Intelligent medication management system for liver, gall and pancreas nursing

Abstract

The invention discloses an intelligent medication management system for liver, gall and pancreas nursing, relates to the technical field of medication management, and improves the medication safety of patients. According to the invention, the association degree among the disease semantic information sets is obtained according to the liver, gall and pancreas disease cases, a liver, gall and pancreas case analysis network and a plurality of interaction chains are established, and a plurality of stage prescription paths are established according to the liver, gall and pancreas disease cases. Matching each stage prescription path with the interaction chain to obtain a corresponding calibration stage prescription path, and inputting the calibration stage prescription path and the real-time body state data into a liver, gall and pancreas case analysis network, so that the liver, gall and pancreas case analysis network outputs a plurality of calibration stage prescription paths, and screening the prescription paths in the calibration stage through the drug allergy information, and selecting one prescription path in the calibration stage by a diagnosis and treatment doctor for execution.

Description

Technical Field

[0001] The present invention relates to the technical field of medication management, and in particular to an intelligent medication management system for hepatobiliary and pancreatic care. Background Art

[0002] As an important digestive and metabolic organ of the human body, the management of diseases of the hepatobiliary and pancreatic system is crucial to the health of patients. With the advancement of medical science and the development of intelligent technology, traditional drug management methods can no longer meet the needs of modern medicine for precise treatment and personalized care.

[0003] Most existing drug management systems rely on manual records and traditional management models, lacking intelligent support. This approach is not only prone to medication errors, but may also affect the patient's treatment effect due to untimely information transmission. For example, in terms of drug interactions, medication time, dosage adjustment, etc., medical staff need to rely on cumbersome manual calculations and experience judgments, which are prone to omissions and errors.

[0004] At the same time, many drug management systems are also deficient in data integration and analysis capabilities, and fail to effectively use big data and artificial intelligence technologies to optimize medication plans. This limits medical staff's understanding and response to patients' individualized medication needs, making it difficult to effectively ensure patients' medication safety.

[0005] Therefore, how to generate corresponding prescriptions according to the patient's drug allergy status while taking into account the drug interactions between drugs is a difficulty in the existing technology. To this end, an intelligent medication management system for hepatobiliary and pancreatic care is provided. Summary of the invention

[0006] In order to solve the above technical problems, the purpose of the present invention is to provide an intelligent medication management system for hepatobiliary and pancreatic care.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] An intelligent medication management system for hepatobiliary and pancreatic care, comprising a cloud computing platform, wherein the cloud computing platform is communicatively connected with a hepatobiliary and pancreatic case analysis module, a prescription analysis module and a medication analysis module;

[0009] The hepatobiliary and pancreatic case analysis module is used to obtain disease semantic information sets of various disease names based on a plurality of pre-stored hepatobiliary and pancreatic disease cases, and then obtain the correlation between each disease semantic information set, establish a lower-layer feature data network and an upper-layer disease association network, and then connect the lower-layer feature data network and the upper-layer disease association network to obtain a hepatobiliary and pancreatic case analysis network, and at the same time, establish a plurality of local disease association networks in the upper-layer disease association network according to the correlation between each disease semantic information set;

[0010] The prescription analysis module is used to establish a plurality of interaction chains according to a plurality of pre-stored drug information data, and to establish a plurality of stage prescription paths according to hepatobiliary and pancreatic disease cases, to match each stage prescription path with the interaction chain, to adjust the stage prescription path according to the matching result to obtain a corresponding calibration stage prescription path, and to input the calibration stage prescription path into the hepatobiliary and pancreatic case analysis network;

[0011] The medication analysis module is used to obtain the patient's real-time physical condition data and drug allergy information, and input the real-time physical condition data into the hepatobiliary and pancreatic case analysis network, which then outputs a number of calibration stage prescription paths based on the local disease association network, and screens the calibration stage prescription paths through drug allergy information, and the treating physician selects a calibration stage prescription path to execute. Whenever a calibration stage prescription path is executed, the patient's real-time physical condition data is collected again and input into the hepatobiliary and pancreatic case analysis network to output the next calibration stage prescription path until the patient recovers.

