Method and system for evaluating influence of anesthetic on learning and memory ability

By constructing causal relationship networks and conditional probability models, the ambiguity in evaluating the effect of anesthetic drugs on learning and memory abilities is resolved, providing a precise evaluation method and system.

CN118888166AActive Publication Date: 2024-11-01THE FIRST AFFILIATED HOSPITAL OF SOOCHOW UNIV
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410902745.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-05
Publication Date
2024-11-01
Estimated Expiration
2044-07-05

AI Technical Summary

Technical Problem

Existing technologies for evaluating the effects of anesthetic drugs on learning and memory abilities rely on regression models of cell apoptosis, activation of important proteins and receptors, resulting in vague and inaccurate results that cannot accurately characterize the specific values ​​of learning and memory abilities.

Method used

By constructing a causal relationship network between anesthetic drugs and memory-affecting indicators, calculating conditional probabilities, using a node-probability regression model to fill in missing values, constructing direct, indirect, and instrumental variable models, and calculating the final learning and memory ability.

Benefits of technology

It enables precise evaluation of the impact of anesthetic drugs on learning and memory abilities, replacing regression models and providing clear evaluation results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118888166A_ABST
    Figure CN118888166A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of Bayesian networks, and discloses a method and system for evaluating the influence of narcotics on learning and memory ability, and the method comprises the steps: generating a memory influence index of the narcotics in an experimental group through employing an experimental group sample and a control group sample; the method comprises the following steps: identifying a causal association network, calculating a conditional probability between anesthetics and a memory influence index in the causal association network, constructing a node-probability regression model between causal nodes and the conditional probability in the causal association network, and constructing a direct influence model between the memory influence index and the learning memory ability of a learning memory part. Constructing an indirect influence model between the memory influence index and the learning memory ability, and constructing a tool variable model between the memory influence index and the learning memory ability; and calculating the final learning memory ability of the learning memory part by using the direct influence model, the indirect influence model and the tool variable model. The method can replace a regression model to evaluate the influence of the anesthetic on the learning and memory ability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of Bayesian network technology, and in particular to a method and system for evaluating the influence of anesthetics on learning and memory abilities. Background Art

[0002] The process of evaluating the effect of anesthetics on learning and memory abilities refers to the process of evaluating the degree of influence of different doses of anesthetics on learning and memory abilities. The current evaluation process is based on measuring biological components related to learning and memory.

[0003] At present, before evaluating the effect of anesthetics on learning and memory ability, the existing technology needs to first measure the number of learning and memory-related cell apoptosis, the increase and decrease of important proteins and the activation of receptors in mice after the use of anesthetics, and then use the number of cell apoptosis, the increase and decrease of important proteins and the activation of receptors to judge the learning and memory ability by artificial experts. The results of the judgment are often vague results such as mild weakening of learning and memory ability, moderate weakening of learning and memory ability and severe weakening of learning and memory ability. If the learning and memory ability is to be characterized by specific numerical values, it is either characterized by parameters such as the number of cell apoptosis, the increase and decrease of important proteins and the activation of receptors related to learning and memory, or by a regression model. For the latter, the independent variables in the regression model are parameters such as the number of cell apoptosis, the increase and decrease of important proteins and the activation of receptors related to learning and memory, and the dependent variable is the learning and memory ability. However, in actual scenarios, cell apoptosis, important proteins and receptors are not curvilinearly correlated with learning and memory ability. The effects of different types of anesthetics on cell apoptosis, important proteins and receptors are probabilistically triggered. By analogy, the effects of changes in substances such as cell apoptosis, important proteins and receptors on learning and memory ability are also probabilistically triggered. Therefore, there is an urgent need for a solution that can replace the regression model to evaluate the effects of anesthetics on learning and memory abilities. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a method and system for evaluating the effect of anesthetics on learning and memory abilities, which can replace the regression model to evaluate the effect of anesthetics on learning and memory abilities.

