WO2021151331A1

WO2021151331A1 - Method, apparatus and device for acquiring parameters of ventilator, and storage medium

Info

Publication number: WO2021151331A1
Application number: PCT/CN2020/124390
Authority: WO
Inventors: 黄思皖
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-09-08
Filing date: 2020-10-28
Publication date: 2021-08-05
Also published as: CN111899862B; CN111899862A

Abstract

A method, apparatus and device for acquiring parameters of a ventilator, and a storage medium, which are applied to the technical field of medical treatment. The method comprises: acquiring multiple sets of state parameters of multiple users using a ventilator, wherein each set of state parameters comprises a first state parameter of a user before a ventilator parameter adjustment policy is used, the used ventilator parameter adjustment policy, and a second state parameter of the user after the ventilator parameter adjustment policy is used; determining, according to the multiple sets of state parameters, reward and punishment information of each ventilator parameter adjustment policy, and determining a state conversion probability corresponding to the first state parameter; and according to a target state parameter of a target user, the reward and punishment information corresponding to each ventilator parameter adjustment policy and the state conversion probability, determining a target ventilator parameter adjustment policy corresponding to the target state parameter. By using the method, the efficiency and reliability of parameter setting of a ventilator can be improved.

Description

Method, device, equipment and storage medium for acquiring parameters of ventilator

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on September 8, 2020, the application number is 202010939041.2, and the invention title is "A method, device, equipment, and storage medium for obtaining parameters of a ventilator", and its entire contents Incorporated in this application by reference.

Technical field

This application relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and storage medium for obtaining parameters of a ventilator.

Background technique

At present, as a common medical device, the use of ventilator is becoming more and more common, and it has been widely used in hospitals and patients' families. The inventor realizes that in the course of use, the ventilator parameters are generally manually set by the medical staff or the patient’s family based on experience or doctor’s advice. The parameter adjustment cannot be made according to the actual situation of the patient, and the reliability of the ventilator parameter setting is poor.

Summary of the invention

The embodiments of the present application provide a method, device, equipment, and storage medium for acquiring parameters of a ventilator, which help to improve the efficiency of setting the parameters of the ventilator and improve the reliability of setting the parameters of the ventilator.

In the first aspect, an embodiment of the present application provides a method for obtaining parameters of a ventilator, including:

Acquire multiple sets of state parameters corresponding to multiple users using the ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter The second state parameter of the user after the policy is adjusted;

Determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine the corresponding state transition probability of the first state parameter under each ventilator parameter adjustment strategy , The state conversion probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter;

Obtain the target state parameters of the target user;

According to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, the target ventilator parameter adjustment strategy corresponding to the target state parameter is determined; wherein, the target ventilator parameter adjustment strategy is used To indicate a parameter adjustment strategy for the ventilator of the target user.

In the second aspect, an embodiment of the present application provides a ventilator parameter acquisition device, including:

The acquiring module is used to acquire multiple sets of state parameters corresponding to multiple users who use the ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter adjustment strategy. The second state parameter of the user after the ventilator parameter adjustment strategy;

The determining module is configured to determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine that the first state parameter is under each ventilator parameter adjustment strategy A corresponding state transition probability, where the state transition probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter;

The acquiring module is also used to acquire target state parameters of the target user;

The determining module is further configured to determine the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability; The target ventilator parameter adjustment strategy is used to indicate a parameter adjustment strategy for the ventilator of the target user.

In a third aspect, an embodiment of the present application provides a ventilator parameter acquisition device. The ventilator parameter acquisition device may include a processor and a memory, and the processor and the memory are connected to each other. Wherein, the memory is used to store a computer program that supports the terminal device to perform the above-mentioned methods or steps, the computer program includes program instructions, and the processor is configured to call the program instructions to perform the following steps:

Obtain the target state parameters of the target user;

In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium that stores a computer program, and the computer program includes program instructions that, when executed by a processor, cause all The processor performs the following steps:

Obtain the target state parameters of the target user;

Optionally, the computer-readable storage medium may be non-volatile or volatile.

The embodiments of the present application improve the efficiency of obtaining parameters of the ventilator, thereby improving the efficiency of setting the parameters of the ventilator, and improving the reliability of setting the parameters of the ventilator.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present application more clearly, the following will briefly introduce the drawings used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. Ordinary technicians can obtain other drawings based on these drawings without creative work.

Figure 1 is a schematic structural diagram of a ventilator parameter acquisition system provided by an embodiment of the present application;

2 is a schematic flowchart of a method for acquiring parameters of a ventilator according to an embodiment of the present application;

Fig. 3a is a schematic flowchart of another method for acquiring parameters of a ventilator according to an embodiment of the present application;

3b is a schematic flowchart of a method for training a parameter model of a ventilator according to an embodiment of the present application;

4 is a schematic structural diagram of a device for acquiring parameters of a ventilator according to an embodiment of the present application;

Fig. 5 is a schematic structural diagram of a ventilator parameter acquisition device provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

The technical solution of this application can be applied to the fields of artificial intelligence, digital medical care, smart city, blockchain and/or big data technology to improve the efficiency and reliability of ventilator parameter settings and realize smart medical care. Optionally, the data involved in this application, such as state parameters, can be stored in a database, or can be stored in a blockchain, which is not limited in this application.

The technical solution of the present application can be applied to a ventilator parameter acquisition system, and can be specifically applied to a ventilator parameter acquisition device (a ventilator parameter acquisition device) for realizing the determination of the user's ventilator parameters such as the ventilator parameter adjustment strategy , To set the parameters of the ventilator based on the ventilator parameter adjustment strategy. Optionally, the ventilator parameter acquisition device may be a terminal, a server, a ventilator, a data platform or other devices. The terminal may include a mobile phone, a tablet computer, a computer, etc., which is not limited in this application. It can be understood that, in other embodiments, the terminal may also be called other names, such as terminal equipment, smart terminal, user equipment, user terminal, etc., which are not listed here.

As a common medical device, ventilator is used more and more widely. The parameters of the ventilator are generally set manually. For example, the inventor realized that respiratory support and respiratory management for mechanically ventilated patients is implemented by medical staff through manual setting and adjustment of ventilator parameters, such as initial ventilation based on the patient's body, disease and condition, and ventilation needs, etc. The parameters of the ventilator are preset, and the manual adjustment of the parameters of the ventilator is realized according to the ventilation effect, arterial blood gas, heart and lung monitoring results, etc. The adjustment efficiency is low. Moreover, for inexperienced medical staff, the parameter adjustment cannot be tailored to the actual situation of the patient, and the reliability of the ventilator parameter adjustment is poor, and it cannot help the patient get the most suitable respiratory support and respiratory management. It may even be to a certain extent. Exacerbate the condition and increase the risk of death. However, this application can obtain the user's current state parameters, and then adjust the strategy based on the reward and punishment information under the first state parameter before adjustment and the second state parameter after adjustment based on each ventilator parameter adjustment strategy, and the first state parameter before adjustment The corresponding state conversion probability under each ventilator parameter adjustment strategy is used to determine the user's ventilator parameter adjustment strategy, thereby improving the efficiency and reliability of the ventilator parameter setting.

