CN114864107A

CN114864107A - Clinical pathway variation analysis method, equipment and storage medium

Info

Publication number: CN114864107A
Application number: CN202110152482.2A
Authority: CN
Inventors: 孙新宝
Original assignee: Alibaba Group Holding Ltd
Current assignee: Alibaba Group Holding Ltd
Priority date: 2021-02-03
Filing date: 2021-02-03
Publication date: 2022-08-05

Abstract

The embodiment of the application provides a clinical pathway variation analysis method, equipment and a storage medium. In the embodiment of the application, the actual diagnosis and treatment path of the target patient can be extracted from the target diagnosis and treatment process record of the target patient, and the target clinical path of the target diagnosis and treatment process video of the target patient can be determined from at least one clinical path; and performing path variation analysis based on the actual diagnosis and treatment path and the target clinical path to determine path variation information of the target patient in the target diagnosis and treatment process. Therefore, in the embodiment of the application, the actual diagnosis and treatment path can be automatically extracted from the target diagnosis and treatment process record of the target patient and used as the basis for the clinical path variation analysis, so that the data value of the diagnosis and treatment process record of the patient can be fully exerted, the path variation of the actual diagnosis and treatment path relative to the clinical path can be rapidly and accurately analyzed, and the clinical path variation analysis can be realized.

Description

Clinical pathway variation analysis method, equipment and storage medium

Technical Field

The present application relates to the field of medical management technologies, and in particular, to a method, an apparatus, and a storage medium for analyzing clinical pathway variation.

Background

With the continuous advance of medical innovation, the state releases clinical path management guidance specifications for many times, and hopes to abstract clinical paths for common diseases and frequently encountered diseases so as to guide the standardized diagnosis and treatment process of medical institutions.

At present, medical staff are usually required to arrange the execution conditions of clinical routes in forms of statistical tables or work reports and the like, which wastes time and labor, and the accuracy and the real-time performance are poor, thereby causing the implementation bottleneck of clinical route management.

Disclosure of Invention

Aspects of the present disclosure provide a method, an apparatus, and a storage medium for analyzing clinical pathway variation, so as to analyze the performance of a clinical pathway more quickly and accurately.

The embodiment of the application provides a clinical pathway variation analysis method, which comprises the following steps:

responding to a variation analysis instruction, and extracting an actual diagnosis and treatment path of a target patient from a target diagnosis and treatment process record of the target patient;

determining a target clinical path matched with the target diagnosis and treatment process of the target patient from at least one clinical path;

and performing path variation analysis based on the actual diagnosis and treatment path and the target clinical path to determine path variation information of the target patient in the target diagnosis and treatment process.

The embodiment of the present application further provides a diagnosis and treatment path guiding method, including:

responding to a path guide instruction, and extracting a historical diagnosis and treatment path of a target patient from a medical history record of the current diagnosis and treatment process of the target patient;

acquiring diagnosis and treatment paths corresponding to at least one reference patient;

determining a specific patient from the at least one reference patient, wherein the diagnosis and treatment path corresponding to the specific patient is matched with the historical diagnosis and treatment record of the target patient;

and guiding the diagnosis and treatment path of the target patient in the current diagnosis and treatment process according to the diagnosis and treatment path of the specific patient.

The embodiment of the present application further provides a method for analyzing path variation, including:

responding to a variation analysis instruction, and extracting an actual processing path corresponding to a target event from a processing process record of the target event;

determining a standard processing path adapted to the target event;

and performing path variation analysis based on the actual processing path and the standard processing path to determine path variation information under the target event.

The embodiment of the application also provides a computing device, which comprises a memory and a processor;

the memory is to store one or more computer instructions;

the processor is coupled with the memory for executing the one or more computer instructions for:

determining a target clinical path which is adaptive to the target diagnosis and treatment process of the target patient from at least one clinical path;

the memory is to store one or more computer instructions;

the processor, coupled with the memory, to execute the one or more computer instructions to:

the memory is to store one or more computer instructions;

determining a standard processing path adapted to the target event;

Embodiments of the present application also provide a computer-readable storage medium storing computer instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the aforementioned clinical pathway variation analysis method, clinical pathway guidance method, or pathway variation analysis method.

Embodiments also provide a computer program product including a computer program/instructions, wherein when the computer program/instructions are executed by a processor, the processor is caused to implement the steps of the clinical pathway variation analysis method, the clinical pathway guidance method, or the pathway variation analysis method.

In the embodiment of the application, the actual diagnosis and treatment path of the target patient can be extracted from the target diagnosis and treatment process record of the target patient, and the target clinical path of the target diagnosis and treatment process video of the target patient can be determined from at least one clinical path; and performing path variation analysis based on the actual diagnosis and treatment path and the target clinical path to determine path variation information of the target patient in the target diagnosis and treatment process. Therefore, in the embodiment of the application, the actual diagnosis and treatment path can be automatically extracted from the target diagnosis and treatment process record of the target patient and used as the basis for the clinical path variation analysis, so that the data value of the diagnosis and treatment process record of the patient can be fully exerted, the path variation of the actual diagnosis and treatment path relative to the clinical path can be rapidly and accurately analyzed, and the clinical path variation analysis can be realized.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:

FIG. 1a is a schematic flow chart of a method for clinical pathway variation analysis according to an exemplary embodiment of the present application;

FIG. 1b is a schematic logic diagram of a clinical pathway variation analysis protocol according to an exemplary embodiment of the present application;

fig. 2a is a schematic flowchart of a method for analyzing path variation according to an exemplary embodiment of the present disclosure;

FIG. 2b is a schematic diagram of the logic of a path variation analysis scheme provided in an exemplary embodiment of the present application;

fig. 3 is a schematic flowchart of a diagnosis and treatment path guidance method according to another exemplary embodiment of the present disclosure;

fig. 4 is a logic diagram of a diagnosis and treatment path guidance plan according to another exemplary embodiment of the present application;

fig. 5a is a schematic logic diagram of a protocol for determining a treatment path of a reference patient according to another exemplary embodiment of the present application;

fig. 5b is a logic diagram illustrating a diagnosis path extraction scheme according to an exemplary embodiment of the present application;

FIG. 6 is a schematic block diagram of a computing device according to yet another exemplary embodiment of the present application;

FIG. 7 is a schematic block diagram of another computing device provided in accordance with yet another exemplary embodiment of the present application;

fig. 8 is a schematic structural diagram of another computing device according to another exemplary embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only a few embodiments of the present application, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

Before describing the technical solution of the embodiments of the present application in detail, the concept of the clinical pathway related to the embodiments of the present application will be briefly described. The clinical pathway refers to a set of standardized diagnosis and treatment procedures established for a certain disease. The clinical route is implemented in comparison with the traditional route, which is a diagnosis and treatment program established by a diagnosis and treatment party according to experience, and different diagnosis and treatment programs may be adopted by different regions, different hospitals, different treatment groups or different doctors and individuals aiming at a certain disease. After the clinical path is adopted, the diagnosis and treatment program of the disease can be standardized, the medical behavior is standardized, the diagnosis and treatment quality is improved, and the medical cost is reduced.