[0005] In a first aspect, the present invention provides a method for evaluating the effect of anesthetics on learning and memory ability, comprising:

[0006] Inquiring the learning and memory sites where the anesthetic acts in the experimental group, collecting the experimental group samples of the learning and memory sites in the experimental group, and sampling the control group samples of the sites corresponding to the learning and memory sites in the control group;

[0007] Using the experimental group samples and the control group samples to generate memory impact indexes of the anesthetic in the experimental group, wherein the memory impact indexes include a structural deformation index, a quantity change index, and a receptor activation index;

[0008] Identify the causal association network between the anesthetic and the memory-affecting index, calculate the conditional probability between the anesthetic and the memory-affecting index in the causal association network, construct a node-probability regression model between the causal nodes in the causal association network and the conditional probability, and complete the conditional probability according to the node-probability regression model to obtain a completed conditional probability;

[0009] Using the complementary conditional probability to construct a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part, using the complementary conditional probability to construct an indirect influence model between the memory influence index and the learning and memory ability, and using the complementary conditional probability to construct an instrumental variable model between the memory influence index and the learning and memory ability;

[0010] The direct influence model, the indirect influence model and the instrumental variable model are used to calculate the final learning and memory ability of the learning and memory part, and the final learning and memory ability is used as the evaluation result of the effect of the anesthetic on the learning and memory ability.

[0011] In a possible implementation of the first aspect, generating the memory impact index of the anesthetic in the experimental group by using the experimental group samples and the control group samples includes:

[0012] The sample difference between the experimental group samples and the control group samples is used as the memory impact index.

[0013] In a possible implementation of the first aspect, the identifying a causal relationship network between the anesthetic and the memory impact indicator includes:

[0014] Querying the starting time of the anesthetic and the memory-affecting index;

[0015] Arrange the anesthetic drugs and the memory-affecting indexes in the time sequence of the starting time to obtain a drug-index sequence;

[0016] A causal association network between the anesthetic drug and the memory-affecting indicator is determined from the drug-indicator sequence.

[0017] In a possible implementation manner of the first aspect, constructing a node-probability regression model between a causal node in the causal association network and the conditional probability includes:

[0018] Acquire a first correlation network, a second correlation network, and a third correlation network in the causal correlation network;

[0019] Acquire all parent nodes of the first child node in the first association network, acquire all child nodes of the second parent node in the second association network, and acquire the historical parent node of the third child node in the third association network;

[0020] Convert the first child node, all the parent nodes, the second parent node, all the child nodes, the third child node, and the historical parent node into a parent-child node pair;

[0021] Based on the parent-child node pairing, a node-probability regression model between the causal node and the conditional probability is constructed.

[0022] In a possible implementation of the first aspect, the using the completion conditional probability to construct a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part includes:

[0023] Based on the completion conditional probability, a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part is constructed.

[0024] In a possible implementation of the first aspect, the using the completion conditional probability to construct an indirect influence model between the memory influence index and the learning and memory ability includes:

[0025] Based on the completion conditional probability, calculating an indirect variable between the memory impact index and the learning and memory ability;

[0026] According to the completion conditional probability and the indirect variable, an indirect influence model between the memory influence index and the learning and memory ability is constructed.

[0027] In a possible implementation of the first aspect, the constructing an instrumental variable model between the memory impact index and the learning and memory ability by using the completed conditional probability includes:

[0028] Calculating the memory impact index under the instrumental variable between the memory impact index and the learning and memory ability according to the completion conditional probability;

[0029] According to the memory influence index under the instrumental variable and the completion conditional probability, an instrumental variable model between the memory influence index and the learning and memory ability is constructed.

[0030] In a possible implementation of the first aspect, the calculating the final learning and memory capacity of the learning and memory part by using the direct influence model, the indirect influence model and the instrumental variable model includes:

[0031] The final learning and memory capacity of the learning and memory part is calculated according to the direct influence model, the indirect influence model and the instrumental variable model.

[0032] In a second aspect, the present invention provides an evaluation system for the effect of anesthetics on learning and memory ability, the system comprising:

[0033] A sample collection module, used to query the learning and memory sites where the anesthetic acts in the experimental group, collect the experimental group samples of the learning and memory sites in the experimental group, and sample the control group samples of the sites corresponding to the learning and memory sites in the control group;

[0034] An index generation module, used to generate a memory impact index of the anesthetic in the experimental group using the experimental group samples and the control group samples, wherein the memory impact index includes a structural deformation index, a quantity change index and a receptor activation index;

[0035] A probability completion module is used to identify the causal association network between the anesthetic and the memory-affecting index, calculate the conditional probability between the anesthetic and the memory-affecting index in the causal association network, construct a node-probability regression model between the causal node in the causal association network and the conditional probability, and complete the conditional probability according to the node-probability regression model to obtain a completed conditional probability;

[0036] A model building module, used to use the complementary conditional probability to build a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part, use the complementary conditional probability to build an indirect influence model between the memory influence index and the learning and memory ability, and use the complementary conditional probability to build an instrumental variable model between the memory influence index and the learning and memory ability;

[0037] The ability evaluation module is used to calculate the final learning and memory ability of the learning and memory part by using the direct influence model, the indirect influence model and the instrumental variable model, and use the final learning and memory ability as the evaluation result of the effect of the anesthetic on the learning and memory ability.