[0012] Furthermore, the hepatobiliary and pancreatic case analysis module pre-stores a number of hepatobiliary and pancreatic disease cases, wherein the hepatobiliary and pancreatic disease cases include the patient's disease name, initial physical condition data, and prescription information;

[0013] The prescription information consists of multiple stage prescriptions, and the stage prescriptions include multiple drug names, drug dosages, drug use sequences, and drug use times.

[0014] Furthermore, the disease semantic information set acquisition process includes:

[0015] Obtaining several normal physical state data intervals through the Internet, comparing the initial physical state data of multiple hepatobiliary and pancreatic disease cases with the same disease name with the corresponding normal physical state data intervals, and if the initial physical state data is not within the corresponding normal physical state data interval, determining that the corresponding initial physical state data is abnormal initial physical state data;

[0016] If the initial body state data is within the corresponding normal body state data interval, the corresponding stage body state data is determined to be normal initial body state data;

[0017] The same type of abnormal stage physical state data in multiple hepatobiliary and pancreatic disease cases with the same disease name are normally distributed, and the characteristic state change data intervals are obtained according to the normal distribution results. The characteristic state change data intervals of each disease name are integrated to establish the corresponding disease semantic information set.

[0018] Furthermore, the process of obtaining the correlation between each disease semantic information set includes:

[0019] Disease semantic information set is represented as {Name m ,{name1,Interval1},…,{name n , Interval n}}, where Name m Indicates the name of the mth disease, name n and Interval n Respectively represent the name and value range of the nth characteristic state change data, and m and n are natural numbers greater than 0;

[0020] Obtain the correlation θ between each disease semantic information set, where the calculation formula of the correlation θ is:

[0021] ;

[0022] in and They represent the correlation between the i-th disease semantic information set and the j-th disease semantic information set, represents the number of data with the same feature state changes in the i-th disease semantic information set and the j-th disease semantic information set, and They respectively represent the monomer association degrees of the kth feature state change data of the ith disease semantic information set relative to the jth disease semantic information set and the jth disease semantic information set relative to the ith disease semantic information set, and i and j are natural numbers greater than 0 and less than or equal to m.

[0023] Furthermore, the process of establishing a local disease association network within the upper disease association network includes:

[0024] According to the number of feature state change data intervals and the number of disease names in the entire disease semantic information set, a number of feature data nodes and disease nodes are established respectively, and at the same time, a lower-layer feature data network and an upper-layer disease association network are established, and each feature state change data name and disease name are input into the feature data node and the disease node respectively;

[0025] All feature data nodes are input into the lower-layer feature data network, and all disease nodes are input into the upper-layer disease association network. According to the feature state change data name of each disease node corresponding to the disease semantic information set, each feature data node is connected to the corresponding disease node, and the feature state change data interval is marked on the connection line, thereby obtaining the hepatobiliary and pancreatic case analysis network;

[0026] In the upper-level disease association network, each disease node is taken as the association center node in turn, and then the association distance between each disease node and the association center node is set according to the correlation between the disease semantic information set corresponding to the association center node and the disease semantic information sets corresponding to other disease nodes, and then several local disease association networks are established in the upper-level disease association network.

[0027] Furthermore, the process of establishing the interaction chain includes:

[0028] The prescription analysis module pre-stores several types of drug information data, and the drug information data includes drug names and drug action records;

[0029] The drug action record includes the interaction effects of multiple drugs in different use orders, and the interaction effects are divided into additive effects, antagonistic effects, and no-effect effects, such as reducing or increasing drug effects and increasing bleeding risks;

[0030] Then, a plurality of drug information nodes are established according to the total number of drug information data, and the drug name and drug treatment effect in each drug information data are respectively input into the drug information node;

[0031] According to the drug action records, an interaction chain is set between each drug information node, and the corresponding interaction effect is marked on the connection line between each drug information node in the interaction path chain, thereby obtaining a drug interaction relationship network.