At present, the execution condition of clinical paths needs to be manually cleared up, time and labor are wasted, and instantaneity and accuracy are poor. To this end, in some embodiments of the present application: the actual diagnosis and treatment path of the target patient can be extracted from the target diagnosis and treatment process record of the target patient, and a target clinical path of a target diagnosis and treatment process video of the target patient can be determined from at least one clinical path; and performing path variation analysis based on the actual diagnosis and treatment path and the target clinical path to determine path variation information of the target patient in the target diagnosis and treatment process. Therefore, in the embodiment of the application, the actual diagnosis and treatment path can be automatically extracted from the target diagnosis and treatment process record of the target patient and used as the basis for the clinical path variation analysis, so that the data value of the diagnosis and treatment process record of the patient can be fully exerted, the path variation of the actual diagnosis and treatment path relative to the clinical path can be rapidly and accurately analyzed, and the clinical path variation analysis can be realized.

The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

Fig. 1a is a flowchart illustrating a method for analyzing clinical pathway variation according to an exemplary embodiment of the present disclosure. FIG. 1b is a logic diagram of a clinical pathway variation analysis protocol according to an exemplary embodiment of the present application. Referring to fig. 1a, the method may be performed by a clinical pathway variation analysis apparatus, which may be implemented as a combination of software and/or hardware, which may be integrated in a computing device. Referring to fig. 1a, the method comprises:

step 100, in response to a variation analysis instruction, extracting an actual diagnosis and treatment path of a target patient from a target diagnosis and treatment process record of the target patient;

step 101, determining a target clinical path adapted to a target diagnosis and treatment process of a target patient from at least one clinical path;

102, performing path variation analysis based on the actual diagnosis and treatment path and the target clinical path to determine path variation information of the target patient in the target diagnosis and treatment process.

In this embodiment, the target patient may be any patient who is visited by a medical institution, and the target diagnosis and treatment process may be any diagnosis and treatment process that occurs in the target patient, that is, the embodiment may perform clinical path variation analysis on any diagnosis and treatment process of the target patient, where one time of the patient is visited corresponds to one diagnosis and treatment process. For example, the target diagnosis process may be a diagnosis process that has been completed by the target patient last time, a diagnosis process that is being performed by the target patient, or a diagnosis process that has been completed by the target patient historically. Each diagnosis and treatment process can have a corresponding diagnosis and treatment process record. The medical procedure record can be used for recording the current visit information of the patient. The medical procedure record includes, but is not limited to, basic diseases of patients, identity information, operation records of current medical treatment, medicines used in current medical treatment, medication orders, examination records of current medical treatment, and the like.

Based on this, referring to fig. 1a and 1b, in step 100, an actual diagnosis path of the target patient may be extracted from a target diagnosis process record of the target patient. The actual diagnosis and treatment path may include a diagnosis and treatment program actually adopted by the target patient in the target diagnosis and treatment process.

In this embodiment, the actual diagnosis and treatment path may be stored according to a specified data structure, and therefore, in this embodiment, the target diagnosis and treatment process record may be structured according to the specified data structure, so as to extract the actual diagnosis and treatment path of the target patient from the target diagnosis and treatment process record. That is, various data required by a specified data structure, such as diagnosis and treatment entities, entity relationships, entity attributes and the like, can be extracted from a target diagnosis and treatment process record of a target patient, and the extracted data is stored according to a specified data format to represent an actual diagnosis and treatment path.

Generally, each medical visit of a patient will generate a medical visit file, which at least includes two parts: history records and current diagnosis and treatment process records. With the popularity of digital medical management in medical institutions, current medical procedure records for target patients are often structured, while medical history records may remain unstructured, e.g., in data formats such as text or images. Based on this, in step 100, path data required for various path construction, such as medical entities and entity relationships, can be directly extracted from the target medical procedure record (formatted) according to a specified data structure to construct an actual medical path. Of course, in some cases, the target procedure record may also be unstructured, for example, text or images may be stored as a storage medium for a relatively long-lived procedure. In these cases, the target procedure record is typically unstructured. In step 100, path data required for various path construction can be identified from the target diagnosis and treatment process record by means of Natural Language Processing (NLP) and the like, so as to construct an actual diagnosis and treatment path, wherein after the NLP is performed, extracted data can be normalized, so that different representations of the same entity are normalized into the same representation, and thus, the accuracy of the actual diagnosis and treatment path is improved. Optionally, in this embodiment, the normalization model may be trained through labeled samples of several clinical entities, so as to perform a normalization operation using the normalization model, where the normalization model may be a machine learning model. It should be understood that the actual clinical path herein is a set of data having a specified data structure.

Referring to fig. 1a and 1b, in step 101, a target clinical path adapted to a target medical procedure of a target patient may be determined from at least one clinical path. Generally, the official release a description file of the clinical path, which includes but is not limited to applicable objects, diagnostic basis, entry path standard, examination items during hospitalization, preoperative examination, medication selection and applicable time, surgery selection, postoperative recovery, discharge standard, and the like. In this embodiment, at least one description file of the clinical path may be obtained, and at least one clinical path may be extracted from the at least one description file of the clinical path according to the designated data structure. Thus, at least one clinical pathway can be stored in advance in the clinical pathway variation analysis device of the present embodiment for future use. In practical applications, considering the limited number of clinical pathways, the clinical pathways can be disassembled from the description file of the clinical pathways in a manual mode. For example, path data required for constructing various paths can be manually extracted from the description file of the clinical path according to a specified data structure and recorded in a storage medium (such as a database) of the clinical path to construct at least one clinical path, which can ensure the accuracy of the clinical path. Of course, in this embodiment, a machine learning manner may also be adopted, and according to a specified data structure, various path data required for constructing the path are extracted from the description files of the clinical paths, and are recorded in the storage medium of the clinical paths, so as to construct at least one clinical path. The construction scheme of the clinical pathway in this embodiment is not limited thereto.

On the basis, referring to fig. 1a and 1b, in step 101, a target clinical pathway can be selected from at least one clinical pathway constructed in advance. The target clinical pathway conforms to the data structure specified in this embodiment. Therefore, the data structure adopted by the actual diagnosis and treatment path corresponding to the target diagnosis and treatment process of the target patient and the target clinical path can be ensured to be consistent.

Referring to fig. 1a and 1b, in the present embodiment, in step 102, a path variation analysis may be performed based on the actual diagnosis and treatment path and the target clinical path to determine path variation information of the target patient during the target diagnosis and treatment process. The path variation analysis may be to analyze a path variation occurring in the actual diagnosis path with respect to the target clinical path. Dimensions of the path variation analysis include, but are not limited to, variation nature, variation controllability, variation origin or variation factor, and the like. Wherein, the variability can include positive or negative variability, and negative variability means that the planned procedure is not performed or results are not produced, or completion is delayed, such as delayed discharge, delayed CT examination, etc.; positive variation is the planned procedure or outcome performed or completed in advance, such as early discharge from hospital, early CT examination, etc. The variability controllability refers to whether the occurring variability is controllable, for example, a new disease condition of a patient leads to an increase of a certain diagnosis and treatment link, which can be determined as uncontrollable, and the hospitalization time is prolonged by the request of family members of the patient, which can be determined as controllable. Sources of variation may include, but are not limited to, patient-specific variations, hospital variations, variations of clinician service providers, and the like. The variable factors may include, but are not limited to, examinations, drugs, consumables, surgeries, etc., and the variable factors may be classified into positive and negative variable factors. Of course, these are merely exemplary, the dimension of the path variance analysis in this embodiment is not limited thereto, and in this embodiment, the dimension of the path variance analysis may be set according to actual needs, and the specified data format may be adjusted reversely based on the information types required by different dimensions of the path variance analysis.