[0038] Compared with the prior art, the technical principle and beneficial effects of this solution are:

[0039] The embodiment of the present invention generates the memory influence index of the anesthetic in the experimental group by using the experimental group samples and the control group samples, so as to observe the fluctuation of the numerical value under the influence of the anesthetic compared with the numerical value under the normal state without the anesthetic. The embodiment of the present invention identifies the causal association network between the anesthetic and the memory influence index to construct a directed acyclic graph between the anesthetic and the memory influence index, so as to pave the way for the subsequent calculation of Bayesian conditional probability. Furthermore, the embodiment of the present invention calculates the conditional probability between the anesthetic and the memory influence index in the causal association network, so as to replace the regression model with the conditional probability. In one step, the embodiment of the present invention uses a node-probability regression model between the causal node and the conditional probability in the causal association network, so that when the change value of the causal node, that is, the value of the index value of the anesthetic is not consistent with the index value of the memory impact index, that is, when a node missing value appears, the node missing value is completed by the node-probability regression model, and then the probability missing value corresponding to the missing value is completed. The embodiment of the present invention constructs a direct influence model between the memory impact index and the learning and memory ability of the learning and memory part by using the completed conditional probability, so as to analyze the direct relationship between the memory impact index and the learning and memory ability under the influence of the completed conditional probability. Therefore, the evaluation method and system of the effect of anesthetics on learning and memory ability proposed in the embodiment of the present invention can replace the regression model to realize the evaluation of the effect of anesthetics on learning and memory ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0042] Figure 1 A schematic flow chart of a method for evaluating the effect of anesthetics on learning and memory abilities provided by one embodiment of the present invention;

[0043] Figure 2 A schematic diagram of a module of a system for evaluating the effect of anesthetics on learning and memory abilities provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0044] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0045] The embodiment of the present invention provides an evaluation method for the effect of anesthetics on learning and memory ability, and the execution subject of the evaluation method for the effect of anesthetics on learning and memory ability includes but is not limited to at least one of the electronic devices that can be configured to perform the method provided by the embodiment of the present invention, such as a server, a terminal, etc. In other words, the evaluation method for the effect of anesthetics on learning and memory ability can be executed by software or hardware installed in a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0046] See also Figure 1 FIG. 1 is a flow chart of a method for evaluating the effect of anesthetics on learning and memory abilities provided by an embodiment of the present invention. Figure 1 The methods described in the evaluation of the effects of anesthetics on learning and memory include:

[0047] S1. Query the learning and memory sites where anesthetics act in the experimental group, collect experimental group samples of the learning and memory sites in the experimental group, and sample control group samples of sites in the control group corresponding to the learning and memory sites.

[0048] In the embodiments of the present invention, the anesthetic refers to an anesthetic drug that affects learning and memory abilities, and the learning and memory sites refer to biological components in the body related to learning and memory abilities, including cells, receptors, proteins, and the like.

[0049] Furthermore, the experimental group refers to a group in which mice are anesthetized with the participation of anesthetics, the control group refers to a group in which numerical changes in the mice's bodies are observed without the participation of anesthetics, the experimental group samples refer to the numerical changes in the learning and memory areas of the mice in the experimental group, and the control group samples refer to the numerical changes in the learning and memory areas of the mice in the control group.

[0050] S2. Generate memory impact indexes of the anesthetic in the experimental group using the experimental group samples and the control group samples, wherein the memory impact indexes include a structural deformation index, a quantity change index, and a receptor activation index.

[0051] The embodiment of the present invention generates the memory impact index of the anesthetic in the experimental group by using the experimental group samples and the control group samples, so as to observe the fluctuation of the value under the influence of the anesthetic compared with the normal value without the anesthetic.