[0032] Furthermore, the process of establishing the prescription path in the calibration phase includes:

[0033] Extract all prescription information from each hepatobiliary and pancreatic disease case, and establish several stage prescription paths according to the prescriptions of each stage in each prescription information by connecting the drug information nodes, and mark the medication time and drug dosage on the connection lines between the drug information nodes associated with the prescription paths of each stage;

[0034] The prescription paths at each stage are matched with the interaction chains in the drug interaction relationship network, and the arrangement order between the drug information nodes corresponding to the stage prescription paths is changed according to the matching results, so as to obtain several calibrated stage prescription paths.

[0035] Further, the process of screening and executing the prescription pathway in the calibration phase includes:

[0036] Obtain the patient's real-time physical condition data and drug allergy information, obtain the name of the drug that the patient cannot use based on the drug allergy information, and input all real-time physical condition data into the hepatobiliary and pancreatic case analysis network;

[0037] Match corresponding feature data nodes in the lower layer feature data network according to the data types of each real-time body status data;

[0038] According to the matching results, the abnormal feature node combination of the patient is matched with the disease node in the upper-layer disease association network, and then it is determined whether each real-time physical state data is within the feature state change data interval on the corresponding connection line, and the association distance threshold is set, and the matched disease node is used as the association center node to retrieve the corresponding local disease association network in the upper-layer disease association network, and then the disease node whose association distance with the association center node is less than or equal to the association distance threshold is selected from the local disease association network;

[0039] The calibration phase prescription paths associated with the associated central node and each selected disease node are obtained, and the corresponding calibration phase prescription paths are eliminated according to the names of the drugs that the patient cannot use. All the retained calibration phase prescription paths are sent to the treating physician in order, and the treating physician then selects a calibration phase prescription path to execute.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The present invention obtains the correlation between various disease semantic information sets according to hepatobiliary and pancreatic disease cases, establishes a hepatobiliary and pancreatic case analysis network, pre-stores several kinds of drug information data to establish several interaction chains, and establishes several stage prescription paths according to the hepatobiliary and pancreatic disease cases, matches each stage prescription path with the interaction chain to obtain a corresponding calibration stage prescription path, and inputs the calibration stage prescription path into the hepatobiliary and pancreatic case analysis network, inputs real-time physical status data into the hepatobiliary and pancreatic case analysis network, and then the hepatobiliary and pancreatic case analysis network outputs several calibration stage prescription paths, and screens the calibration stage prescription paths through drug allergy information, and the treating doctor selects a calibration stage prescription path to execute, thereby realizing the generation of corresponding prescriptions according to the patient's drug allergy status while considering the drug interactions between drugs. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention.

[0043] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0044] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0045] like Figure 1 As shown, an intelligent medication management system for hepatobiliary and pancreatic care includes a cloud computing platform, wherein the cloud computing platform is communicatively connected to a hepatobiliary and pancreatic case analysis module, a prescription analysis module, and a medication analysis module;

[0046] The hepatobiliary and pancreatic case analysis module is used to obtain disease semantic information sets of various disease names based on a plurality of pre-stored hepatobiliary and pancreatic disease cases, and then obtain the correlation between each disease semantic information set, establish a lower-layer feature data network and an upper-layer disease association network, and then connect the lower-layer feature data network and the upper-layer disease association network to obtain a hepatobiliary and pancreatic case analysis network, and at the same time, establish a plurality of local disease association networks in the upper-layer disease association network according to the correlation between each disease semantic information set;

[0047] The prescription analysis module is used to establish a plurality of interaction chains according to a plurality of pre-stored drug information data, and to establish a plurality of stage prescription paths according to hepatobiliary and pancreatic disease cases, to match each stage prescription path with the interaction chain, to adjust the stage prescription path according to the matching result to obtain a corresponding calibration stage prescription path, and to input the calibration stage prescription path into the hepatobiliary and pancreatic case analysis network;

[0048] The medication analysis module is used to obtain the patient's real-time physical condition data and drug allergy information, and input the real-time physical condition data into the hepatobiliary and pancreatic case analysis network, which then outputs a number of calibration stage prescription paths based on the local disease association network, and screens the calibration stage prescription paths through drug allergy information, and the treating physician selects a calibration stage prescription path to execute. Whenever a calibration stage prescription path is executed, the patient's real-time physical condition data is collected again and input into the hepatobiliary and pancreatic case analysis network to output the next calibration stage prescription path until the patient recovers.