The technical solution of this application can be applied to the fields of artificial intelligence, smart city, blockchain and/or big data technology. For example, it can be realized through a data platform or other equipment. The data involved can be stored through blockchain nodes, or can be stored in The database is not limited in this application.

The embodiments of the present application provide a ventilator parameter acquisition system, method, device, equipment, medium, etc., so as to help improve the efficiency of ventilator parameter acquisition and improve the reliability of breathing parameter settings. Detailed descriptions are given below.

Please refer to FIG. 1, which is a schematic structural diagram of a ventilator parameter acquisition system provided by an embodiment of the present application. As shown in FIG. 1, the ventilator parameter acquisition system may include a ventilator parameter acquisition device (a ventilator parameter acquisition device) 101 and a ventilator 102. in,

The ventilator parameter obtaining device 101 is used to obtain multiple sets of state parameters corresponding to multiple users who use the ventilator, and each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy and the adopted ventilator parameter adjustment Strategy and the second state parameter of the user after adopting the ventilator parameter adjustment strategy; determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and second state parameter according to the multiple sets of state parameters, and determine The state transition probability corresponding to the first state parameter under each ventilator parameter adjustment strategy, the state transition probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter; obtain the target user According to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, determine the target ventilator parameter adjustment strategy corresponding to the target state parameter; wherein the target ventilator parameter adjustment strategy It is used to indicate the parameter adjustment strategy for the ventilator of the target user.

The ventilator 102 is configured to set the ventilator parameters according to the target ventilator parameter adjustment strategy. Alternatively, the ventilator parameter acquisition device 101 may also be used to control the ventilator 102 to set the ventilator parameters according to the target ventilator parameter adjustment strategy.

It can be understood that the ventilator and the ventilator parameter acquisition device may be independent devices, that is, independently deployed, and the ventilator and the ventilator parameter adjustment strategy may be connected to each other. Alternatively, the ventilator and the ventilator parameter acquisition device can also be deployed in the same device, for example, a ventilator that automatically determines the ventilator parameters for the user, the ventilator parameter acquisition device can be set in the ventilator, etc., This application is not limited. Figure 1 shows a standalone deployment scenario.

In some embodiments, the ventilator parameter acquisition system may further include a storage device, and the storage device may be used to store data related to this application. For example, the storage device can store the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and second state parameter (for example, the ventilator parameter adjustment strategy, the first state parameter, the second state parameter and their rewards and punishments) Information binding storage), the corresponding state transformation probability of the first state parameter under each ventilator parameter adjustment strategy (such as binding and storing the ventilator parameter adjustment strategy, the first state parameter, the second state parameter and its state transformation probability) and many more. For another example, the storage device may store the identification of the target user, target state parameters, and target ventilator parameter adjustment strategies, etc., which are not listed here. After the ventilator parameter acquisition device determines these data, it will soon be able to store these data in the storage device.

Optionally, the storage device may be a blockchain node, and the multiple sets of state parameters and other related data may be obtained from the blockchain. By storing the ventilator parameter-related data in the blockchain, and determining the ventilator parameter adjustment strategy for the user based on the data obtained from the blockchain, it helps to improve the reliability of the obtained data, which in turn helps to improve the ventilator The efficiency and reliability of parameter setting.

In some embodiments, the ventilator parameter acquisition system may further include a variety of wearable devices, and the wearable may be interconnected with the ventilator acquisition device. For example, the wearable device can be used to collect state parameters of the user. Therefore, the user state parameters can be obtained in real time, so that the ventilator parameter acquisition device can quickly determine the target ventilator parameter adjustment strategy for the user according to the collected user state parameters, thereby helping to improve the target ventilator determined by the ventilator parameter acquisition device The timeliness of the parameter adjustment strategy helps to alleviate the user's condition.

Refer to FIG. 2, which is a schematic flowchart of a method for acquiring parameters of a ventilator according to an embodiment of the present application. The method may be executed by the above-mentioned ventilator parameter acquisition device. As shown in FIG. 2, the ventilator parameter acquisition method may include the following steps:

201. Acquire multiple sets of state parameters corresponding to multiple users who use the ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter The second state parameter of the user after the policy is adjusted.

In some embodiments, the ventilator parameter obtaining device may obtain multiple sets of state parameters by obtaining historical diagnosis and treatment data of multiple users (patients) using the ventilator, and determining the state parameters of each user according to the diagnosis and treatment data of each user.

Wherein, the diagnosis and treatment data may include physical sign data and/or examination data and so on. Optionally, the physical sign data may include heart rate, blood oxygen, respiratory rate, etc., and the inspection data may include blood gas indicators, lactic acid, and other data. For example, the data can be extracted from the monitoring system.

Optionally, the diagnosis and treatment data may be obtained by processing collected raw medical data. For example, the patient's original medical data can be obtained, including the patient's historical baseline data, etc. The historical baseline data can include multiple visit records, and each visit record may include various demographic information, diagnosis, inspection, and surgical items. Further, the historical baseline data can be pre-processed. For example, the physical sign data can be obtained by sampling the collected original physical sign data in a preset time unit (for example, in a unit of 1 hour), and the original physical sign data is continuous data; For another example, you can use multiple interpolations to fill in missing values for inspection data, such as blood gas indicators, lactic acid, etc. In order to obtain the pre-processed diagnosis and treatment data. Further optionally, each group of state parameters may be partial data filtered from the diagnosis and treatment data, or may be the diagnosis and treatment data or a vector corresponding to the partial data.

Optionally, for each group of state parameters, the time interval between the collection time of the first state parameter and the time when the ventilator parameter adjustment strategy is adopted, and the time when the ventilator parameter adjustment strategy is adopted and the collection of the second state parameter The time interval between the times may both be less than the preset first time threshold; and/or, the collection time of the first state parameter and the collection time of the second state parameter are less than the preset second time threshold. This helps to better determine the influence of the adopted ventilator parameter adjustment strategy on the user's state, thereby helping to further improve the reliability of the ventilator parameter setting.

In some embodiments, the diagnosis and treatment data can be obtained from the blockchain, that is, the diagnosis and treatment data of each patient can be stored in the blockchain in advance. By obtaining the user's diagnosis and treatment data from the blockchain node (such as the storage device mentioned above), it helps to improve the reliability of the obtained diagnosis and treatment data, and thus the reliability of the determined user state parameters, which helps To improve the reliability of the ventilator parameter adjustment strategy determined based on the user's state parameters.

202. Determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine the corresponding state transition of the first state parameter under each ventilator parameter adjustment strategy Probability.

Among them, the state conversion probability can be used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter.

Optionally, the reward and punishment information may include a reward value or a penalty value, or it may be called a reward coefficient or a penalty coefficient, or it may be called a reward point or a penalty point, etc., which are not listed here. This application uses the reward value and The penalty value is explained as an example. The reward value can indicate that the ventilator parameter adjustment strategy is an adjustment strategy with better parameters in the corresponding first state. For example, the higher the reward value corresponding to the ventilator parameter adjustment strategy, the better the ventilator parameter adjustment strategy; otherwise, the penalty The value may indicate that the ventilator parameter adjustment strategy is an adjustment strategy with poor parameters in the corresponding first state. For example, the higher the penalty value corresponding to the ventilator parameter adjustment strategy, the worse the ventilator parameter adjustment strategy.

203. Obtain target state parameters of the target user.

Wherein, the target state parameter is a state parameter before the ventilator parameter setting/adjustment is performed on the target user, for example, it may correspond to the above-mentioned first state parameter.

Optionally, when obtaining the target state parameter, the target state parameter of the target user may be determined according to the diagnosis and treatment data of the target user by obtaining the diagnosis and treatment data of the target user.

Among them, the diagnosis and treatment data may include physical sign data and/or post-inspection test data and so on. Optionally, the physical sign data may include heart rate, blood oxygen, respiratory rate, etc., and the inspection data may include blood gas indicators, lactic acid, and other data. For example, the data can be extracted from the monitoring system, or collected through a wearable device, etc., which is not limited in this application.

Optionally, the diagnosis and treatment data may be obtained by processing collected raw medical data. For example, the original medical data of the target user (the current patient) can be obtained, such as records of multiple visits. Each visit record may include various demographic information, diagnosis, inspection, and surgical items. Further, the original medical data can be preprocessed to obtain the preprocessed diagnosis and treatment data, which will not be repeated here.

The target state parameter may be part of the data filtered from the diagnosis and treatment data, or may be the diagnosis and treatment data or the vector corresponding to the part of the data, so that the initial vector of the diagnosis and treatment data of the target user at the current time t'can be obtained.

In some embodiments, the aforementioned state parameters such as the first state parameter and the second state parameter may be a multi-dimensional matrix state space. Optionally, when obtaining the target state parameters of the target user, the diagnosis and treatment data of the target user may be obtained, and the diagnosis and treatment data includes multiple feature data; the multiple feature data are respectively converted into feature vectors to obtain multiple feature vectors; The multi-dimensional matrix state space of the target user is constructed according to the multiple feature vectors, and the multi-dimensional matrix state space of the target user is determined as the target state parameter. Thus, the multi-dimensional matrix state space corresponding to the target user can be obtained, so as to determine the ventilator parameter adjustment strategy of the target user based on the multi-dimensional matrix state space.

For example, by obtaining the respiratory rate and blood gas measurement value of the target user at time t'(s _t′ ), the multi-dimensional matrix state space of the target user is constructed and obtained

That is, the target state parameter of the target user is determined.

204. According to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, determine the target ventilator parameter adjustment strategy corresponding to the target state parameter.

Wherein, the target ventilator parameter adjustment strategy may be used to indicate a parameter adjustment operation/strategy to be performed on the ventilator of the target user.

In some embodiments, the ventilator parameter acquisition device may train to obtain the ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability. Further, when determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, the ventilator parameter acquisition device may input the target state parameter into the ventilator parameter model to obtain the target ventilator parameter adjustment strategy corresponding to the target state parameter . That is to say, the multiple sets of state parameters obtained above can be used as training samples, and the reward and punishment information corresponding to each ventilator parameter adjustment strategy, the state conversion probability, etc. can be determined to achieve model training to obtain the ventilator parameter model.

Optionally, the ventilator parameter acquisition device may perform ventilator parameter model training according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, according to the multiple sets of state parameters and each The reward and punishment information corresponding to the ventilator parameter adjustment strategy determines the reward and punishment function, and the state conversion function is determined according to the multiple sets of state parameters and the state conversion probability, and then the value of the ventilator parameter model can be determined according to the reward and punishment function and the state conversion function Function to train the parameter model of the ventilator.

Further optionally, the reward and punishment information may include a reward value or a penalty value. When the target ventilator parameter adjustment strategy corresponding to the target state parameter is determined based on the ventilator parameter model, the ventilator parameter acquisition device may input the target state parameter The ventilator parameter model calls the value function to calculate the reward and punishment information of the target state parameter under multiple ventilator parameter adjustment strategies, and then determines the breath with the maximum reward value of the target state parameter under the multiple ventilator parameter adjustment strategies The parameter adjustment strategy of the ventilator is used, and the parameter adjustment strategy of the ventilator with the largest reward value is used as the target ventilator parameter adjustment strategy.

In some embodiments, when the ventilator parameter acquisition device determines the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, One or more sets of state parameters that match the first state parameter and the target state parameter can be determined from the multiple sets of state parameters; and the corresponding adjustment strategy for each ventilator parameter in the matched one or more sets of state parameters can be determined. The reward and punishment information of and the state transition probability corresponding to the matched one or more sets of state parameters are determined, and the target ventilator parameter adjustment strategy corresponding to the target state parameter is determined. Optionally, the matching of the first state parameter and the target state parameter may mean that, for example, the state parameter has the greatest similarity, or the similarity is greater than the similarity threshold, and so on.

In some embodiments, the reward and punishment information includes a reward value or a punishment value. Further, when determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, the reward and punishment information of the target state parameter under multiple ventilator parameter adjustment strategies can be calculated; it is determined that the target state parameter is in the multiple ventilator parameter adjustment strategies. Under the parameter adjustment strategy, the ventilator parameter adjustment strategy with the largest reward value is used, and the ventilator parameter adjustment strategy with the largest reward value is used as the target ventilator parameter adjustment strategy.

That is to say, when determining the target ventilator parameter adjustment strategy based on the reward and punishment information and the state transformation function, it can be combined with the state transformation probability to calculate the ventilator parameter adjustment strategy corresponding to the matched one or more sets of state parameters. Regarding the reward and punishment value under the first state parameter and the possible second state parameter, the ventilator parameter adjustment strategy with the largest reward value is used as the target ventilator parameter adjustment strategy.