[0052] In one embodiment of the present invention, the use of the experimental group samples and the control group samples to generate the memory impact index of the anesthetic in the experimental group includes: using the sample difference between the experimental group samples and the control group samples as the memory impact index.

[0053] Among them, the structural deformation index includes values ​​such as length, width, volume and area, the quantity change index includes values ​​such as quantity decrease and quantity increase, and the receptor activation index includes values ​​such as activated receptor type and activated receptor number.

[0054] S3. Identify the causal association network between the anesthetic and the memory-affecting index, calculate the conditional probability between the anesthetic and the memory-affecting index in the causal association network, construct a node-probability regression model between the causal nodes in the causal association network and the conditional probability, and complete the conditional probability according to the node-probability regression model to obtain the completed conditional probability.

[0055] The embodiment of the present invention identifies the causal association network between the anesthetic and the memory-affecting index to construct a directed acyclic graph between the anesthetic and the memory-affecting index, thereby paving the way for the subsequent calculation of Bayesian conditional probability.

[0056] The causal association network refers to a directed acyclic graph consisting of the change value of the anesthetic and the memory impact index. In the directed acyclic graph, each node represents the index value of the change value of the anesthetic and the memory impact index.

[0057] In one embodiment of the present invention, the identifying of the causal association network between the anesthetic and the memory affecting indicator includes: querying the starting time of the anesthetic and the memory affecting indicator; sorting the anesthetic and the memory affecting indicator in chronological order of the starting time to obtain a drug-indicator sequence; and determining the causal association network between the anesthetic and the memory affecting indicator from the drug-indicator sequence.

[0058] Optionally, the process of determining the causal association network between the anesthetic and the memory-affecting index from the drug-index sequence refers to taking the first-appearing value as the parent node of the later-appearing value.

[0059] Furthermore, the embodiment of the present invention calculates the conditional probability between the anesthetic and the memory-affecting index in the causal association network to replace the regression model with the conditional probability.

[0060] In one embodiment of the present invention, the calculating of the conditional probability between the anesthetic and the memory impact index in the causal association network includes: calculating the marginal probability between the anesthetic and the memory impact index using the following formula:

[0061]

[0062] Among them, P(A l ) represents the marginal probability, K represents the total number of experimental groups, and n l represents the number of occurrences of the lth memory impact indicator in the K experimental groups, A l Indicates the occurrence of the event of the lth node;

[0063] Distinguish a first association network, a second association network, and a third association network in the causal association network; and calculate a first conditional probability in the first association network based on the marginal probability using the following formula:

[0064] P(a,b,c)=P(a)*P(b)*P(c∣a,b)

[0065]

[0066] Among them, P(c|a,b) represents the first conditional probability, P(a,b,c) represents the joint probability of events a, b, and c occurring simultaneously, P(a) represents the marginal probability of event a, and P(b) represents the marginal probability of event b;

[0067] Based on the edge probability, the second conditional probability in the second association network is calculated using the following formula:

[0068]

[0069] Wherein, P(a,b|c) represents the second conditional probability, P(c) represents the marginal probability of event c, P(a|c) represents the conditional probability that the parent node in the second association network is event c and the child node is event a, and P(b|c) represents the conditional probability that the parent node in the second association network is event c and the child node is event b;

[0070] Based on the edge probability, the third conditional probability in the third association network is calculated using the following formula:

[0071]

[0072] Among them, P(b|a,c) represents the third conditional probability, P(a,b,c) represents the joint probability of events a, b, and c occurring simultaneously, P(a) represents the marginal probability of event a, and P(c) represents the marginal probability of event c;

[0073] The first conditional probability, the second conditional probability and the third conditional probability are used to determine the conditional probability between the anesthetic and the memory impact indicator in the causal association network.

[0074] The first association network refers to a network in which multiple parent nodes point to the same child node, the second association network refers to a network in which the same parent node points to multiple child nodes, and the third association network refers to a network consisting of a chain structure between nodes.

[0075] Optionally, the process of using the first conditional probability, the second conditional probability and the third conditional probability to determine the conditional probability between the anesthetic and the memory influencing index in the causal association network means that when it is necessary to calculate the conditional probability between a parent node and a child node, if the network structure between the parent node and the child node conforms to any of the network structures in the above-mentioned first association network, the second association network and the third association network, the conditional probability between the parent node and the child node is calculated using the calculation method of the conditional probability corresponding to this network structure.