Referring to fig. 1a and 1b, based on a consistent data structure between an actual diagnosis path and a target clinical path corresponding to a target diagnosis process of a target patient, in step 102, the actual diagnosis path and the clinical path may be aligned according to a specified data structure; and comparing the aligned actual diagnosis and treatment path with the clinical path to analyze the difference between the actual diagnosis and treatment path and the clinical path. The alignment refers to aligning data structures of an actual diagnosis path and a clinical path, for example, time alignment, entity attribute alignment, and the like. This ensures the rationality and accuracy of the mutation analysis.

Accordingly, in this embodiment, the path variation information of at least one dimension can be obtained in step 102.

In this embodiment, the path variation information can be visually displayed. In an optional display scheme, an actual diagnosis and treatment path and a target clinical path corresponding to a target diagnosis and treatment process of a target patient can be displayed respectively, and path variation labels are presented at variation positions on the actual diagnosis and treatment path and the target clinical path corresponding to the target diagnosis and treatment process of the target patient. The path mutation tag can be used for showing the content of the path mutation information. The actual diagnosis and treatment path and the target clinical path may be displayed in a knowledge graph manner, but the embodiment is not limited thereto. In addition, the present embodiment may also adopt other display schemes to visually display the path variation information, for example, a chart and the like may be adopted, which is not limited in the present embodiment.

In addition, in the embodiment, cost variation information of the target patient in the target diagnosis and treatment process can be calculated according to the path variation information of the target patient in the target diagnosis and treatment process. Since the cost variation may be caused by the path variation, the cost variation information may be calculated according to the path variation information in the embodiment. The cost variation information may include, but is not limited to, variation information in terms of consumable material cost, traditional Chinese medicine cost, western medicine cost, image inspection cost, laboratory inspection cost, or the like. For the cost aspect related to the medicine, the method can also be used for subdividing the cost of the monitored medicine, the cost of the non-monitored medicine, the cost of the auxiliary medicine, the cost of the non-auxiliary medicine and the like. The cost variation information can also be calculated from the level of coarser granularity such as treatment cost, rehabilitation cost and the like. The cost variation information in this embodiment is not limited to this, and the cost variation information to be calculated in this embodiment may be determined according to actual needs.

In summary, in the present embodiment, the actual diagnosis and treatment path can be automatically extracted from the target diagnosis and treatment process record of the target patient as the basis for the clinical path mutation analysis, so that the data value of the diagnosis and treatment process record of the patient can be fully exerted, and the path mutation of the actual diagnosis and treatment path relative to the clinical path can be rapidly and accurately analyzed, thereby implementing the clinical path mutation analysis. In addition, the actual diagnosis and treatment path and the target clinical path can be stored according to a uniform data structure, so that the comparison between the actual diagnosis and treatment path and the target clinical path can be better supported, and the efficiency and the accuracy of the path variation analysis are improved. The technical scheme provided by the embodiment can be used for analyzing whether the completed diagnosis and treatment process or the completed diagnosis and treatment program in the ongoing diagnosis and treatment process meets the diagnosis and treatment specification or not, and provides support for implementation management of a clinical path.

In the above or following embodiments, the target clinical path adapted to the target diagnosis and treatment process of the target patient may be determined from the at least one clinical path according to the actual condition of the target patient. Wherein, the actual condition of the target patient can be characterized by at least one patient information of the target patient.

In this embodiment, a medical history record of a target patient can be obtained; identifying at least one patient information affecting the clinical pathway assignment from the target medical procedure record and/or the medical history record; determining a target clinical pathway from the at least one clinical pathway according to the at least one patient information, wherein the at least one patient information meets the admission condition of the target clinical pathway.

The medical history of the target patient may refer to a historical medical information record of the target patient before the target diagnosis and treatment process. For example, the target patient's disease history, surgical history, examination history, allergy history, medication history, medical order history, and the like. As mentioned above, the history of the target patient is typically unstructured. For example, the medical history record can be a medical history portion of the medical record, typically based on text spoken by the patient, or medical record images downloaded from other hospitals, etc. In this embodiment, at least one patient information that may affect the clinical pathway assignment may be extracted from the medical history of the target patient by the NLP technology. After NLP is carried out, the extracted patient information can be normalized, so that different characteristics of the same patient information are normalized into the same characteristics, the accuracy of the patient information is improved, and similarly, a normalization model can be used for executing normalization operation. In addition to the medical history of the target patient, patient information affecting the clinical path allocation may also exist in the target diagnosis process record of the target patient, in this embodiment, at least one type of patient information may also be extracted from the target diagnosis process record of the target patient, wherein, as mentioned above, the target diagnosis process record is usually structured, so that the required patient information may be directly extracted from the target diagnosis process record.

In this way, at least one patient information of the target patient that may affect the clinical pathway assignment, which may include, but is not limited to, current symptoms, complaining symptoms, complications, historical diagnosis, occupation, performed surgery, used drugs, etc., is accessible and is not exhaustive.

In this embodiment, the admission condition that at least one patient information of the target patient meets may be found by referring to the respective admission condition of at least one clinical pathway, so as to determine the target clinical pathway. Where the admission conditions for different clinical pathways may not be exactly the same.

In an alternative implementation, a description file of at least one clinical pathway may be obtained; and respectively marking patient information required for accessing at least one clinical path from the description file of at least one clinical path as an access condition. In the foregoing, the clinical pathway has been disassembled from the description file corresponding to the clinical pathway, and here, the admission condition is continuously disassembled from the description file corresponding to the clinical pathway. Similarly, considering that the number of clinical paths is limited, in practical application, an artificial manner may be adopted to construct the admission condition corresponding to each of the at least one clinical path, and certainly, a machine learning manner may also be adopted to identify various admission factors and relationships between the admission factors from the description file of the at least one clinical path, and further to construct the admission condition of the at least one clinical path, which is not limited in this embodiment.

In this embodiment, the admission condition corresponding to at least one clinical path may be in the form of a decision tree, and of course, this embodiment is not limited to this, and other data forms may also be used to carry the admission condition here.

Taking the decision tree as an example, when a target clinical path adapted to a target diagnosis and treatment process of a target patient is determined, the decision trees corresponding to at least one clinical path can be respectively operated based on at least one patient information of the target patient, and if the decision tree capable of being executed exists, the clinical path corresponding to the executed decision tree is taken as the target clinical path, so that the target clinical path corresponding to the target patient is determined.

It should be noted that, as mentioned above, the target diagnosis and treatment process may be a diagnosis and treatment process being performed by a target patient, and in such a case, a problem that a clinical path to be adopted by the target patient changes due to a sudden disease change of the target patient may occur. This may ensure the rationality and real-time nature of the clinical pathway variation analysis. Of course, this embodiment does not limit this.

Therefore, in the embodiment, the target clinical path adapted to the target diagnosis and treatment process of the target patient can be automatically and accurately determined, and the path variation condition of the actual diagnosis and treatment path of the target patient in the target diagnosis and treatment process relative to the target clinical path is analyzed by taking the target clinical path as a reference, so that the real-time performance and the accuracy of the clinical path variation analysis are ensured.

In the above or following embodiments, the actual treatment path of the target patient in the target treatment process may be constructed in various ways.

In an alternative implementation, the specified data structure may comprise a sequence of data. Based on this, in the implementation manner, event representation data corresponding to at least one diagnosis and treatment event can be extracted from the target diagnosis and treatment process record; and sequencing the event representation data corresponding to at least one diagnosis and treatment event according to the time sequence to obtain an event representation data sequence as an actual diagnosis and treatment path.