For example, filter out the matching two sets of state parameters, the reward and punishment information corresponding to group 1 is reward value 4, the state conversion probability is 80%, the reward and punishment information corresponding to group 2 is reward value 3, and the state conversion probability is 90%, then group 1 The reward value of the ventilator parameter adjustment strategy can be regarded as 4*80%=3.2, the reward value of the ventilator parameter adjustment strategy of group 2 can be regarded as 3*90%=2.7, then the ventilator parameters of group 1 can be adjusted The strategy is determined as the target ventilator parameter adjustment strategy. Another example is to filter out the matching two sets of state parameters. The reward and punishment information corresponding to group 1 is reward value 1, and the state conversion probability is 80%. The reward and punishment information corresponding to group 2 is punishment value 1, and the state conversion probability is 90%. The reward value of the ventilator parameter adjustment strategy of 1 can be regarded as 4*80%=0.8, and the penalty value of the ventilator parameter adjustment strategy of group 2 can be regarded as 1*90%=0.9 (or the reward value can be regarded as -1 *90%=-0.9), then the ventilator parameter adjustment strategy of group 1 can be determined as the target ventilator parameter adjustment strategy. I will not list them all here.

Further optionally, the reward and punishment value determined based on the reward and punishment information and the state conversion probability can also be weighted. For example, the higher the state conversion probability, the larger the corresponding weighting coefficient; the higher the similarity of the state parameters, the larger the corresponding weighting coefficient; or the weighting coefficient can be confirmed based on other methods, which is not limited in this application. This helps to further improve the reliability of the determined target ventilator parameter adjustment strategy.

In some embodiments, after determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, the ventilator parameter acquisition device may also verify the target ventilator parameter adjustment strategy; when adjusting the target ventilator parameter When the strategy verification is passed, the target user's ventilator is adjusted according to the target ventilator parameter adjustment strategy. Optionally, the verification operation is used to determine whether the acquired target ventilator parameter adjustment strategy is reliable. For example, whether the ventilator parameters adjusted by the target ventilator parameter adjustment strategy are within the normal parameter range, or whether they are within the normal parameter range corresponding to the target state parameter, where the normal parameter range can be preset; if it is normal Within the parameter range, the verification is determined to be successful, otherwise, the verification fails, the parameter adjustment can be performed without following the target ventilator parameter adjustment strategy, or, if the verification fails, the target ventilator parameter adjustment strategy can be sent to the designated user (For example, a designated doctor) confirms, and after the confirmation is passed, the parameters of the ventilator of the target user are adjusted according to the target ventilator parameter adjustment strategy. For another example, the verification operation can be to send the target ventilator parameter adjustment strategy to a designated user (such as a designated doctor) for confirmation. The target user's ventilator performs parameter adjustment; otherwise, the parameter adjustment may not be performed according to the target ventilator parameter adjustment strategy, etc., which are not listed here. Therefore, the reliability of the parameter setting/adjustment operation of the ventilator can be improved.

In some embodiments, after the target ventilator parameter adjustment strategy is determined, the target user's identification, the target state parameter, and the target ventilator parameter adjustment strategy may be bound and uploaded to the blockchain. Through the operation of binding and uploading the blockchain, it can be viewed by target users and related medical staff at any time, can be traced to the source, and provide a basis for subsequent treatment.

In this embodiment of the application, the ventilator parameter obtaining device may obtain multiple sets of state parameters corresponding to multiple users using the ventilator, and determine the reward and punishment information of each ventilator parameter adjustment strategy according to the multiple sets of state parameters, and determine the first The state transformation probability corresponding to the state parameter can then quickly determine the target state parameter corresponding to the target state parameter of the target user, according to the target state parameter of the target user, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state transformation probability. The target ventilator parameter adjustment strategy, which improves the efficiency of ventilator parameter acquisition, thereby improves the efficiency of ventilator parameter setting, and improves the reliability of ventilator parameter setting.

Referring to Fig. 3a, Fig. 3a is a schematic flowchart of another method for acquiring parameters of a ventilator according to an embodiment of the present application. As shown in Fig. 3a, the method for acquiring parameters of a ventilator may include the following steps:

301. Acquire multiple sets of state parameters corresponding to multiple users using a ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter The second state parameter of the user after the policy is adjusted.

302. Determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine the corresponding state transition of the first state parameter under each ventilator parameter adjustment strategy Probability.

Optionally, for the description of steps 301-302, reference may be made to the relevant description of the foregoing embodiment, which will not be repeated here.

303. According to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, train to obtain a ventilator parameter model.

Optionally, the ventilator parameter module may be a Markov model.

After the training data is obtained, a Markov decision process can be constructed based on the diagnosis and treatment data, and behavior strategy prediction can be performed, and the value function calculation can be used to maximize the expected cumulative reward to determine the most beneficial ventilator parameter adjustment strategy for the patient. For example, please also refer to Figure 3b, which is a schematic flowchart of a training method for a ventilator parameter model provided by an embodiment of the present application, which may include the following steps:

3031. Determine a reward and punishment function according to the multiple sets of state parameters and reward and punishment information corresponding to each ventilator parameter adjustment strategy.

The reward and punishment function may be used to indicate the reward and punishment index corresponding to the ventilator parameter adjustment strategy, and the reward and punishment index may be determined based on the state parameters of multiple users before the adjustment of the ventilator parameter adjustment strategy and the state parameters after the adjustment.

Optionally, each group of state parameters may include a first state parameter ( _st ) at a first time (t), a ventilator parameter adjustment strategy adopted, and a second state parameter (t+1) at a second time (t+1). s _t+1 ). Furthermore, a multi-dimensional matrix state space can be constructed according to the first state parameter. For example, obtain the initial vector of the patient's diagnosis and treatment data at time t, and construct the patient's multi-dimensional matrix state space based on the initial vector

For example, the diagnosis and treatment data may include the patient's respiratory rate and blood gas measurement value at time t, and then the corresponding vector (s _t ) of the patient's respiratory rate and blood gas measurement value at time t can be obtained to obtain a multi-dimensional matrix state space.

Further optionally, a strategy matrix space used to indicate a strategy for adjusting parameters of the ventilator may be determined. For example, by using an adjustable tidal volume, respiratory rate, inspiratory flow rate, inspiratory oxygen concentration, PEEP strategy parameters corresponding to (a _t), to build the initial policy matrix space (matrix operation set space)

a _t will affect the patient's next state s _t+1 .