[0049] Further, the working principle of the present invention is described below by way of examples:

[0050] The hepatobiliary and pancreatic case analysis module pre-stores a number of hepatobiliary and pancreatic disease cases, wherein the hepatobiliary and pancreatic disease cases include the patient's gender, age, disease name, initial physical condition data, drug allergy history, and prescription information;

[0051] The initial physical condition data refers to the physical condition data of the patient before treatment, wherein the data types include liver enzyme level, blood sugar level, cholesterol level, etc.;

[0052] The prescription information is composed of multiple stage prescriptions, and the stage prescriptions include multiple drug names, drug dosages, drug order, and drug time;

[0053] Set multiple age ranges and obtain several normal physical condition data ranges of different genders in each age range through the Internet;

[0054] Comparing each item of initial physical state data in multiple hepatobiliary and pancreatic disease cases with the same disease name with the corresponding normal physical state data interval, if the initial physical state data is not within the corresponding normal physical state data interval, then judging the corresponding initial physical state data as abnormal initial physical state data;

[0055] If the initial body state data is within the corresponding normal body state data interval, the corresponding stage body state data is determined to be normal initial body state data;

[0056] Normally distribute the same type of abnormal stage physical status data in multiple hepatobiliary and pancreatic disease cases with the same disease name, and then select the central part of the normal distribution result, and select the interval equivalent to one quarter of the whole part from both the left and right sides, and record the interval as a characteristic state change data interval corresponding to the disease name;

[0057] Through the above process of obtaining the characteristic state change data interval of the disease name, all the characteristic state change data intervals of each disease name are obtained, and the corresponding disease semantic information set is established according to the characteristic state change data intervals of each disease name, where the disease semantic information set can be expressed as {Name m ,{name1,Interval1},…,{name n , Interval n}}, where Name m Indicates the name of the mth disease, name n and Interval n Respectively represent the name and value range of the nth characteristic state change data, where m and n are natural numbers greater than 0;

[0058] Obtain the correlation θ between each disease semantic information set, where the calculation formula of the correlation θ is:

[0059] ;

[0060] in and They represent the correlation between the i-th disease semantic information set and the j-th disease semantic information set, represents the number of data with the same feature state changes in the i-th disease semantic information set and the j-th disease semantic information set, and They respectively represent the monomer association degree of the kth feature state change data of the ith disease semantic information set relative to the jth disease semantic information set and the jth disease semantic information set relative to the ith disease semantic information set, i and j are natural numbers greater than 0 and less than or equal to m;

[0061] It should be noted that and To illustrate, for example, two disease semantic information sets with different disease names both carry blood sugar characteristic state change data intervals. The blood sugar characteristic state change data interval of one of the disease semantic information sets is (7.1-8.0 mg / dL), and the blood sugar characteristic state change data interval of the other disease semantic information set is (7.3-7.6 mg / dL). Then the monomer associations of the two disease semantic information sets for the blood sugar characteristic state change data intervals are 1 and 0.5 respectively.