[0076] Furthermore, the embodiment of the present invention uses a node-probability regression model between the causal node and the conditional probability in the causal association network to complete the missing value of the node through the node-probability regression model when the change value of the anesthetic and the index value of the memory impact index are not consistent, that is, when there is a missing value of the node. Then, the missing value of the node is completed by the node-probability regression model, and the missing value of the probability corresponding to the missing value is completed.

[0077] In one embodiment of the present invention, the construction of a node-probability regression model between a causal node in the causal association network and the conditional probability includes: obtaining a first association network, a second association network, and a third association network in the causal association network; obtaining all parent nodes of a first child node in the first association network, obtaining all child nodes of a second parent node in the second association network, and obtaining a historical parent node of a third child node in the third association network; converting the first child node, all parent nodes, the second parent node, all child nodes, the third child node, and the historical parent node into a parent-child node pairing; based on the parent-child node pairing, constructing a node-probability regression model between the causal node and the conditional probability using the following formula:

[0078]

[0079] in, represents the node-probability regression model, P represents the conditional probability between the parent node and the child node in the parent-child node pairing, b represents the bias factor, represents the weight factor, represents the parent node of the d1th parent-child node pairing in the kth experimental group, D1 represents the total number of parent-child node pairs, Represents the child node of the d1th parent-child node pairing in the kth experimental group.

[0080] Among them, the all parent nodes refer to multiple parent nodes pointing to the same child node in the first association network, the all child nodes refer to multiple child nodes with the same parent node in the second association network, the historical parent node refers to all nodes on the previous chain of the third child node in the chain structure of the third association network, and the parent-child node pairing refers to the three parent-child node pairings of the first child node and all parent nodes, the second parent node and all child nodes, and the third child node and the historical parent node.

[0081] Optionally, the process of completing the conditional probability according to the node-probability regression model to obtain the completed conditional probability refers to interpolating missing values ​​of node values ​​in the node-probability regression model, and then using the node-probability regression model to calculate the conditional probability corresponding to the interpolated missing values.

[0082] S4. Use the completed conditional probability to construct a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part, use the completed conditional probability to construct an indirect influence model between the memory influence index and the learning and memory ability, and use the completed conditional probability to construct an instrumental variable model between the memory influence index and the learning and memory ability.

[0083] The embodiment of the present invention constructs a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part by utilizing the completion conditional probability, so as to analyze the direct relationship between the memory influence index and the learning and memory ability under the influence of the completion conditional probability.

[0084] Among them, the learning and memory ability refers to the ability of complex high-level brain activities formed by the interaction and mutual connection between various neurons and synapses, which can be obtained by testing the maze-solving ability of mice. The direct impact model refers to the explicit direct relationship between the biological tissues in the mouse body and the learning and memory ability.

[0085] In one embodiment of the present invention, the use of the complementary conditional probability to construct a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part includes: based on the complementary conditional probability, using the following formula to construct a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part:

[0086] Q it1 =e1+e2P'x it +∑f1m it+α i +β t +γ it

[0087] Among them, Q it1 =e1+e2x it +∑f1m it +α i +β t +γ it represents the direct impact model, i represents the serial number of the memory impact index, t represents the time, Q it1 It directly affects the learning and memory ability of the model, x it Represents the sub-memory influence index corresponding to the completion conditional probability, m it represents the control variable, e1 represents the intercept term, e2 represents the regression coefficient of the memory impact index, f1 represents the regression coefficient of the control variable, α j Represents x it The corresponding fixed effect of learning and memory location, β t represents the time fixed effect, γ jt represents the random error parameter, and P' represents the completion conditional probability.

[0088] Among them, the control variable refers to the variables of other biological tissues that may have an impact that need to be controlled when controlling the relationship between the memory impact index under study and the learning and memory ability, including structural changes and quantity changes in other parts of the body near the learning and memory site, and the sub-memory impact index refers to the sub-node.