The diagnosis and treatment event may refer to a relatively independent medical event occurring in the diagnosis and treatment process, for example, a CT examination or the like. Correspondingly, in this embodiment, event representation data corresponding to at least one diagnosis and treatment event can be extracted from the target diagnosis and treatment process record of the target patient.

The generation scheme of the event characterization data of the diagnosis and treatment event will be described below by taking the first diagnosis and treatment event as an example, and it should be understood that the first diagnosis and treatment event may be any one of at least one diagnosis and treatment event included in the target diagnosis and treatment process.

In the implementation mode, event description information corresponding to a first diagnosis and treatment event can be acquired from a target diagnosis and treatment process record; identifying at least one medical entity and an entity relationship between the at least one medical entity under the event description information; and according to the at least one diagnosis and treatment entity and the entity relationship between the at least one diagnosis and treatment entity, establishing event representation data corresponding to the first diagnosis and treatment event.

The medical entity can be any object related to a medical procedure, such as operation, medicine, examination, disease, body part, and the like.

As mentioned above, the target diagnosis and treatment process record is usually structured, and in practical applications, the target diagnosis and treatment process record is usually stored according to diagnosis and treatment events, in which case, the event description information corresponding to the first diagnosis and treatment event can be directly obtained from the target diagnosis and treatment process record, and the event description information obtained according to the event description information is also structured. On the basis, at least one diagnosis and treatment entity and the entity relationship between the at least one diagnosis and treatment entity can be identified from the structured event description information, so that event representation data corresponding to the first diagnosis and treatment event is constructed. For example, in a structured event description of a CT exam event, patient name is recorded under the object field, CT is recorded under the exam field, lung is recorded under the site field, and nodule is recorded under the symptom field, the clinical entity can be extracted from it: "patient", "CT", "lung", "nodule", the physical relationship between these several entities includes but is not limited to [ CT, exam, patient ], [ CT, site, lung ], [ CT, outcome, nodule ], [ patient, symptom, nodule ], etc.

Of course, the target diagnosis process record may also be unstructured, in which case at least one diagnosis event may be identified from the target diagnosis process record. An alternative diagnosis and treatment event identification scheme may be: and identifying an event description text corresponding to the first diagnosis and treatment event from the target diagnosis and treatment process record of the target patient by using a preset diagnosis and treatment event identification model as event description information. The diagnosis and treatment event identification model can adopt a machine learning model, and in practical application, a plurality of unstructured samples can be marked to represent the boundaries of diagnosis and treatment events in the unstructured samples; and inputting the unstructured sample into the diagnosis and treatment event recognition model so that the diagnosis and treatment event recognition model can learn diagnosis and treatment event demarcation knowledge, and thus, diagnosis and treatment events can be recognized in the target diagnosis and treatment process record by using the trained diagnosis and treatment event recognition model.

Then, under the first diagnosis and treatment event, at least one diagnosis and treatment entity and an entity relation between the at least one diagnosis and treatment entity can be extracted from the target diagnosis and treatment process record based on a Natural Language Processing (NLP) technology. In practical applications, a machine learning model can be trained for identifying clinical entities and extracting entity relationships from unstructured target clinical process records, and conventional schemes can be adopted for training and using the machine learning model, which will not be described in detail herein. Corresponding to the above example, if the event description information corresponding to one CT examination event is "CT examination is performed on a patient and a nodule is found in a lung", that is, a section of non-structural text, the diagnosis and treatment entities "patient", "CT", "lung", "nodule" and the entity relationship between these diagnosis and treatment entities can be extracted based on the NLP technology.

Therefore, at least one diagnosis and treatment entity and an entity relation corresponding to the first diagnosis and treatment event can be obtained, and accordingly, event representation data corresponding to the first diagnosis and treatment event can be constructed according to the at least one diagnosis and treatment entity and the entity relation between the at least one diagnosis and treatment entity.

Because the diagnosis and treatment events have a time sequence, in this embodiment, the event representation data corresponding to at least one diagnosis and treatment event in the target diagnosis and treatment process can be sorted according to the time sequence, so as to obtain an event representation data sequence of a time dimension. In this embodiment, the event representation data sequence may be used as an actual diagnosis path.

Accordingly, in this implementation, the actual clinical path may include at least one time phase, for example, the actual clinical path may include four time nodes, i.e., a pre-treatment phase, a mid-treatment phase, a post-treatment phase, and a rehabilitation phase, which is merely exemplary. And each time phase can contain at least one diagnosis and treatment event, so that the diagnosis and treatment events can be grouped according to the time phase and sorted among groups, and a data sequence formed by the event representation data corresponding to the at least one diagnosis and treatment event can be obtained.

In this implementation, the diagnosis events involved may also be ordered within a group within a single time period included in the actual diagnosis path, for example, randomly ordered, sorted by attribute, grouped again, or ordered by time, etc.

In an exemplary sort by attribute scheme: determining at least one attribute to which each of at least one target diagnosis and treatment event contained in a single time period belongs; and sequencing the event representation data corresponding to the at least one target diagnosis and treatment event according to a preset attribute level relation and at least one attribute to which the at least one target diagnosis and treatment event belongs so as to determine an event representation data subsequence in a single time stage. The at least one attribute of the at least one target diagnosis and treatment event contained in the single time period may be extracted from the target diagnosis and treatment process record, and of course, the at least one attribute of the at least one target diagnosis and treatment event contained in the single time period may also be determined in other manners such as manual tagging. In addition, the designated data structure may contain the attribute level relationship within a single time period, for example, three attributes, which may be necessary, optional, and general, may be configured in the designated data structure, and these three attributes may correspond to a plurality of subordinate attributes, such as examination, inspection, medicine, consumable, surgery, or other treatment, respectively. Thus, a single clinical event may belong to multiple attributes simultaneously, e.g., a CT examination event, may belong to both "required" and "examination" attributes simultaneously. On the basis, the event characterization data corresponding to the related diagnosis and treatment events can be sequenced in a single time phase according to the attribute level relation, so that the related diagnosis and treatment events can be sequenced in a group in the single time phase.

Accordingly, from the aspect of a single attribute, the single attribute may include event characterization data of at least one diagnosis and treatment event, which provides a basis for alignment of an actual diagnosis and treatment path and a target clinical path in a subsequent path variation analysis process, and based on this, alignment of dimensions such as a time phase and an attribute can be performed on the actual diagnosis and treatment path and the target clinical path, so as to ensure accuracy and rationality of comparison.

Of course, in this embodiment, other implementation manners may also be used to construct the actual diagnosis and treatment path of the target patient in the target diagnosis and treatment process, for example, other data structures such as a multi-level index are used to represent the diagnosis and treatment program in the target diagnosis and treatment process to construct the actual diagnosis and treatment path, which is not limited in this embodiment.

In the above or following embodiments, path variation information corresponding to each of a plurality of patients using a designated clinical path may also be obtained; and if the variation degree of the path variation information corresponding to the plurality of patients meets the preset condition, correcting the appointed clinical path according to the path variation information corresponding to the plurality of patients. Wherein the specified clinical pathway may be any of the at least one clinical pathway.

A plurality of patients using the designated clinical path usually have the same disease types, so that the rationality of the clinical path corresponding to a single disease type can be analyzed through batch path variation information under the single disease type, thereby being beneficial to discovering the dissatisfaction in the clinical path, continuously modifying and improving the existing labeled clinical path and promoting the normalization of diagnosis and treatment. For example, if the degree of variation corresponding to the path variation information of the patients with a ratio exceeding a preset ratio is higher than a preset threshold value for the disease type of lung cancer, it may be determined that there may be unreasonable items in the existing clinical paths for lung cancer, and the clinical paths for lung cancer may be corrected based on the path variation information of the patients to improve the diagnosis quality and the diagnosis cost for lung cancer.

