on the wound type, wound location, wound period, and wound area size. Specifically, the wound severity can be represented by a severity score, and the greater the score, the more severe the wound. The severity score corresponding to each wound type, wound location, wound period, and wound area size can be preset. For example, the severity scores corresponding to burn, scald, and abrasion in the wound type are 5, 4, and 1, respectively; the severity scores corresponding to face, abdomen, neck, and hand in the wound location are 8, 4, 6, and 1, respectively; the severity scores corresponding to infection period, recovery period, and repair period in the wound period are 8, 3, and 2, respectively; and the severity scores corresponding to 0-1 square centimeters, 1 (not including) -5 square centimeters, and 5 (not including) -10 square centimeters in the wound area size are 1, 4, and 7, respectively.

[0095] For example, if the wound type of a patient is scald, the wound is located on the hand, is in the recovery period, and has an area size of 4.6 square centimeters, then the severity score of the wound is 4+1+3+4=12, which represents that the severity of the wound is higher than that of other wounds with a severity score less than 12.

[0096] In some embodiments, for wounds located at specific parts, the weight values of the loss terms corresponding to the first prediction layer and the second prediction layer in the loss function can be preset in advance. The specific parts can be relatively fragile parts such as the ear, eye, and front neck, so that the wind power for the wound at the specific part can be stabilized within a preset range, and the plasma gas generation speed can also be stabilized within a safe range to avoid causing additional damage to the wound.

[0097] In some embodiments of the present specification, the model is trained by joint training, and the output of part of the layers is used as training sample data to train other layers during training, which effectively improves the training efficiency and the accuracy of the output of the model.

[0098] In some embodiments of the present specification, the weight value of the loss term in the loss function during model training based on the wound location can make the output of the model more in line with the needs of wound treatment.

[0099] In some embodiments of the present specification, by dividing the model into multiple levels, the accuracy of the output of the model is greatly improved.

[0100] In some embodiments of the present disclosure, the content of the model determined part of the instructions can be determined by processing the relevant sample data to obtain the required result data quickly, thereby saving time and improving the accuracy and adaptability of the result.

[0101] In some embodiments, the control module 140 can be further configured to generate second control instructions.

[0102] The second control instructions can be instructions for controlling the working time of the fan in the plasma module 120 and the negative pressure module 130.

[0103] In some embodiments, the second control instructions can be determined in various ways based on the relevant information of the wound surface. For example, the second control instructions can be determined by looking up a predetermined instruction table, or by processing the relevant information of the wound surface based on a machine learning model.

[0104] In some embodiments, the working time in the second control instructions can be based on the similarity between the wound surface data and the historical wound surface data, and the historical working time corresponding to the highest similarity historical wound surface data can be selected as the working time.

[0105] Figure 4 is an exemplary flowchart of a method for determining the working time according to some embodiments of the present disclosure.

[0106] As shown in Figure 4 , step 400 can include the following steps.

[0107] Step 410, obtaining wound surface data and constructing a wound surface vector based on the wound surface data.

[0108] For more information about the wound surface data and the wound surface vector, please refer to Figure 3 and the related description.

[0109] In some embodiments, the wound surface vector can be constructed by a model based on processing of the wound surface data. The model can be a machine learning model. The input of the model can be the wound surface data, and the output can be the wound surface vector.

[0110] In some embodiments, the wound surface vector can be constructed by an embedding layer based on processing of the wound surface data. The embedding layer can be a convolutional neural network model. The input of the embedding layer can be the wound surface data, and the output can be the wound surface vector.

[0111] In some embodiments, the embedding layer can be obtained by synchronous training based on a training judgment model, and the judgment model is a machine learning model.

[0112] Figure 5is a structural diagram of a judgment model according to some embodiments shown in the specification. As shown in Figure 5 The judgment model 500 can include two embedding layers A 520-1 and B 520-2 with the same parameters, and a judgment layer 540. The judgment model 500 can determine the parameter similarity 550 of the two wound data based on the processing of the wound data A 510-1 and the wound data B 510-2. The related description of the parameter similarity 550 can be referred to in the following description.

[0113] The embedding layer A 520-1 and the embedding layer B 520-2 of the judgment model 500 can be two embedding layers with the same parameters. The embedding layer A 520-1 can determine the wound vector A 530-1 based on the processing of the wound data A 510-1. The embedding layer B 520-2 can determine the wound vector B 530-2 based on the processing of the wound data B 510-2.