For example, assuming that only two strategies (actions) are set up, breathing rate and inhaled oxygen concentration adjustment, a _t [0]={0,1} means that the respiratory rate exceeds 20 times/minute is 1, and less than 20 times/minute is 0 tael. A degree, a _t [1] = {0,1,2} indicates that the inhaled oxygen concentration is 100%, 90%, 50%, then the action set matrix space

Further, the ventilator parameter acquisition device may determine the reward and punishment function _{according to the first state parameter (st} ), the adopted ventilator parameter adjustment strategy, and the second state parameter ( _st+1)

Further, after obtaining the diagnosis and treatment data of multiple patients using the ventilator, the punishment (value) and reward (value) can be determined, for example, the ventilator parameters related to the diagnosis and treatment data and the patient's artery after adjusting the parameters can be obtained. Blood qi, vital signs and other indicators define punishment and rewards (rewards). Among them, the ventilator parameters may include tidal volume, respiratory frequency, inspiratory flow rate, inhaled oxygen concentration, positive end-expiratory pressure, and so on. For example, the punishment can be determined based on the prolonged use of the ventilator, the deterioration of blood gas indicators, the need for intubation, etc., and the reward can be determined based on the improvement of blood gas indicators, no longer using the ventilator, etc.

For example, suppose the rewards and punishments are whether intubation is required (intubation is not required)

Need intubation

) And whether the vital signs are stable

The reward and punishment function at each time can be a combination of a sigmoid function and a piecewise linear function, as shown below:

Among them, v _t may be the patient's vital sign value at time t, and [v _min , v _max ] may refer to the ideal range of vital sign values;

in,

Is the vital sign value directly related to whether the patient can be extubated,

It is in line with the range of vital signs for extubation; C ₁ to C ₇ are the relative weights of rewards and punishments, which can be set in advance.

3032. Determine a state transition function according to the multiple sets of state parameters and the state transition probability.

Wherein, the state conversion function can be used to indicate the probability of the state parameter of the user at the second time after adopting the ventilator parameter adjustment strategy under the state parameter of the first time.

The conversion status determining function, according to a first status parameter (s _t), the parameters used by the ventilator adjustment strategy (a _t) and a second state parameter (s _{t + 1),} determines the state transformation function P (s _{t + _{_{1 | s t, a t)}}} . This state transition function can be used to indicate the probability of the second state parameter (s _t+1 _{) at the second time after the user adopts the ventilator parameter adjustment strategy under the first state parameter (s t) at the first time} , That is, the state of the patient at time t and the action to be taken, and the probability of the state that the patient enters at time t+1.

3033. Determine the value function of the ventilator parameter model according to the reward and punishment function and the state conversion function, so as to obtain the ventilator parameter model through training.

Further, the defined reward and punishment function can be spliced with the defined patient's state space at time t through the state conversion function to form a value function, as shown below:

Among them, π(s _t ) can be taken under _st

In one of the strategies, γ is the relative weight of short-term rewards and long-term rewards. For example, increasing the inhaled oxygen concentration can relieve the patient’s oxygenation index in the short term, but increasing the oxygen concentration for a long time will reduce the patient’s own skills. Get it in advance.

Therefore, the strategy π(s _t ) can be adopted (according to

chosen

A certain action set in) maximizes the patient's reward in a period of time T. Specifically, the value function can be used to calculate the maximum cumulative reward expectation to obtain the best ventilator parameter adjustment strategy. For example, it is possible to traverse the patient's medical records and use the value function to calculate the maximum cumulative reward expectation through dynamic programming to obtain the best action strategy.

For example, the ventilator parameter acquisition device can give an initial strategy and the calculated initial Q based on the initial state parameters of the patient. Among them, Q-function is the expected value Q ^π (s, a) of a state-policy (state-action pairs):

It is the set of values corresponding to each group of states-strategies calculated by the above-mentioned value function. Further, according to the calculated initial Q and the new strategy, a new Q can be calculated, new estimates can be made, and iterated continuously, as shown below:

Among them, α is the learning rate, that is, the relative weight of the current estimate and the previous estimate, and γ is the relative weight of the aforementioned short-term reward and long-term reward, which can be preset.

Optionally, the iteration can be stopped when a preset condition is met, such as stopping after a preset number of iterations such as K times, or stopping when the value of the value function is greater than a preset threshold, or if the iteration time exceeds the preset time threshold Stop when it is time, or stop after iterating all the strategies in the strategy matrix space, and so on, not listed here.

Further, after the iteration stops, the optimal ventilator parameter adjustment strategy corresponding to the initial state parameter can be determined, and the optimal ventilator parameter adjustment strategy is the strategy that maximizes the value function

That is:

Optionally, the optimal ventilator parameter adjustment strategy

It can be the strategy that maximizes the value function in K iterations, the strategy that has the value function greater than the preset threshold, the strategy that maximizes the value function within the iteration time, or the strategy that maximizes the value function in the strategy matrix space. The strategy with the largest value function, etc., is not limited in this application.

As a result, the optimal ventilator parameter adjustment strategy corresponding to each state of the user can be determined, so as to train to obtain the ventilator parameter model. This enables subsequent input of the state parameters of the user into the ventilator parameter model to obtain the corresponding optimal ventilator parameter adjustment strategy, and then the ventilator parameter adjustment can be performed according to the optimal ventilator parameter adjustment strategy. This improves the efficiency of ventilator parameter adjustment, improves the reliability of ventilator parameter adjustment, and helps reduce the risk of patient death.

304. Obtain target state parameters of the target user.

Optionally, the description of this step 304 can refer to the related description of step 203 in the embodiment shown in FIG. 2, which will not be repeated here.

That is, the target state parameter of the target user is determined.

305. Input the target state parameter into the ventilator parameter model to obtain a target ventilator parameter adjustment strategy corresponding to the target state parameter.

That is to say, the ventilator parameter acquisition device can input the target state parameter into the ventilator parameter model, call the value function to calculate the reward and punishment information of the target state parameter under various ventilator parameter adjustment strategies, and determine that the target state parameter is in The ventilator parameter adjustment strategy with the largest reward value under the multiple ventilator parameter adjustment strategies, and the ventilator parameter adjustment strategy with the largest reward value is used as the target ventilator parameter adjustment strategy. It is similar to the above-mentioned strategy for determining the optimal ventilator parameter adjustment, and will not be repeated here. Alternatively, the ventilator parameter model may be called to directly determine the optimal ventilator parameter adjustment strategy corresponding to the target state parameter, as the target ventilator parameter adjustment strategy.

In this embodiment of the application, the ventilator parameter acquisition device can determine the reward and punishment function and the state conversion function based on the patient's multi-dimensional state space and strategy matrix space by acquiring multiple sets of state parameters corresponding to multiple users using the ventilator, to Reconstruct the value function of the model, and determine the optimal ventilator parameter adjustment strategy of the patient in the current state based on the value function, thereby improving the efficiency of ventilator parameter setting, and further improving the efficiency of ventilator parameter setting. The reliability of the parameter settings of the ventilator helps to improve the treatment effect and reduce the risk of death of the patient.

It can be understood that the above method embodiments are all examples of the ventilator parameter acquisition method or system of the present application, and the description of each embodiment has its own focus. For parts that are not described in detail in an embodiment, please refer to other embodiments. Related description.