[0062] Further, a number of feature data nodes and disease nodes are established respectively according to the number of feature state change data intervals and the number of disease names in the entire disease semantic information set, and a lower-layer feature data network and an upper-layer disease association network are established at the same time, and each feature state change data name and disease name are input into the feature data node and the disease node respectively;

[0063] All feature data nodes are input into the lower-layer feature data network, and all disease nodes are input into the upper-layer disease association network. According to the feature state change data name of each disease node corresponding to the disease semantic information set, each feature data node is connected to the corresponding disease node, and the feature state change data interval is marked on the connection line, thereby obtaining the hepatobiliary and pancreatic case analysis network;

[0064] In the upper-layer disease association network, each disease node is taken as the association center node in turn, and then the association distance between each disease node and the association center node is set according to the association degree between the disease semantic information set corresponding to the association center node and the disease semantic information set corresponding to other disease nodes, and then several local disease association networks are established in the upper-layer disease association network;

[0065] It should be noted that the greater the correlation degree, the smaller the correlation distance between the corresponding disease node and the correlation center node, but the correlation distance is not equal to 0.

[0066] Further, the hepatobiliary and pancreatic case analysis module sends all hepatobiliary and pancreatic disease cases to the prescription analysis module;

[0067] The prescription analysis module pre-stores several types of drug information data, and the drug information data includes drug names, drug treatment effects, and drug action records;

[0068] The drug action record includes the interaction effects of multiple drugs in different use orders, and the interaction effects are divided into additive effects, antagonistic effects, and no-effect effects, such as reducing or increasing drug effects and increasing bleeding risks;

[0069] Then, a plurality of drug information nodes are established according to the total number of drug information data, and the drug name and drug treatment effect in each drug information data are respectively input into the drug information node;

[0070] According to the drug action records, an interaction chain is set between each drug information node, and the corresponding interaction effect is marked on the connection line between each drug information node in the interaction path chain, thereby obtaining a drug interaction relationship network;

[0071] It should be noted that the interaction chain has a situation where multiple drug information nodes are simultaneously connected to one drug information node;

[0072] At the same time, all prescription information is extracted from each hepatobiliary and pancreatic disease case, and according to the prescriptions of each stage in each prescription information, the drug information nodes are connected to establish several stage prescription paths, and the medication time and drug dosage are marked on the connection lines between each drug information node associated with each stage prescription path;

[0073] Match the prescription paths of each stage with the interaction chains in the drug interaction relationship network, and obtain the interaction information between adjacent drug information nodes in the prescription paths of each stage according to the matching results. If it is judged that there is an antagonistic effect between adjacent drug information nodes according to the matching, the arrangement order between the corresponding drug information nodes in the prescription paths of the stage is changed so that the two are not adjacent.

[0074] If there is an additive effect between adjacent drug information nodes according to the matching, but the arrangement order of the corresponding adjacent drug information nodes in the original stage prescription path makes the two unable to adapt to the additive effect, then the arrangement order between the corresponding drug information nodes in the stage prescription path is swapped to make the additive effect adapt;

[0075] If it is determined based on the matching that there is only no interaction effect between adjacent drug information nodes, no operation is performed;

[0076] Repeat the above process of adjusting the order of drug information nodes in the prescription path of the stage, and then obtain several prescription paths of the calibration stage.

[0077] Furthermore, the medication analysis module obtains the hepatobiliary and pancreatic case analysis network and all calibration phase prescription paths from the hepatobiliary and pancreatic case analysis module and the prescription analysis module, and binds each calibration phase prescription path to the corresponding disease node according to the prescription information in the hepatobiliary and pancreatic disease cases;

[0078] The medication analysis module obtains the patient's real-time physical condition data and drug allergy information, obtains the name of the drug that the patient cannot use based on the drug allergy information, and inputs all real-time physical condition data into the hepatobiliary and pancreatic case analysis network;

[0079] Match corresponding feature data nodes in the lower layer feature data network according to the data types of each real-time body status data;

[0080] According to the matching results, the abnormal feature nodes of the patient are combined and matched with the disease nodes in the upper disease association network, and then it is determined whether the real-time physical status data are within the feature status change data interval of the corresponding connection line;

[0081] If more than 90% of the real-time body status data are within the characteristic state change data interval on the corresponding connection line, it is determined to match the corresponding disease node, otherwise it is determined not to match;