[0089] In one embodiment of the present invention, the indirect influence model between the memory influence index and the learning and memory ability is constructed by using the complementary conditional probability, including: based on the complementary conditional probability, the indirect variable between the memory influence index and the learning and memory ability is calculated by using the following formula:

[0090] I it =g1+h1P'x it +∑f1m it +α i +β t +γ it

[0091] Among them, I it represents the indirect variable, g1 represents the intercept term, h1 represents the regression coefficient of the memory impact index, x it represents the sub-memory influence index corresponding to the completion conditional probability, P' represents the completion conditional probability, m it represents the control variable, f1 represents the regression coefficient of the control variable, α j Represents x itThe corresponding fixed effect of learning and memory location, β t represents the time fixed effect, γ jt represents the random error parameter;

[0092] According to the completion conditional probability and the indirect variable, the indirect influence model between the memory influence index and the learning and memory ability is constructed using the following formula:

[0093] Q it2 =e3+e4P'x it +qI it +∑f1m it +α i +β t +γ it

[0094] Among them, Q it2 =e3+e4P'x it +qI it +∑f1m it +α i +β t +γ it represents the indirect effect model, e3 represents the intercept term, e4 represents the regression coefficient of the memory effect index, q represents the regression coefficient of the indirect variable, I it Indicates an indirect variable, x it represents the sub-memory influence index corresponding to the completion conditional probability, P' represents the completion conditional probability, m it represents the control variable, f1 represents the regression coefficient of the control variable, α j Represents x it The corresponding fixed effect of learning and memory location, β t represents the time fixed effect, γ jt represents the random error parameter, Q it2 Represents the learning and memory ability under the indirect influence model.

[0095] The fixed effect is essentially a control variable, which indicates the factors that remain unchanged under control, or the factors that change with control, and the indirect variable refers to the mediator that affects the dependent variable through the independent variable.

[0096] In one embodiment of the present invention, the use of the completed conditional probability to construct an instrumental variable model between the memory impact index and the learning and memory ability includes: according to the completed conditional probability, calculating the memory impact index under the instrumental variable between the memory impact index and the learning and memory ability using the following formula:

[0097] x it '=e5+e6P'X it +∑f1m it+α i +β t +γ it

[0098] Among them, x it ' represents the memory impact index under the instrumental variable, e5 represents the intercept term, e6 represents the regression coefficient of the instrumental variable, X it represents the instrumental variable, P' represents the complete conditional probability, m it represents the control variable, f1 represents the regression coefficient of the control variable, α j Represents x it The corresponding fixed effect of learning and memory location, β t represents the time fixed effect, γ jt represents the random error parameter;

[0099] According to the memory influence index under the instrumental variable and the completion conditional probability, the instrumental variable model between the memory influence index and the learning and memory ability is constructed using the following formula:

[0100] Q it3 =e7+e8P'x it '+∑f1m it +α i +β t +γ it

[0101] Among them, Q it3 =e7+e8x it '+∑f1m it +α i +β t +γ it represents the instrumental variable model, x it ' represents the memory impact index under the instrumental variable, e7 represents the intercept term, e8 represents the regression coefficient of the memory impact index under the instrumental variable, P' represents the completion conditional probability, m it represents the control variable, f1 represents the regression coefficient of the control variable, α j Represents x it The corresponding fixed effect of learning and memory location, β t represents the time fixed effect, γ jt represents the random error parameter, Q it3 Represents learning and memory ability under the instrumental variable model.

[0102] The instrumental variable refers to a variable used to estimate the causal relationship of the model when a controlled experiment cannot be implemented.

[0103] S5. Calculate the final learning and memory capacity of the learning and memory part using the direct influence model, the indirect influence model and the instrumental variable model, and use the final learning and memory capacity as the evaluation result of the effect of the anesthetic on the learning and memory capacity.

[0104] In one embodiment of the present invention, the use of the direct influence model, the indirect influence model and the instrumental variable model to calculate the final learning and memory ability of the learning and memory part includes: according to the direct influence model, the indirect influence model and the instrumental variable model, the final learning and memory ability of the learning and memory part is calculated using the following formula:

[0105] Q=Q it1 +Q it2 +Q it3

[0106] Among them, Q represents the final learning and memory ability, Q it1 It directly affects the learning and memory ability of the model, Q it2 represents the learning and memory ability under the indirect influence model, Q it3 Represents learning and memory ability under the instrumental variable model.