The embodiment of the present application also provides a device for acquiring parameters of a ventilator. The device may include a module for executing the method described in FIG. 2 or FIG. 3a. Please refer to FIG. 4, which is a schematic structural diagram of a ventilator parameter acquisition device provided by an embodiment of the present application. The ventilator parameter obtaining device described in this embodiment may be configured in a ventilator parameter obtaining device. As shown in FIG. 4, the ventilator parameter obtaining device 400 of this embodiment may include: an obtaining module 401 and a processing module 402. in,

The obtaining module 401 can be used to obtain multiple sets of state parameters corresponding to multiple users using ventilators, each set of state parameters includes the user's first state parameters before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and The second state parameter of the user after adopting the ventilator parameter adjustment strategy;

The processing module 402 can be used to determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine the first state parameter in each ventilator parameter adjustment strategy Download the corresponding state transition probability, where the state transition probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter;

The acquiring module 401 may also be used to acquire target state parameters of the target user;

The processing module 402 may also be configured to determine the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability; wherein, The target ventilator parameter adjustment strategy is used to indicate a parameter adjustment strategy for the ventilator of the target user.

In some embodiments, the processing module 402 may also be used to train to obtain a ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability;

Further, when the processing module 402 determines the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state transformation probability, Specifically used for:

The target state parameter is input into the ventilator parameter model to obtain a target ventilator parameter adjustment strategy corresponding to the target state parameter.

In some embodiments, when the processing module 402 obtains the ventilator parameter model by training according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, it may be specifically used to:

Determine the reward and punishment function according to the multiple sets of state parameters and the reward and punishment information corresponding to each ventilator parameter adjustment strategy;

Determine a state transition function according to the multiple sets of state parameters and the state transition probability;

According to the reward and punishment function and the state conversion function, the value function of the ventilator parameter model is determined to obtain the ventilator parameter model through training.

In some embodiments, the processing module 402 determines the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability. When, it can be specifically used for:

Determining one or more sets of state parameters matching the first state parameter and the target state parameter from the plurality of sets of state parameters;

Determine the target corresponding to the target state parameter according to the reward and punishment information corresponding to each ventilator parameter adjustment strategy in the matched one or more sets of state parameters and the state transition probability corresponding to the matched one or more sets of state parameters Ventilator parameter adjustment strategy.

In some embodiments, the first state parameter and the second state parameter are a multi-dimensional matrix state space; when acquiring the target state parameter of the target user, the acquiring module 401 may be specifically used to:

Acquiring diagnosis and treatment data of the target user, where the diagnosis and treatment data includes a plurality of characteristic data;

Respectively transforming the multiple feature data into feature vectors to obtain multiple feature vectors;

Constructing the multi-dimensional matrix state space of the target user according to the multiple feature vectors, and determining the multi-dimensional matrix state space of the target user as the target state parameter.

In some embodiments, the reward and punishment information includes a reward value or a penalty value; the processing module 402 determines the reward and punishment information corresponding to the target state parameter, each ventilator parameter adjustment strategy, and the state conversion probability. The target ventilator parameter adjustment strategy corresponding to the target state parameter can be specifically used for:

Calculating the reward and punishment information of the target state parameter under multiple ventilator parameter adjustment strategies;

Determine the ventilator parameter adjustment strategy with the largest reward value of the target state parameter under the multiple ventilator parameter adjustment strategies, and use the ventilator parameter adjustment strategy with the largest reward value as the target ventilator parameter adjustment strategy.

In some embodiments, the processing module 402 may also be used to verify the target ventilator parameter adjustment strategy after determining the target ventilator parameter adjustment strategy corresponding to the target state parameter; When the parameter adjustment strategy of the target ventilator passes the verification, the ventilator of the target user is adjusted according to the parameter adjustment strategy of the target ventilator; the identification of the target user, the target state parameter, and the target The ventilator parameter adjustment strategy is bound and uploaded to the blockchain.

It is understandable that the functional modules of the ventilator parameter acquisition device in this embodiment can be specifically implemented according to the method in Figure 2 or Figure 3a of the above method embodiment, and the specific implementation process can refer to Figure 2 or Figure 3a of the above method embodiment. The related description of, I won’t repeat it here.

In the embodiment of the present application, the ventilator parameter obtaining device 400 may obtain multiple sets of state parameters corresponding to multiple users using the ventilator through the obtaining module 401, and determine each ventilator parameter adjustment strategy according to the multiple sets of state parameters through the processing module 402 The reward and punishment information and the state transformation probability corresponding to the first state parameter are determined, so that when the target state parameter of the target user is obtained, the reward and punishment information and state transformation corresponding to each ventilator parameter adjustment strategy can be adjusted according to the target state parameter of the target user Probability, and quickly determine the target ventilator parameter adjustment strategy corresponding to the target state parameter, thereby improving the efficiency of ventilator parameter acquisition, thereby improving the efficiency of ventilator parameter setting, and improving the reliability of ventilator parameter setting.

Please refer to FIG. 5, which is a schematic structural diagram of a ventilator parameter obtaining device provided by an embodiment of the present application. As shown in FIG. 5, the ventilator parameter acquisition device may include: a processor 501 and a memory 502. Optionally, the ventilator parameter acquisition device may further include a communication interface 503. The above-mentioned processor 501, memory 502, and communication interface 503 may be connected through a bus or in other ways. In FIG. 5 shown in the embodiment of the present application, the connection through a bus is taken as an example. Wherein, the communication interface 503 can be controlled by the processor to send and receive messages, the memory 502 can be used to store a computer program, the computer program includes program instructions, and the processor 501 is used to execute the program instructions stored in the memory 502. Wherein, the processor 501 is configured to call the program instructions to execute the following steps:

Obtain the target state parameters of the target user;

In some embodiments, the processor 501 may further execute the following steps:

Training to obtain a ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability;

When the processor 501 determines the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, the following steps are specifically performed :

In some embodiments, the processor 501 specifically executes the following steps when training to obtain a ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability:

In some embodiments, the processor 501 determines the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability. When, perform the following steps:

In some embodiments, the first state parameter and the second state parameter are a multi-dimensional matrix state space; the processor 501 specifically executes the following steps when acquiring the target state parameter of the target user:

In some embodiments, the reward and punishment information includes a reward value or a penalty value; the processor 501 determines the reward and punishment information corresponding to the target state parameter, each ventilator parameter adjustment strategy, and the state conversion probability. When the target ventilator parameter adjustment strategy corresponding to the target state parameter, the following steps are specifically executed:

In some embodiments, after determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, the processor 501 further performs the following steps:

Verifying the target ventilator parameter adjustment strategy;

When the verification of the target ventilator parameter adjustment strategy is passed, adjust the parameters of the ventilator of the target user according to the target ventilator parameter adjustment strategy;

The identification of the target user, the target state parameter, and the target ventilator parameter adjustment strategy are bound and uploaded to the blockchain.