[0082] Set an association distance threshold, use the matched disease node as the association center node to retrieve the corresponding local disease association network in the upper-layer disease association network, and then select the disease node whose association distance with the association center node is less than or equal to the association distance threshold from the local disease association network;

[0083] Then, the calibration phase prescription paths associated with the associated central node and each selected disease node are obtained, and the names of the drugs that the patient cannot use are matched with the prescription paths of each calibration phase. If there is a drug information node with a drug name that the patient cannot use in the prescription path of the calibration phase, the corresponding prescription path of the calibration phase is removed, otherwise no operation is performed;

[0084] All the reserved calibration phase prescription paths are sent to the treating doctors in order, and then the treating doctors select a calibration phase prescription path to execute. It should be noted that if the reserved calibration phase prescription path has a calibration phase prescription path associated with the associated center node, the calibration phase prescription path is placed first, and the remaining calibration phase prescription paths are sorted in order according to the size of the associated distance between the corresponding disease node and the associated center node;

[0085] When the prescription path of the calibration phase is executed, the medication analysis module re-collects the patient's real-time physical condition data, generates and executes the prescription path of the next calibration phase until all the patient's real-time physical condition data have no matching results with the underlying feature data network.

[0086] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent medication management system for hepatobiliary and pancreatic care, comprising a cloud computing platform, characterized in that: The cloud computing platform is communicatively connected to a hepatobiliary and pancreatic case analysis module, a prescription analysis module, and a medication analysis module; The hepatobiliary and pancreatic case analysis module is used to obtain disease semantic information sets of various disease names based on a plurality of pre-stored hepatobiliary and pancreatic disease cases, and then obtain the correlation between each disease semantic information set, establish a lower-layer feature data network and an upper-layer disease association network, and then connect the lower-layer feature data network and the upper-layer disease association network to obtain a hepatobiliary and pancreatic case analysis network, and at the same time, establish a plurality of local disease association networks in the upper-layer disease association network according to the correlation between each disease semantic information set; The prescription analysis module is used to establish a plurality of interaction chains according to a plurality of pre-stored drug information data, and to establish a plurality of stage prescription paths according to hepatobiliary and pancreatic disease cases, to match each stage prescription path with the interaction chain, to adjust the stage prescription path according to the matching result to obtain a corresponding calibration stage prescription path, and to input the calibration stage prescription path into the hepatobiliary and pancreatic case analysis network; The medication analysis module is used to obtain the patient's real-time physical condition data and drug allergy information, input the real-time physical condition data into the hepatobiliary and pancreatic case analysis network, and then output a number of calibration stage prescription paths based on the local disease association network, and screen the calibration stage prescription paths through the drug allergy information, and the treating physician selects a calibration stage prescription path to execute. Whenever a calibration stage prescription path is executed, the patient's real-time physical condition data is re-collected and the next calibration stage prescription path is obtained until the patient recovers.

2. According to claim 1, an intelligent medication management system for hepatobiliary and pancreatic care is characterized in that: The hepatobiliary and pancreatic disease cases include the patient's disease name, initial physical condition data and prescription information, and the prescription information consists of multiple stage prescriptions, and the stage prescriptions include multiple drug names, drug dosages, drug order and medication time.

3. The intelligent medication management system for hepatobiliary and pancreatic care according to claim 2 is characterized in that: The disease semantic information set acquisition process includes: Obtaining several normal physical state data intervals through the Internet, comparing the initial physical state data of multiple hepatobiliary and pancreatic disease cases with the same disease name with the corresponding normal physical state data intervals, and dividing the initial physical state data into abnormal initial physical state data and normal initial physical state data according to the comparison results; The same type of abnormal stage physical state data in multiple hepatobiliary and pancreatic disease cases with the same disease name are normally distributed, and the characteristic state change data intervals are obtained according to the normal distribution results. The characteristic state change data intervals of each disease name are integrated to establish the corresponding disease semantic information set.