[0107] It can be seen that the embodiment of the present invention generates the memory influence index of the anesthetic in the experimental group by using the experimental group samples and the control group samples, so as to observe the fluctuation of the numerical value under the influence of the anesthetic compared with the normal numerical value without anesthetic. The embodiment of the present invention identifies the causal association network between the anesthetic and the memory influence index to construct a directed acyclic graph between the anesthetic and the memory influence index, so as to pave the way for the subsequent calculation of Bayesian conditional probability. Furthermore, the embodiment of the present invention calculates the conditional probability between the anesthetic and the memory influence index in the causal association network, so as to use the conditional probability instead of the regression model. Furthermore, the embodiment of the present invention uses a node-probability regression model between the causal node and the conditional probability in the causal association network to complete the missing node value through the node-probability regression model when the change value of the causal node, that is, the value of the index value of the memory influencing index is not consistent, that is, when a node missing value occurs, and then completes the probability missing value corresponding to the missing value. The embodiment of the present invention constructs a direct influence model between the memory influencing index and the learning and memory ability of the learning and memory part by using the completed conditional probability, so as to analyze the direct relationship between the memory influencing index and the learning and memory ability under the influence of the completed conditional probability. Therefore, the evaluation method and system for the effect of anesthetics on learning and memory ability proposed in the embodiment of the present invention can replace the regression model to realize the evaluation of the effect of anesthetics on learning and memory ability.

[0108] like Figure 2 The figure shows a functional module diagram of an evaluation system for the effect of an anesthetic on learning and memory abilities according to the present invention.

[0109] The evaluation system 200 for the effect of anesthetics on learning and memory ability of the present invention can be installed in an electronic device. According to the functions to be implemented, the evaluation system for the effect of anesthetics on learning and memory ability can include a sample collection module 201, an indicator generation module 202, a probability completion module 203, a model construction module 204 and an ability evaluation module 205. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0110] In the embodiment of the present invention, the functions of each module / unit are as follows:

[0111] The sample collection module 201 is used to query the learning and memory sites where the anesthetic acts in the experimental group, collect the experimental group samples of the learning and memory sites in the experimental group, and sample the control group samples of the sites corresponding to the learning and memory sites in the control group;

[0112] The index generation module 202 is used to generate the memory impact index of the anesthetic in the experimental group using the experimental group samples and the control group samples, wherein the memory impact index includes a structural deformation index, a quantity change index and a receptor activation index;

[0113] The probability completion module 203 is used to identify the causal association network between the anesthetic and the memory-affecting index, calculate the conditional probability between the anesthetic and the memory-affecting index in the causal association network, construct a node-probability regression model between the causal node in the causal association network and the conditional probability, and complete the conditional probability according to the node-probability regression model to obtain the completed conditional probability;

[0114] The model building module 204 is used to use the complementary conditional probability to build a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part, use the complementary conditional probability to build an indirect influence model between the memory influence index and the learning and memory ability, and use the complementary conditional probability to build an instrumental variable model between the memory influence index and the learning and memory ability;

[0115] The ability evaluation module 205 is used to calculate the final learning and memory ability of the learning and memory part using the direct influence model, the indirect influence model and the instrumental variable model, and use the final learning and memory ability as the evaluation result of the effect of the anesthetic on the learning and memory ability.

[0116] In detail, the modules in the evaluation system 300 for the effect of anesthetics on learning and memory abilities in the embodiment of the present invention are used in the same manner as described above. Figure 1 The same technical means are used as the method for evaluating the effects of anesthetics on learning and memory abilities described in the previous section, and can produce the same technical effects, so they will not be repeated here.

[0117] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0118] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0119] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0120] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0121] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for evaluating the effect of anesthetics on learning and memory ability, characterized in that: The method comprises: Inquiring the learning and memory sites where the anesthetic acts in the experimental group, collecting the experimental group samples of the learning and memory sites in the experimental group, and sampling the control group samples of the sites corresponding to the learning and memory sites in the control group; Using the experimental group samples and the control group samples to generate memory impact indexes of the anesthetic in the experimental group, wherein the memory impact indexes include a structural deformation index, a quantity change index, and a receptor activation index; Identify the causal association network between the anesthetic and the memory-affecting index, calculate the conditional probability between the anesthetic and the memory-affecting index in the causal association network, construct a node-probability regression model between the causal nodes in the causal association network and the conditional probability, and complete the conditional probability according to the node-probability regression model to obtain a completed conditional probability; Using the complementary conditional probability to construct a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part, using the complementary conditional probability to construct an indirect influence model between the memory influence index and the learning and memory ability, and using the complementary conditional probability to construct an instrumental variable model between the memory influence index and the learning and memory ability; The direct influence model, the indirect influence model and the instrumental variable model are used to calculate the final learning and memory ability of the learning and memory part, and the final learning and memory ability is used as the evaluation result of the effect of the anesthetic on the learning and memory ability.