Fig. 2a is a schematic flow chart of a method for analyzing path variation according to an exemplary embodiment of the present disclosure. Fig. 2b is a logic diagram of a path variation analysis scheme according to an exemplary embodiment of the present application. Referring to fig. 2a, the method may be performed by a path variation analysis apparatus, which may be implemented as a combination of software and/or hardware, which may be integrated in a computing device. Referring to fig. 2a, the method comprises:

step 200, responding to a variation analysis instruction, and extracting an actual processing path corresponding to a target event from a processing process record of the target event;

step 201, determining a standard processing path adapted to a target event;

step 202, based on the actual processing path and the standard processing path, performing a path variation analysis to determine path variation information under the target event.

The path variation analysis method provided by this embodiment may be applied to various scenarios that require management of an event processing process, for example, an education scenario, an application development or test scenario, a product development, test or release scenario, a product quality detection or industrial production scenario, and the like. In different application scenarios, the target events in this embodiment may not be identical. For example, in an educational scenario, the target event may be a teaching event. Of course, the application scenario may also be a medical scenario in the foregoing, and the target diagnosis and treatment process of the target patient in the medical scenario may correspond to the target event in this embodiment.

Referring to fig. 2a and 2b, in this embodiment, in step 200, an actual processing path corresponding to the target event may be extracted from the processing procedure record of the target event. In general, the processing of a target event may involve one or more processing elements that may constitute a processing path. In a medical scenario, one or more diagnosis and treatment events may correspond to a processing link in the present embodiment, and a diagnosis and treatment path may correspond to a processing path in the present embodiment.

In this embodiment, the actual processing path of the target event may be stored according to a specified data structure. To this end, in step 200, the processing procedure record of the target event may be structured according to a specified data structure, so as to extract an actual processing path from the processing procedure record of the target event. That is, various data required by the specified data structure, such as entities, entity relationships, entity attributes, and the like, may be extracted from the processing records of the target event, and the extracted data may be stored according to the specified data format to characterize the actual processing path.

In an optional implementation manner, the specified data structure may be a data sequence, and based on this, in this embodiment, link characterization data corresponding to at least one processing link may be extracted from the processing procedure record; and sequencing link representation data corresponding to at least one processing link according to the time sequence to obtain a link representation data sequence as an actual processing path. For different application scenarios, the processing links may not be completely divided in the same manner, for example, in a medical scenario, the processing links may correspond to diagnosis and treatment events, and in an application program development scenario, the processing links may correspond to development stages, and the like. In addition, under different application scenarios, the content, the internal structure, and the like included in the link characterization data may not be completely the same, and are not exhaustive here.

Of course, in this embodiment, the specified data structure is not limited to this, and a structure such as an array including a sequence identifier may also be adopted, which is not limited in this embodiment.

In step 201, a standard processing path adapted to the target event may be determined. Wherein the standard processing path may be an official or conventionally prescribed processing path. In this embodiment, the standard processing path may also be stored in a specified data structure. Typically, the standard processing path is presented in the form of a description file, such as text or pictures. For this reason, in the present embodiment, the description file of the standard processing path may be structured according to a specified data structure to extract the standard processing path from the description file of the standard processing path. In this way, the standard processing path and the actual processing path of the target event may be stored as a unified data structure.

In step 202, a path variation analysis may be performed based on the actual processing path and the standard processing path to determine path variation information under the target event. Based on the specified data structure, in this embodiment, the actual processing path and the standard processing may be aligned according to the specified data structure; and comparing the aligned actual processing path with the standard processing path to analyze the difference between the actual processing path and the standard processing path, wherein the difference is used as path variation information under the target event.

In addition, in this embodiment, the actual processing path and the standard processing path of the target event may be visually displayed, for example, in a form of a knowledge graph, which is not limited in this embodiment. The path variation information at the target time may also be visually displayed, for example, in the form of animation or icon, which is not limited in this embodiment.

In summary, in the embodiment, the actual processing path can be automatically extracted from the processing procedure record of the target event as the basis for the path mutation analysis, so that the data value of the processing procedure record can be fully exerted, and the path mutation of the actual processing path relative to the standard processing path can be quickly and accurately analyzed, thereby implementing the path mutation analysis. Moreover, the actual treatment path and the standard treatment path can be stored according to a uniform data structure, so that the comparison between the actual treatment path and the standard treatment path can be better supported, and the efficiency and the accuracy of the path variation analysis are improved. The technical scheme provided by the embodiment can be used for analyzing whether the completed processing process or the completed processing link in the ongoing processing process meets the processing specification or not, and provides support for implementation management of the standard path.

It should be noted that, for the sake of brevity, many technical details of the path variance analysis method provided in this embodiment are not described in detail. These technical details may be obtained by referring to the detailed description of the embodiments of clinical pathway variation analysis described above, or by adaptively extending the clinical pathway variation analysis scheme to other application scenarios in the embodiments, but this should not be construed as a loss of the scope of the present application.

Fig. 3 is a flowchart illustrating a diagnosis and treatment path guidance method according to another exemplary embodiment of the present application, and fig. 4 is a logic diagram illustrating a diagnosis and treatment path guidance scheme according to another exemplary embodiment of the present application. Referring to fig. 3, the method may be performed by a clinical pathway guidance apparatus, which may be implemented as a combination of software and/or hardware, which may be integrated in a computing device. Referring to fig. 3, the method includes:

step 300, responding to the path guide instruction, and extracting a historical diagnosis and treatment path of the target patient from the medical history record of the current diagnosis and treatment process of the target patient;

301, acquiring diagnosis and treatment paths corresponding to at least one reference patient;

step 302, determining a specific patient from at least one reference patient, wherein the diagnosis and treatment path corresponding to the specific patient is matched with the historical diagnosis and treatment record of the target patient;

and 303, guiding the diagnosis and treatment path of the target patient in the current diagnosis and treatment process according to the diagnosis and treatment path of the specific patient.

The diagnosis and treatment path guide scheme provided by the embodiment can be used for providing diagnosis and treatment path guide for the current diagnosis and treatment process of the target patient. The method is particularly suitable for disease types with unpublished clinical routes, and of course, for disease types with clinical routes, the diagnosis and treatment route guidance scheme provided by the embodiment can be adopted to supplement or modify the existing clinical routes.

In step 300, a history diagnosis path of the target patient may be extracted from the medical history record of the target patient. In this embodiment, the historical diagnosis and treatment path may be stored according to a specified data structure, and for this reason, in step 300, the medical history record of the current diagnosis and treatment process of the target patient may be structured according to the specified data structure, so as to extract the historical diagnosis and treatment path of the target patient from the medical history record. That is, various data required by a specified data structure, such as diagnosis and treatment entities, entity relationships, entity attributes and the like, are extracted from the medical history record of the target patient, and the extracted data are stored according to a specified data format to represent the historical diagnosis and treatment path. The historical diagnosis and treatment path may include diagnosis and treatment programs historically adopted by the target patient, and the diagnosis and treatment programs in the historical diagnosis and treatment path may cover a plurality of historical diagnosis and treatment processes.