4. The intelligent medication management system for hepatobiliary and pancreatic care according to claim 3 is characterized in that: The process of obtaining the correlation between each disease semantic information set includes: Disease semantic information set is represented as {Name m ,{name1,Interval1},…,{name n , Interval n }}, where Name m Indicates the name of the mth disease, name n and Interval n Respectively represent the name and value range of the nth characteristic state change data, m and n are natural numbers greater than 0, and the correlation θ between each disease semantic information set is obtained, where the calculation formula of the correlation θ is: ； in and They represent the correlation between the i-th disease semantic information set and the j-th disease semantic information set, represents the number of data with the same feature state changes in the i-th disease semantic information set and the j-th disease semantic information set, and They respectively represent the monomer association degrees of the kth feature state change data of the ith disease semantic information set relative to the jth disease semantic information set and the jth disease semantic information set relative to the ith disease semantic information set, and i and j are natural numbers greater than 0 and less than or equal to m.

5. The intelligent medication management system for hepatobiliary and pancreatic care according to claim 4 is characterized in that: The process of establishing a local disease association network within the upper disease association network includes: Establish a number of feature data nodes, disease nodes, lower-layer feature data networks, and upper-layer disease association networks, and input each feature state change data name and disease name into the feature data node and disease node respectively; All feature data nodes are input into the lower-layer feature data network, and all disease nodes are input into the upper-layer disease association network. According to the feature state change data name of each disease node corresponding to the disease semantic information set, each feature data node is connected to the corresponding disease node, and the feature state change data interval is marked on the connection line, thereby obtaining the hepatobiliary and pancreatic case analysis network; In the upper-level disease association network, each disease node is taken as the association center node in turn. According to the correlation between the disease semantic information set corresponding to the association center node and the disease semantic information sets corresponding to other disease nodes, the association distance between each disease node and the association center node is set, and then several local disease association networks are established in the upper-level disease association network.

6. The intelligent medication management system for hepatobiliary and pancreatic care according to claim 5, characterized in that: The process of establishing the interaction chain includes: The prescription analysis module pre-stores several types of drug information data, and the drug information data includes drug names and drug action records; The drug action record includes the interaction effects of multiple drugs in different usage orders, and then a number of drug information nodes are established according to the total number of drug information data, and the drug name and drug treatment effect in each drug information data are respectively input into the drug information node; According to the drug action records, an interaction chain is set between each drug information node, and the corresponding interaction effect is marked on the connection line between each drug information node in the interaction path chain, thereby obtaining a drug interaction relationship network.

7. The intelligent medication management system for hepatobiliary and pancreatic care according to claim 6, characterized in that: The process of establishing the prescription path in the calibration phase includes: According to the prescriptions at each stage in the prescription information, the drug information nodes are connected to obtain several stage prescription paths, and the prescription paths at each stage are matched with the interaction chains in the drug interaction relationship network. According to the matching results, the arrangement order between the drug information nodes corresponding to the stage prescription paths is changed, so as to obtain several calibrated stage prescription paths.

8. The intelligent medication management system for hepatobiliary and pancreatic care according to claim 7, characterized in that: The process of screening and executing the prescription pathway in the calibration phase includes: Obtain various real-time physical status data and drug allergy information of the patient, obtain the name of the drug that the patient cannot use according to the drug allergy information, and match the corresponding feature data node in the lower-layer feature data network according to the data type of various real-time physical status data; According to the matching results, the abnormal feature node combination of the patient is matched with the disease node in the upper-layer disease association network, and then it is determined whether each real-time physical state data is within the feature state change data interval on the corresponding connection line, and the association distance threshold is set, and the matched disease node is used as the association center node to retrieve the corresponding local disease association network in the upper-layer disease association network, and then the disease node whose association distance with the association center node is less than or equal to the association distance threshold is selected from the local disease association network; The calibration phase prescription paths associated with the associated central node and each selected disease node are obtained, and the corresponding calibration phase prescription paths are eliminated according to the names of the drugs that the patient cannot use. All the retained calibration phase prescription paths are sent to the treating physician in order, and the treating physician then selects a calibration phase prescription path to execute.

Images