In the embodiment of this application, the processor 501 may be a central processing unit (Central Processing Unit, CPU), and the processor 501 may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSPs), and application-specific integrated circuits. (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like.

The memory 502 may include a read-only memory and a random access memory, and provides instructions and data to the processor 501. A part of the memory 502 may also include a non-volatile random access memory. For example, the memory 502 may also store state parameters of the user and so on.

The communication interface 503 may include an input device and/or an output device. For example, the input device may be a control panel, a microphone, a receiver, etc., and the output device may be a display screen, a transmitter, etc., which are not listed here.

In specific implementation, the processor 501, the memory 502, and the communication interface 503 described in the embodiment of the present application can perform the implementation described in the method embodiment described in FIG. 2 or FIG. 3a provided by the embodiment of the present application, and may also perform The implementation of the ventilator parameter acquisition device described in the embodiment of the present application will not be repeated here.

An embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, the computer program includes program instructions, and when the program instructions are executed by a processor, the above-mentioned ventilator can be executed Part or all of the steps performed in the embodiment of the parameter obtaining method, such as part or all of the steps performed by the ventilator parameter obtaining device. Optionally, the computer-readable storage medium may be non-volatile or volatile.

The embodiments of the present application also provide a computer program product. The computer program product includes computer program code. When the computer program code runs on a computer, the computer is caused to execute the above-mentioned ventilator parameter acquisition device method embodiment. step.

In some embodiments, the computer-readable storage medium may mainly include a storage program area and a storage data area, where the storage program area may store an operating system, an application program required by at least one function, etc.; the storage data area may store Data created based on the use of blockchain nodes, etc.

Among them, the blockchain referred to in this application is a new application mode of computer technology such as distributed data storage, point-to-point transmission, consensus mechanism, and encryption algorithm. Blockchain, essentially a decentralized database, is a series of data blocks associated with cryptographic methods. Each data block contains a batch of network transaction information for verification. The validity of the information (anti-counterfeiting) and the generation of the next block. The blockchain can include the underlying platform of the blockchain, the platform product service layer, and the application service layer.

A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing relevant hardware through a computer program. The program can be stored in a computer readable storage medium. During execution, it may include the procedures of the above-mentioned method embodiments. Wherein, the storage medium may be a magnetic disk, an optical disc, a read-only memory (Read-Only Memory, ROM), or a random access memory (Random Access Memory, RAM), etc.

What is disclosed above is only a preferred embodiment of this application. Of course, it cannot be used to limit the scope of rights of this application. A person of ordinary skill in the art can understand all or part of the process of implementing the above-mentioned embodiments and follow the rights of this application. The equivalent changes required are still within the scope of the invention.

Claims

A method for obtaining parameters of a ventilator includes:

Acquire multiple sets of state parameters corresponding to multiple users using the ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter The second state parameter of the user after the policy is adjusted;

Determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine the corresponding state transition probability of the first state parameter under each ventilator parameter adjustment strategy , The state conversion probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter;

Obtain the target state parameters of the target user;

According to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, the target ventilator parameter adjustment strategy corresponding to the target state parameter is determined; wherein, the target ventilator parameter adjustment strategy is used To indicate a parameter adjustment strategy for the ventilator of the target user.
The method according to claim 1, wherein the method further comprises:

Training to obtain a ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability;

The determining the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability includes:

The target state parameter is input into the ventilator parameter model to obtain a target ventilator parameter adjustment strategy corresponding to the target state parameter.
The method according to claim 2, wherein the training to obtain a ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability comprises:

Determine the reward and punishment function according to the multiple sets of state parameters and the reward and punishment information corresponding to each ventilator parameter adjustment strategy;

Determine a state transition function according to the multiple sets of state parameters and the state transition probability;

According to the reward and punishment function and the state conversion function, the value function of the ventilator parameter model is determined to obtain the ventilator parameter model through training.
The method of claim 1, wherein the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability are used to determine the target ventilator parameter adjustment corresponding to the target state parameter Strategies, including:

Determining one or more sets of state parameters matching the first state parameter and the target state parameter from the plurality of sets of state parameters;

Determine the target corresponding to the target state parameter according to the reward and punishment information corresponding to each ventilator parameter adjustment strategy in the matched one or more sets of state parameters and the state transition probability corresponding to the matched one or more sets of state parameters Ventilator parameter adjustment strategy.
The method according to any one of claims 1 to 4, wherein the first state parameter and the second state parameter are a multi-dimensional matrix state space; the obtaining the target state parameter of the target user comprises:

Acquiring diagnosis and treatment data of the target user, where the diagnosis and treatment data includes a plurality of characteristic data;

Respectively transforming the multiple feature data into feature vectors to obtain multiple feature vectors;

Constructing the multi-dimensional matrix state space of the target user according to the multiple feature vectors, and determining the multi-dimensional matrix state space of the target user as the target state parameter.
The method according to any one of claims 1 to 4, wherein the reward and punishment information includes a reward value or a penalty value; the reward and punishment information corresponding to the target state parameter, each ventilator parameter adjustment strategy, and the state The conversion probability, determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, includes:

Calculating the reward and punishment information of the target state parameter under multiple ventilator parameter adjustment strategies;

Determine the ventilator parameter adjustment strategy with the largest reward value of the target state parameter under the multiple ventilator parameter adjustment strategies, and use the ventilator parameter adjustment strategy with the largest reward value as the target ventilator parameter adjustment strategy.
The method according to any one of claims 1 to 4, wherein after the determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, the method further comprises:

Verifying the target ventilator parameter adjustment strategy;

When the verification of the target ventilator parameter adjustment strategy is passed, adjust the parameters of the ventilator of the target user according to the target ventilator parameter adjustment strategy;

The identification of the target user, the target state parameter, and the target ventilator parameter adjustment strategy are bound and uploaded to the blockchain.
A device for acquiring parameters of a ventilator includes:

The acquiring module is used to acquire multiple sets of state parameters corresponding to multiple users who use the ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter adjustment strategy. The second state parameter of the user after the ventilator parameter adjustment strategy;

The determining module is configured to determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine that the first state parameter is under each ventilator parameter adjustment strategy A corresponding state transition probability, where the state transition probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter;

The acquiring module is also used to acquire target state parameters of the target user;

The determining module is further configured to determine the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability; The target ventilator parameter adjustment strategy is used to indicate a parameter adjustment strategy for the ventilator of the target user.
A ventilator parameter acquisition device includes a processor and a memory, the processor and the memory are connected to each other, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured Used to call the program instructions to execute the following methods:

Acquire multiple sets of state parameters corresponding to multiple users using the ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter The second state parameter of the user after the policy is adjusted;

Determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine the corresponding state transition probability of the first state parameter under each ventilator parameter adjustment strategy , The state conversion probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter;

Obtain the target state parameters of the target user;

According to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, the target ventilator parameter adjustment strategy corresponding to the target state parameter is determined; wherein, the target ventilator parameter adjustment strategy is used To indicate a parameter adjustment strategy for the ventilator of the target user.
The device according to claim 9, wherein the processor is further configured to execute:

Training to obtain a ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability;

When determining the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, the following is specifically executed:

The target state parameter is input into the ventilator parameter model to obtain a target ventilator parameter adjustment strategy corresponding to the target state parameter.
The device according to claim 9, wherein the target state parameter is determined according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, and the target ventilator parameter adjustment corresponding to the target state parameter During the strategy, the specific implementation:

Determining one or more sets of state parameters matching the first state parameter and the target state parameter from the plurality of sets of state parameters;

Determine the target corresponding to the target state parameter according to the reward and punishment information corresponding to each ventilator parameter adjustment strategy in the matched one or more sets of state parameters and the state transition probability corresponding to the matched one or more sets of state parameters Ventilator parameter adjustment strategy.
The device according to any one of claims 9-11, wherein the first state parameter and the second state parameter are a multi-dimensional matrix state space; when the target state parameter of the target user is obtained, the following is specifically executed:

Acquiring diagnosis and treatment data of the target user, where the diagnosis and treatment data includes a plurality of characteristic data;

Respectively transforming the multiple feature data into feature vectors to obtain multiple feature vectors;

Constructing the multi-dimensional matrix state space of the target user according to the multiple feature vectors, and determining the multi-dimensional matrix state space of the target user as the target state parameter.
The device according to any one of claims 9-11, wherein the reward and punishment information includes a reward value or a penalty value; the reward and punishment information corresponding to the target state parameter, each ventilator parameter adjustment strategy, and the state Conversion probability, when determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, specifically execute:

Calculating the reward and punishment information of the target state parameter under multiple ventilator parameter adjustment strategies;

Determine the ventilator parameter adjustment strategy with the largest reward value of the target state parameter under the multiple ventilator parameter adjustment strategies, and use the ventilator parameter adjustment strategy with the largest reward value as the target ventilator parameter adjustment strategy.
The device according to any one of claims 9-11, wherein, after the determination of the target ventilator parameter adjustment strategy corresponding to the target state parameter, the processor is further configured to execute:

Verifying the target ventilator parameter adjustment strategy;

When the verification of the target ventilator parameter adjustment strategy is passed, adjust the parameters of the ventilator of the target user according to the target ventilator parameter adjustment strategy;

The identification of the target user, the target state parameter, and the target ventilator parameter adjustment strategy are bound and uploaded to the blockchain.
A computer-readable storage medium, the computer-readable storage medium stores a computer program, and the computer program includes program instructions that, when executed by a processor, cause the processor to perform the following method:

Acquire multiple sets of state parameters corresponding to multiple users using the ventilator, each set of state parameters includes the user's first state parameter before adopting the ventilator parameter adjustment strategy, the adopted ventilator parameter adjustment strategy, and the adopted ventilator parameter The second state parameter of the user after the policy is adjusted;

Determine the reward and punishment information of each ventilator parameter adjustment strategy under the corresponding first state parameter and the second state parameter according to the multiple sets of state parameters, and determine the corresponding state transition probability of the first state parameter under each ventilator parameter adjustment strategy , The state conversion probability is used to indicate the probability that the user obtains the second state parameter after adopting the ventilator parameter adjustment strategy under the first state parameter;

Obtain the target state parameters of the target user;

According to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, the target ventilator parameter adjustment strategy corresponding to the target state parameter is determined; wherein, the target ventilator parameter adjustment strategy is used To indicate a parameter adjustment strategy for the ventilator of the target user.
The computer-readable storage medium according to claim 15, wherein the program instructions, when executed by the processor, are also used to cause the processor to execute:

Training to obtain a ventilator parameter model according to the multiple sets of state parameters, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability;

When determining the target ventilator parameter adjustment strategy corresponding to the target state parameter according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability, the following is specifically executed:

The target state parameter is input into the ventilator parameter model to obtain a target ventilator parameter adjustment strategy corresponding to the target state parameter.
The computer-readable storage medium according to claim 15, wherein the target state parameter corresponding to the target state parameter is determined according to the target state parameter, the reward and punishment information corresponding to each ventilator parameter adjustment strategy, and the state conversion probability The specific implementation of the ventilator parameter adjustment strategy:

Determining one or more sets of state parameters matching the first state parameter and the target state parameter from the plurality of sets of state parameters;

Determine the target corresponding to the target state parameter according to the reward and punishment information corresponding to each ventilator parameter adjustment strategy in the matched one or more sets of state parameters and the state transition probability corresponding to the matched one or more sets of state parameters Ventilator parameter adjustment strategy.
The computer-readable storage medium according to any one of claims 15-17, wherein the first state parameter and the second state parameter are a multi-dimensional matrix state space; when the target state parameter of the target user is acquired, Specific implementation:

Acquiring diagnosis and treatment data of the target user, where the diagnosis and treatment data includes a plurality of characteristic data;

Respectively transforming the multiple feature data into feature vectors to obtain multiple feature vectors;

Constructing the multi-dimensional matrix state space of the target user according to the multiple feature vectors, and determining the multi-dimensional matrix state space of the target user as the target state parameter.
The computer-readable storage medium according to any one of claims 15-17, wherein the reward and punishment information includes a reward value or a punishment value; the reward and punishment information corresponding to the adjustment strategy according to the target state parameter and each ventilator parameter , The state conversion probability, when determining the target ventilator parameter adjustment strategy corresponding to the target state parameter, specifically execute:

Calculating the reward and punishment information of the target state parameter under multiple ventilator parameter adjustment strategies;

Determine the ventilator parameter adjustment strategy with the largest reward value of the target state parameter under the multiple ventilator parameter adjustment strategies, and use the ventilator parameter adjustment strategy with the largest reward value as the target ventilator parameter adjustment strategy.
The computer-readable storage medium according to any one of claims 15-17, wherein, after the determination of the target ventilator parameter adjustment strategy corresponding to the target state parameter, the program instructions, when executed by the processor, also Used to make the processor execute:

Verifying the target ventilator parameter adjustment strategy;

When the verification of the target ventilator parameter adjustment strategy is passed, adjust the parameters of the ventilator of the target user according to the target ventilator parameter adjustment strategy;

The identification of the target user, the target state parameter, and the target ventilator parameter adjustment strategy are bound and uploaded to the blockchain.