In this embodiment, the target patient may be any patient that is present at a medical facility. The medical history of the target patient may refer to the historical medical information of the target patient before the current medical procedure. For example, the target patient's disease history, surgical history, examination history, allergy history, medication history, medical order history, and the like. Typically, the medical history of the target patient is unstructured. The medical file is generated in the current medical treatment process of the target patient, and the medical file at least comprises two parts: history records and current diagnosis and treatment process records. With the development of digital medical management, the current medical procedure record in the medical record is usually structured, and the medical history record is usually based on the text of the patient's dictation record, or medical record images downloaded from other hospitals, and so on, so the medical history record is usually unstructured. In contrast, in this embodiment, according to the specified data structure, based on the natural language processing NLP technology, various path data required for constructing a path, such as medical entities and entity relationships, are extracted from the medical history record of the target patient, so as to construct a historical medical path of the target patient. After NLP is carried out, the extracted data can be normalized, so that different characteristics of the same entity are normalized into the same characteristic, and the accuracy of a historical diagnosis and treatment path is improved. Optionally, in this embodiment, the normalization model may be trained through labeled samples of several clinical entities, so as to perform a normalization operation using the normalization model, where the normalization model may be a machine learning model. It should be understood that the historical clinical path of the target patient is a set of data having a specified data structure.

In addition, in this embodiment, if the target patient has completed a part of the diagnosis and treatment procedure in the current diagnosis and treatment process, the current diagnosis and treatment path may be extracted from the current diagnosis and treatment process record of the target patient, and the current diagnosis and treatment path of the target patient is added to the historical diagnosis and treatment path. Of course, this embodiment is not limited thereto. The current diagnosis and treatment process record of the target patient is usually structured, so that various path data required for constructing paths can be directly extracted from the current diagnosis and treatment process record to construct the current diagnosis and treatment path.

In this embodiment, the reference patient may be any historical patient, the reference patient may be a historical patient who has been visited by the medical institution where the target patient is located, and of course, the reference patient may also be a historical patient who has been visited by another medical institution and whose visiting file can be acquired by the medical institution where the target patient is located. Referring to fig. 3 and 4, in step 301, a diagnosis path corresponding to each of at least one reference patient may be obtained. In practical application, the disease category information of the disease category determined by the target patient in the current diagnosis and treatment process can be determined, and the patient which accords with the disease category information corresponding to the target patient is screened from at least one historical patient to serve as a reference patient, so that the magnitude order of the reference patient can be preliminarily reduced, wherein the disease category of the target patient can be one or more. Of course, this embodiment is not limited thereto.

In this embodiment, the diagnosis and treatment path corresponding to the reference patient can be extracted from the reference patient hospitalization file according to the designated data format. Generally, there are a plurality of reference medical records of patients, in this embodiment, the reference medical records of patients can be comprehensively referred to determine the diagnosis and treatment path of the patients, and the specific scheme will be detailed in the following embodiments. In this way, the historical treatment path of the target patient and the treatment path of the reference patient are consistent in data format.

In step 302, a comparison algorithm may be used to quickly screen out a specific patient similar to the target patient from at least one reference patient as a similar case of the target patient. And calculating the similarity between the historical diagnosis and treatment path of the target patient and the diagnosis and treatment path of at least one reference patient by adopting a comparison algorithm, and determining the reference patient with the similarity meeting the specified requirement as the characteristic patient. The dimension of the alignment may be varied, and may include, but is not limited to, the type of examination included in the path, the surgical procedure employed, the type of disease diagnosed or the drugs used, etc. In practical applications, the similarity may be determined by combining comparison results of multiple dimensions, and of course, the embodiment is not limited thereto.

Different from the clinical pathway variation analysis scheme, the present embodiment finds similar diagnosis and treatment pathways by comparison, and the clinical pathway variation analysis scheme finds variation of the diagnosis and treatment pathways relative to the clinical pathways by comparison. In addition, the comparison path is longer, the time span is larger, and historical diagnosis and treatment programs of patients are involved, so that the specific patient corresponding to the target patient can be matched in at least one reference patient quickly and accurately.

Referring to fig. 3 and 4, in step 303, a diagnosis path of a target patient during a current diagnosis process may be guided according to a diagnosis path of a specific patient. As mentioned above, the current diagnosis process of the target patient is not yet finished, and the diagnosis path of the specific patient can provide guidance for the subsequent diagnosis scheme of the target patient.

In this embodiment, the diagnosis and treatment path of the specific patient and the historical diagnosis and treatment path of the target patient may be visually displayed, optionally, the diagnosis and treatment path of the specific patient and the historical diagnosis and treatment path of the target patient may be visually displayed in the form of a knowledge graph, a time axis line, a chart or the like, and the visualization display form of the diagnosis and treatment path of the specific patient and the historical diagnosis and treatment path of the target patient is not limited in this embodiment. Through the visual display of the diagnosis and treatment path of the specific patient and the historical diagnosis and treatment path of the target patient, the similar situation between the diagnosis and treatment path of the specific patient and the historical diagnosis and treatment path of the target patient can be briefly and clearly presented, and medical workers can determine the subsequent diagnosis and treatment program of the target patient in the current diagnosis and treatment process by referring to the diagnosis and treatment path of the specific patient.

Accordingly, this embodiment can be used to find similar cases. The historical diagnosis and treatment path of the target patient can be extracted from the medical history record of the target patient, and the diagnosis and treatment path of at least one reference patient can be respectively extracted from the hospitalization file of the at least one reference patient; based on the method, the specific patient corresponding to the target patient can be quickly and accurately screened out to serve as the similar case of the target patient. Therefore, the decision reference of the diagnosis and treatment program can be provided for the current diagnosis and treatment process of the target patient based on the diagnosis and treatment path of the specific patient, and therefore the diagnosis and treatment quality of the current diagnosis and treatment process of the target patient can be improved. Moreover, the historical diagnosis and treatment path of the target patient and the diagnosis and treatment path of the reference patient can adopt a unified data structure, so that comparison between the diagnosis and treatment paths can be better supported, and a specific patient can be found more quickly and accurately.

In the above or below embodiments, the diagnosis and treatment path corresponding to each of the at least one reference patient may be determined based on the medical record corresponding to each of the at least one reference patient. The following will describe a scheme for determining a diagnosis and treatment path corresponding to a first reference patient by taking the first reference patient as an example, and it should be understood that the first reference patient may be any one of at least one reference patient.

Fig. 5a is a schematic logic diagram of a scenario for determining a treatment path of a reference patient according to another exemplary embodiment of the present application. Referring to fig. 5a, in this embodiment, a basic diagnosis and treatment path may be extracted from the last hospitalization file of the first reference patient according to a specified data structure; extracting at least one diagnosis and treatment data conflicting with the basic diagnosis and treatment path from other hospitalizing files of the first reference patient; and based on the diagnosis and treatment data, correcting the basic diagnosis and treatment path to obtain a diagnosis and treatment path corresponding to the first reference patient.

As mentioned above, the first reference patient may have a plurality of medical records. In this embodiment, the basic diagnosis and treatment path can be constructed by taking the last hospitalization file of the first reference patient as a main basis.

The basic diagnosis and treatment path construction scheme can be as follows:

according to a specified data structure, extracting a first diagnosis and treatment path for representing a historical diagnosis and treatment program from a medical history record in a last hospitalization file of a first reference patient;

according to a specified data structure, extracting a second diagnosis and treatment path for representing a current diagnosis and treatment program from a current diagnosis and treatment process record in a latest hospitalizing file of a first reference patient;

and integrating the first diagnosis and treatment path and the second diagnosis and treatment path to obtain a basic diagnosis and treatment path corresponding to the first reference patient.