2. The method according to claim 1, characterized in that The step of using the experimental group samples and the control group samples to generate the memory impact index of the anesthetic in the experimental group includes: The sample difference between the experimental group samples and the control group samples is used as the memory impact index.

3. The method according to claim 1, characterized in that The identifying a causal relationship network between the anesthetic and the memory-affecting index comprises: Querying the starting time of the anesthetic and the memory-affecting index; Arrange the anesthetic drugs and the memory-affecting indexes in the time sequence of the starting time to obtain a drug-index sequence; A causal association network between the anesthetic drug and the memory-affecting indicator is determined from the drug-indicator sequence.

4. The method according to claim 1, characterized in that: The step of constructing a node-probability regression model between the causal nodes in the causal association network and the conditional probabilities includes: Acquire a first correlation network, a second correlation network, and a third correlation network in the causal correlation network; Acquire all parent nodes of the first child node in the first association network, acquire all child nodes of the second parent node in the second association network, and acquire the historical parent node of the third child node in the third association network; Convert the first child node, all the parent nodes, the second parent node, all the child nodes, the third child node, and the historical parent node into a parent-child node pair; Based on the parent-child node pairing, a node-probability regression model between the causal node and the conditional probability is constructed.

5. The method according to claim 1, characterized in that The method of using the completion conditional probability to construct a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part includes: Based on the completion conditional probability, a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part is constructed.

6. The method according to claim 1, characterized in that The method of using the completion conditional probability to construct an indirect influence model between the memory influence index and the learning and memory ability includes: Based on the completion conditional probability, calculating an indirect variable between the memory impact index and the learning and memory ability; According to the completion conditional probability and the indirect variable, an indirect influence model between the memory influence index and the learning and memory ability is constructed.

7. The method according to claim 1, characterized in that The method of using the completed conditional probability to construct an instrumental variable model between the memory impact index and the learning and memory ability includes: Calculating the memory impact index under the instrumental variable between the memory impact index and the learning and memory ability according to the completion conditional probability; According to the memory influence index under the instrumental variable and the completion conditional probability, an instrumental variable model between the memory influence index and the learning and memory ability is constructed.

8. The method according to claim 1, characterized in that The method of calculating the final learning and memory ability of the learning and memory part by using the direct influence model, the indirect influence model and the instrumental variable model includes: The final learning and memory capacity of the learning and memory part is calculated according to the direct influence model, the indirect influence model and the instrumental variable model.

9. An evaluation system for the effect of anesthetics on learning and memory ability, characterized in that: The system comprises: A sample collection module, used to query the learning and memory sites where the anesthetic acts in the experimental group, collect the experimental group samples of the learning and memory sites in the experimental group, and sample the control group samples of the sites in the control group corresponding to the learning and memory sites; An index generation module, used to generate a memory impact index of the anesthetic in the experimental group using the experimental group samples and the control group samples, wherein the memory impact index includes a structural deformation index, a quantity change index and a receptor activation index; A probability completion module is used to identify the causal association network between the anesthetic and the memory-affecting index, calculate the conditional probability between the anesthetic and the memory-affecting index in the causal association network, construct a node-probability regression model between the causal node in the causal association network and the conditional probability, and complete the conditional probability according to the node-probability regression model to obtain a completed conditional probability; A model building module, used to use the complementary conditional probability to build a direct influence model between the memory influence index and the learning and memory ability of the learning and memory part, use the complementary conditional probability to build an indirect influence model between the memory influence index and the learning and memory ability, and use the complementary conditional probability to build an instrumental variable model between the memory influence index and the learning and memory ability; The ability evaluation module is used to calculate the final learning and memory ability of the learning and memory part using the direct influence model, the indirect influence model and the instrumental variable model, and use the final learning and memory ability as the evaluation result of the effect of the anesthetic on the learning and memory ability.

Citation Information

Patent Citations

  • Brain circuit causation analysis method based on heroin addiction model

    CN103235869A

  • Mobile wearable monitoring systems

    CN112998649A

  • Operation anesthesia selection recommendation method based on clinical medical data causal machine learning

    CN117275749A

  • Disease probability estimation method, system and equipment based on causal learning and medium

    CN117711634A

  • Causal analysis method based on Bayesian network

    CN118153688A