[0114] The judgment layer 540 of the judgment model 500 can determine the parameter similarity 550 of the two wound data based on the processing of the wound vector A 530-1 and the wound vector B 530-2. The parameter similarity 550 can refer to the degree of similarity between the wound vector A 530-1 and the wound vector B 530-2. The parameter similarity 550 can be calculated based on the vector distance between the wound vector A 530-1 and the wound vector B 530-2. The smaller the vector distance, the higher the parameter similarity 550, that is, the parameter similarity 550 can be negatively correlated with the vector distance. For example, the relationship between the parameter similarity 550 and the vector distance can be shown in the following formula (2):

[0115]

[0116] wherein, is the parameter similarity; is the vector distance; is a constant, which can be pre-set.

[0117] The calculation method of the vector distance can include but is not limited to Euclidean distance, Manhattan distance, Chebyshev distance, etc.

[0118] In some embodiments, the embedding layer A 520-1, the embedding layer B 520-2 and the judgment layer 540 of the judgment model 500 can be jointly trained. The sample data for jointly training the embedding layer A, the embedding layer B and the judgment layer includes sample wound data, and the label includes sample parameter similarity. Part of the sample wound data is input into the embedding layer A to obtain a wound vector A output by the embedding layer A; the remaining part of the sample wound data is input into the embedding layer B to obtain a wound vector B output by the embedding layer B; the wound vector A and the wound vector B are taken as training sample data and input into the judgment layer to obtain a parameter similarity output by the judgment layer; a loss function is constructed based on the sample parameter similarity and the parameter similarity output by the judgment layer, and the parameters of the embedding layer A, the embedding layer B and the judgment layer are synchronously updated. Through parameter updating, the trained embedding layer A, the trained embedding layer B and the trained judgment layer are obtained.

[0119] In some embodiments, the sample data for jointly training can be historical wound data, and the label can be obtained by manual annotation.

[0120] In some embodiments of the present specification, by placing the embedding layer into other models and synchronously training the embedding layer based on the training of other models, the situation that the embedding layer is difficult to train due to the fact that the training sample data cannot meet the training quantity requirement or the sample label is difficult to obtain when the embedding layer is trained alone is effectively avoided, and the requirement for training data is reduced.

[0121] In some embodiments of the present specification, the wound vector is determined based on the wound data by a specific model, and the accuracy of the information reflected by the wound vector is effectively improved.

[0122] Step 420, searching in the wound database to determine a reference vector.

[0123] The reference vector can refer to vector form data determined based on historical wound vectors in the wound database, which is used to determine the reference duration subsequently. The historical wound vector can refer to vector form data determined based on historical wound data, which can reflect the characteristics of the historical wound. The form of the historical wound vector can be the same as that of the wound vector. For example, the content of a certain historical wound vector can be (1, 2, 3, 1), which represents that the wound type is burn, the wound is located in the abdomen, is in the repair period, and the area size is between 0-1 square centimeters. The correspondence between the numerical value of each dimension of the historical wound vector and the wound data can be the same as the correspondence between the numerical value of each dimension of the wound vector and the wound data. The related description of the correspondence between the numerical value of each dimension of the wound vector and the wound data can be referred to in the related description of the wound vector and the like. Figure 2

[0124] ​In some embodiments, a historical wound vector satisfying a preset condition can be selected from a plurality of historical wound vectors in the wound database as a reference vector. For example, the preset condition can be that the historical wound vector has the maximum similarity with the wound vector, or the similarity between the historical wound vector and the wound vector is greater than a similarity threshold, etc. The similarity threshold can be preset. The similarity can refer to the degree of similarity between the historical wound vector and the wound vector, which can be calculated based on the vector distance between the wound vector and each historical wound vector. The smaller the vector distance, the higher the similarity, i.e., the similarity can be negatively correlated with the vector distance. For example, the relationship between the similarity and the vector distance can be shown in the following formula (3):

[0125]

[0126] wherein, is the similarity; is the vector distance; is a constant, which can be preset.

[0127] The method for calculating the vector distance can include, but is not limited to, Euclidean distance, Manhattan distance, Chebyshev distance, etc.

[0128] In some embodiments, the reference vector can be constructed only based on the historical wound vector corresponding to the data of the historical wound that has not been treated twice within a preset time period. That is, when selecting the reference vector from the historical wound vector, the historical wound vector generated based on the secondary treatment process is not considered.