Compared with other hospitalizing archives, the information in the medical history record in the last hospitalizing archive is most abundant and comprehensive, so that the first diagnosis and treatment path determined based on the medical history record in the last hospitalizing archive can basically comprehensively and accurately represent the historical diagnosis and treatment program of the first reference patient. However, as mentioned above, the medical history is usually generated by the patient dictating, which results in the accuracy and comprehensiveness of the part of the medical history in the last visit record being still insufficient. Therefore, in the embodiment, in the process of correcting the basic diagnosis and treatment path of the first reference patient, the first diagnosis and treatment path included in the basic diagnosis and treatment path of the first reference patient may be mainly corrected.

Referring to fig. 5a, an alternative modification may be: extracting at least one diagnosis and treatment data with content conflict with the medical history record in the last medical record of the first reference patient from other medical records of the first reference patient; and based on the diagnosis and treatment data, correcting the first diagnosis and treatment path of the first reference patient so as to correct the basic diagnosis and treatment path. Referring to fig. 5a, in practical applications, only the current medical procedure record (usually structured and accurate) in other medical records of the first reference patient may be analyzed, various medical data may be extracted therefrom, and various path data extracted from the medical history record in the last medical record of the first reference patient for constructing the first medical path may be fused and disambiguated to correct the path data, so as to construct the first medical path of the first reference patient by using the corrected path data, which may effectively improve the accuracy and comprehensiveness of the first medical path.

Referring to fig. 5a, in this embodiment, according to the specified data structure, the second clinical path for characterizing the current clinical procedure may be extracted from the current clinical process record in the last visit archive of the first reference patient. In this embodiment, various path data required for constructing the second diagnosis and treatment path can be directly extracted from the current diagnosis and treatment process record in the last visit file of the first reference patient to construct the second diagnosis and treatment path.

In addition, referring to fig. 5a, in this embodiment, the first clinical path and the second clinical path may be spliced to obtain the clinical path of the first reference patient, but the integration manner of the first clinical path and the second clinical path is not limited thereto, and a manner of reordering clinical events according to time may also be adopted, which is not limited in this embodiment.

Of course, in this embodiment, the implementation manner of determining the diagnosis and treatment path of the reference patient is not limited thereto, and other implementation manners may be adopted to determine the diagnosis and treatment path of the reference patient, for example, the current diagnosis and treatment path is extracted from the multiple-time medical record of the reference patient including the current diagnosis and treatment process record, at least one current diagnosis and treatment path is integrated to obtain the diagnosis and treatment path of the reference patient, and the medical history record in the medical record is no longer used as a basis, and the like.

Therefore, in the embodiment, the diagnosis and treatment path of the reference patient can be determined more accurately and comprehensively, and an accurate comparison basis is provided for searching for the specific patient.

In the above or the following embodiments, the process related to the diagnosis and treatment path extraction includes, but is not limited to: extracting the historical diagnosis and treatment path of the target patient, extracting the first diagnosis and treatment path of the reference patient, extracting the second diagnosis and treatment path of the reference patient and the like. Fig. 5b is a logic diagram of a diagnosis path extraction scheme according to an exemplary embodiment of the present application, and the diagnosis path extraction scheme will be described below with reference to fig. 5b by taking the example of extracting a historical diagnosis path of a target patient.

In this embodiment, the specified data structure includes a data sequence in a time dimension.

Based on this, in this embodiment, event representation data corresponding to at least one diagnosis and treatment event can be extracted from the medical history record of the target patient; and sequencing the event representation data corresponding to at least one diagnosis and treatment event according to the time sequence to obtain an event representation data sequence as a historical diagnosis and treatment path of the target patient.

In the process of extracting event representation data corresponding to at least one diagnosis and treatment event from the medical history record of the target patient, acquiring event description information corresponding to a first diagnosis and treatment event from the medical history record of the target patient; identifying at least one medical entity and an entity relationship between the at least one medical entity under the event description information; according to at least one diagnosis and treatment entity and an entity relation between the at least one diagnosis and treatment entity, event representation data corresponding to a first diagnosis and treatment event are constructed; wherein the first clinical event is any one of the at least one clinical event.

In this embodiment, the history of the target patient may or may not be structured. In the case of structuring, the structured storage is usually performed in units of clinical events, and therefore, at least one clinical event can be directly determined. Whereas for the unstructured case, referring to fig. 5b, a delineation of the clinical events may be performed first, thereby identifying at least one clinical event. An alternative diagnosis and treatment event identification scheme may be: and identifying an event description text corresponding to the first diagnosis and treatment event from the medical history record of the target patient by using a preset diagnosis and treatment event identification model as event description information. The diagnosis and treatment event identification model can adopt a machine learning model, and in practical application, a plurality of unstructured samples can be marked to represent the boundaries of diagnosis and treatment events in the unstructured samples; and inputting the unstructured sample into the diagnosis and treatment event recognition model so that the diagnosis and treatment event recognition model can learn diagnosis and treatment event demarcation knowledge, and therefore, the trained diagnosis and treatment event recognition model can be used for recognizing the diagnosis and treatment event in the medical history record of the target patient. Then, under a single diagnosis and treatment event, various path data required for constructing the path, such as diagnosis and treatment entities, entity relations and the like, are extracted by using the NLP.

The extraction process of the event characterization data may refer to the relevant descriptions in the embodiments of the clinical pathway variation analysis method, and may be adaptively adjusted according to the application scenario, the processing object, and the like. For brevity, no further description is provided herein. In addition, the above-mentioned other diagnosis and treatment path extraction processes may refer to a process of extracting a historical diagnosis and treatment path of a target patient, and are not described in detail herein.

Therefore, in the embodiment, the diagnosis and treatment path conforming to the specified data structure can be extracted quickly and accurately in the extraction process of each diagnosis and treatment path related in the diagnosis and treatment path guide scheme.

In the above or following embodiments, the at least one patient to be analyzed that matches the designated patient may be selected from the at least one reference patient and the target patient; clustering at least one patient to be analyzed based on a diagnosis and treatment path corresponding to the patient to be analyzed to obtain at least one patient group; acquiring diagnosis and treatment effect information corresponding to at least one patient group; and comparing the diagnosis and treatment effect information corresponding to at least one patient group to determine the diagnosis and treatment effect difference of different diagnosis and treatment paths under the appointed disease category.

The scheme provided by the embodiment can be used for subject research of a single disease, a plurality of patients matched with the specified disease are used as the patients to be analyzed, the patients to be analyzed can be divided into a plurality of patient groups through clustering of diagnosis and treatment paths, the diagnosis and treatment paths of the patients to be analyzed in a single patient group are similar, and equivalently, the diagnosis and treatment schemes used by the patients to be analyzed in the single patient group are similar. Therefore, the patient samples of different diagnosis and treatment schemes can be obtained for a single disease, diagnosis and treatment effect evaluation can be performed on the different diagnosis and treatment schemes, diagnosis and treatment programs of the single disease can be optimized, and a basis is provided for diagnosis and treatment research of the single disease.

It should be noted that the execution subjects of the steps of the methods provided in the above embodiments may be the same device, or different devices may be used as the execution subjects of the methods. For example, the execution subjects of steps 101 to 103 may be device a; for another example, the execution subject of

steps

101 and 102 may be device a, and the execution subject of step 103 may be device B; and so on.

In addition, in some of the flows described in the above embodiments and the drawings, a plurality of operations are included in a specific order, but it should be clearly understood that the operations may be executed out of the order presented herein or in parallel, and the sequence numbers of the operations, such as 101, 102, etc., are merely used for distinguishing different operations, and the sequence numbers do not represent any execution order per se. Additionally, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first", "second", and the like in this document are used to distinguish different application terminals, messages, modules, and the like, and do not represent a sequence, and do not limit that "first" and "second" are different types.