[0129] The secondary treatment can refer to the process of adjusting the data of the corresponding historical wound for a second time plasma wound treatment after the wound has not achieved the expected treatment effect after the first plasma wound treatment.

[0130] In some embodiments of the present specification, by screening the historical data in the wound database, the adaptability of the selected reference data to the current wound is effectively enhanced.

[0131] Step 430, determining a reference duration based on the reference vector.

[0132] The reference duration can refer to the duration data determined based on the reference vector for determining the working duration subsequently.

[0133] In some embodiments, if the number of reference vectors is only one, the historical reference duration corresponding to the reference vector can be used as the reference duration; if the number of reference vectors is more than one, the mean value of the historical reference durations corresponding to all reference vectors can be used as the reference duration. In some other embodiments, the mean value of the historical reference durations corresponding to all reference vectors can also be a weighted mean value, and the weight value of the historical reference duration corresponding to each reference vector can be positively correlated with the similarity between the reference vector and the wound vector.

[0134] Step 440, determining the working time length based on the reference time length.

[0135] In some embodiments, the reference time length can be determined as the working time length after artificial confirmation.

[0136] In some embodiments of the present specification, by determining the working time length based on the historical data in the wound database, the adaptability of the determined time length to the current wound is enhanced, and the workload in determining the working time length is reduced.

[0137] In some embodiments of the present specification, by determining the content of part of the instructions through the correlation of the vector data, the content of the instructions can be quickly determined, saving the time cost and enhancing the adaptability of the instructions to the wound.

[0138] The above has described the basic concept, and it is obvious that the above detailed disclosure is only used as an example for the person skilled in the art, and does not constitute a limitation on the present specification. Although it is not explicitly stated here, the person skilled in the art can make various modifications, improvements and corrections to the present specification. Such modifications, improvements and corrections are suggested in the present specification, so such modifications, improvements and corrections still belong to the spirit and scope of the exemplary embodiments of the present specification.

[0139] At the same time, specific words are used in the present specification to describe the embodiments of the present specification. As "one embodiment", "an embodiment", and / or "some embodiments" means a certain feature, structure or characteristic related to at least one embodiment of the present specification. Therefore, it should be emphasized and noted that the "an embodiment" or "one embodiment" or "one alternative embodiment" mentioned in different places in the present specification does not necessarily refer to the same embodiment. In addition, some features, structures or characteristics in one or more embodiments of the present specification can be properly combined.

[0140] In addition, unless the claim explicitly states, the order of the processing elements and sequences described in the present specification, the use of numerals and letters, or the use of other names, is not intended to limit the order of the processes and methods of the present specification. Although some currently considered useful embodiments of the invention are discussed in the above disclosure through various examples, it should be understood that such details are only for the purpose of illustration, and the additional claims are not limited to the disclosed embodiments, on the contrary, the claims are intended to cover all modifications and equivalent combinations that meet the spirit and scope of the embodiments of the present specification. For example, although the system components described above can be realized by hardware devices, they can also be realized by only software solutions, such as installing the described system on existing servers or mobile devices.

[0141] For simplicity and to facilitate understanding of one or more embodiments, a description of an embodiment sometimes refers to a plurality of features in a single embodiment, drawing, or description of an embodiment. However, this method of disclosure is not to be interpreted as meaning that the claimed embodiment requires more features than are explicitly recited in the claims. In fact, claims that do not specifically claim a combination of features are intended to cover the various possible combinations of features as would be understood by a person of ordinary skill in the art.

[0142] Some embodiments use numerical values to describe components, quantities of attributes. It should be understood that such numerical values used in the description of embodiments are, in some examples, modified by the adjectives "about," "approximately," or "substantially." Unless otherwise stated, "about," "approximately," or "substantially" indicate that the described numerical value allows for a variation of ±20%. Accordingly, numerical values used in the specification and claims of some embodiments are approximations that can vary depending on the desired characteristics of the individual embodiments. In some embodiments, numerical values used in the specification and claims are approximations that can vary depending on the desired characteristics of the individual embodiments. In some embodiments, numerical values should be considered in the context of the number of significant digits used in the number and the accepted bits of precision of the number. Although the numerical ranges and parameters setting forth the broadest scope of some embodiments of the specification are approximations, the numerical values set forth in the specific examples are reported as precisely as reasonably possible. The application is not limited to the specific numerical values set forth in the examples.

[0143] Each patent, patent application, patent publication, and other material cited in this specification is hereby incorporated by reference in its entirety. In the event of inconsistencies between the disclosure of this specification and the materials incorporated by reference, the disclosure of this specification shall prevail. In the event of inconsistencies between the disclosure of this specification and the claims, the claims shall prevail. In the event of inconsistencies between the disclosure of this specification and the materials incorporated by reference (whether attached hereto or subsequently added), the disclosure of this specification shall prevail. In the event of inconsistencies between the disclosure of this specification and the description, definitions, and / or terminology used in the materials incorporated by reference, the description, definitions, and / or terminology used in this specification shall prevail.

[0144] Finally, it should be understood that the embodiments described herein are merely exemplary of the principles of the embodiments described herein. Other variations having essentially the same structure and function but different values for components, and / or different arrangements of the components can also be utilized. Accordingly, the embodiments described herein are not to be considered as limited to the examples described herein.

Claims

1. A negative pressure cryoplasma wound treatment system comprising: a space module, a plasma module, a negative pressure module, a control module, the space module comprises a closed space with adjustable size, the closed space is used for placing a wound surface, the closed space comprises at least one movable surface, and the movable surface changes the size of the closed space based on the movement of a lifting device; the plasma module is used for generating plasma gas, and the plasma module is connected with the space module through a flexible pipeline to deliver the plasma gas into the closed space; the negative pressure module comprises a fan and an exhaust port, the exhaust port is arranged on at least one surface of the closed space, and the fan exhausts the gas in the closed space; the control module is used for generating control instructions and controlling the working parameters of the closed space, the plasma module and the fan through the control instructions, including: generating a second control instruction, the second control instruction is used for controlling the working time length of the plasma module and the fan; wherein the working time length is determined based on the following manner: obtaining wound surface data, inputting the wound surface data into an embedding layer, the embedding layer outputs a wound surface vector, the embedding layer is obtained based on the synchronous training of a judgment model, the judgment model comprises two embedding layers with the same parameters and a judgment layer, the judgment model is used for determining the parameter similarity of two wound surface data; retrieving in a wound surface database to determine a reference vector; the wound surface database contains a plurality of historical wound surface vectors and corresponding historical processing time lengths; determining a reference time length based on the reference vector; determining the working time length based on the reference time length.

2. The system of claim 1, wherein the control module is configured to perform the following operations: obtaining a wound surface image captured by a camera arranged in the closed space; generating a first control instruction based on the wound surface image; sending the first control instruction to the space module to control the working parameters of the closed space based on the first control instruction.

3. The system of claim 2, wherein the control module is further configured to perform the following operations: sending a reminder information to a user terminal, the reminder information is used for reminding the user to input wound surface parameters and reminding the user to confirm the wound surface placement position; obtaining the wound surface parameters input by the user through the user terminal; generating a first control instruction based on the wound surface parameters and the wound surface image.

4. The system of claim 2, wherein the closed space comprises at least one movable surface, and the working parameters of the closed space comprise a working mode and / or a lifting parameter of the movable surface.

5. The system of claim 1, wherein the control module is further configured to generate a third control instruction for controlling a speed of the plasma gas generation and an extraction power of the fan, wherein the third control instruction is determined by a first prediction model comprising the embedding layer, the feature extraction layer, the first prediction layer, and the second prediction layer, wherein the first prediction model determines the speed of the plasma gas generation and the fan power based on processing of the wound data, the environmental parameters, and the enclosure parameters, and determines the third control instruction, comprising: determining the speed of the plasma gas generation based on the wound vector by the first prediction layer; determining environmental features based on processing of the environmental parameters by the feature extraction layer; and determining the fan power based on processing of the environmental features, the speed of the plasma gas generation, and the enclosure parameters by the second prediction layer.

6. A negative pressure low temperature plasma wound treatment device comprising a processor configured to generate a control instruction and control working parameters of an enclosure, the plasma module, and the fan of the negative pressure low temperature plasma wound treatment system of any one of claims 1-5 by the control instruction.

7. A computer readable storage medium storing computer instructions, wherein when a computer reads the computer instructions in the storage medium, the computer executes to generate a control instruction and control working parameters of an enclosure, the plasma module, and the fan of the negative pressure low temperature plasma wound treatment system of any one of claims 1-5 by the control instruction. ​ ​ ​

Images