Fig. 6 is a schematic structural diagram of a computing device according to yet another exemplary embodiment of the present application. As shown in fig. 6, the computing device includes: a memory 60 and a processor 61.

A processor 61, coupled to the memory 60, for executing computer programs in the memory 60 for:

responding to the variation analysis instruction, and extracting an actual diagnosis and treatment path of the target patient from a target diagnosis and treatment process record of the target patient;

determining a target clinical path matched with a target diagnosis and treatment process of a target patient from at least one clinical path;

In an alternative embodiment, the processor 61, when extracting the actual treatment path of the target patient from the target treatment process record of the target patient, is configured to:

and according to the appointed data structure, carrying out structural processing on the target diagnosis and treatment process record so as to extract the actual diagnosis and treatment path of the target patient from the target diagnosis and treatment process record.

In an alternative embodiment, the specified data structure is a data sequence, and the processor 61, when performing a structured processing on the target clinical process record according to the specified data structure to generate the actual clinical path of the target patient, is configured to:

extracting event representation data corresponding to at least one diagnosis and treatment event from a target diagnosis and treatment process record;

and sequencing the event representation data corresponding to at least one diagnosis and treatment event according to the time sequence to obtain an event representation data sequence as an actual diagnosis and treatment path.

In an optional embodiment, when extracting the event characterization data corresponding to each of the at least one diagnosis and treatment event from the target diagnosis and treatment process record, the processor 61 is configured to:

acquiring event description information corresponding to a first diagnosis and treatment event from a target diagnosis and treatment process record;

identifying at least one medical entity and an entity relationship between the at least one medical entity under the event description information;

according to at least one diagnosis and treatment entity and an entity relation between the at least one diagnosis and treatment entity, event representation data corresponding to a first diagnosis and treatment event are constructed;

wherein the first clinical event is any one of the at least one clinical event.

In an alternative embodiment, the actual clinical path includes at least one time phase, and the processor 61 is further configured to:

extracting at least one attribute to which at least one target diagnosis and treatment event contained in a target time phase belongs from the target diagnosis and treatment process record;

and sequencing the event representation data corresponding to the at least one target diagnosis and treatment event according to a preset attribute level relation and at least one attribute to which the at least one target diagnosis and treatment event belongs so as to determine an event representation data subsequence in a target time phase.

In an alternative embodiment, the processor 61, when determining the target clinical path adapted to the target clinical procedure of the target patient from the at least one clinical path, is configured to:

acquiring a medical history record of a target patient;

identifying at least one patient information affecting the clinical pathway assignment from the target medical procedure record and/or the medical history record;

determining a target clinical pathway from the at least one clinical pathway according to the at least one patient information, wherein the at least one patient information meets the admission condition of the target clinical pathway.

In an alternative embodiment, the processor 61 is further configured to:

obtaining a description file of at least one clinical pathway;

and respectively marking patient information required for accessing at least one clinical path from the description file of at least one clinical path as an access condition.

In an alternative embodiment, the processor 61 is further configured to:

obtaining a description file of at least one clinical pathway;

and according to a specified data format, structuring the description file of the at least one clinical path respectively to extract the at least one clinical path from the description file of the at least one clinical path.

In an alternative embodiment, the processor 61, when performing the mutation analysis based on the actual clinical path and the clinical path, is configured to:

aligning the actual diagnosis and treatment path with the clinical path according to a specified data structure;

and comparing the aligned actual diagnosis and treatment path with the clinical path to analyze the difference between the actual diagnosis and treatment path and the clinical path.

In an alternative embodiment, the processor 61 is further configured to:

and calculating the cost variation information of the target patient in the target diagnosis and treatment process according to the path variation information of the target patient in the target diagnosis and treatment process.

In an alternative embodiment, the path variation information includes one or more of variation nature, variation controllability, variation source and variation factor; the cost variation information comprises one or more of consumable cost, traditional Chinese medicine cost, western medicine cost, image inspection cost and laboratory inspection cost.

In an alternative embodiment, the processor 61 is further configured to:

and visually displaying the path variation information.

In an alternative embodiment, the processor 61 is further configured to:

acquiring path variation information corresponding to a plurality of patients using a specified clinical path;

and if the variation degree of the path variation information corresponding to the plurality of patients meets the preset condition, correcting the appointed clinical path according to the path variation information corresponding to the plurality of patients.

In an alternative embodiment, the processor 61 is further configured to:

and constructing the actual diagnosis and treatment path and the target clinical path into a knowledge graph for displaying.

Further, as shown in fig. 6, the computing device further includes: communication components 62, power components 63, and the like. Only some of the components are schematically shown in fig. 6, and the computing device is not meant to include only the components shown in fig. 6.

It should be noted that, for the sake of brevity, the above description of the technical details of the embodiments of the computing apparatus may be referred to in the related descriptions of the embodiments of the clinical pathway variation analysis method, which should not be repeated herein, but should not cause a loss of the scope of the present application.

Accordingly, the present application further provides a computer-readable storage medium storing a computer program, where the computer program can implement the steps that can be executed by a computing device in the foregoing method embodiments when executed.

Fig. 7 is a schematic structural diagram of another computing device according to yet another exemplary embodiment of the present application. As shown in fig. 7, the computing device includes: a memory 70 and a processor 71.

A processor 71, coupled to the memory 70, for executing computer programs in the memory 70 for:

responding to the variation analysis instruction, and extracting an actual processing path corresponding to the target event from the processing process record of the target event;

determining a standard processing path adapted to the target event;

In an alternative embodiment, when the processor 71 extracts the actual processing path corresponding to the target event from the processing procedure record of the target event, it is configured to:

and according to the specified data structure, performing structuring processing on the processing procedure record to extract an actual processing path from the processing procedure record.

In an alternative embodiment, the specified data structure is a data sequence, and the processor 71, when performing the structuring process on the process record according to the specified data structure to extract the actual processing path from the process record, is configured to:

extracting link representation data corresponding to at least one processing link from the processing process record;

and sequencing link representation data corresponding to at least one processing link according to the time sequence to obtain a link representation data sequence as an actual processing path.

In an alternative embodiment, processor 71 is further configured to:

acquiring a description file of a standard processing path;

and according to the specified data format, carrying out structuring processing on the description file of the standard processing path so as to extract the standard processing path from the description file of the standard processing path.

In an alternative embodiment, the processor 71 performs a path variance analysis based on the actual processing path and the standard processing path, including:

aligning the actual processing path with the standard processing according to the specified data structure;

and comparing the aligned actual processing path with the standard processing path to analyze the difference between the actual processing path and the standard processing path.

Further, as shown in fig. 7, the computing device further includes: communication components 72, power components 73, and the like. Only some of the components are schematically shown in fig. 7, and the computing device is not meant to include only the components shown in fig. 7.

It should be noted that, for the technical details in the embodiments of the computing device, reference may be made to the related descriptions in the embodiments of the path variation analysis method, which are not described herein for brevity, but this should not cause a loss of the scope of the present application.

Fig. 8 is a schematic structural diagram of another computing device according to another exemplary embodiment of the present application. As shown in fig. 8, the computing device includes: a memory 80 and a processor 81.

A processor 81, coupled to the memory 80, for executing the computer program in the memory 80 for:

responding to the path guide instruction, and extracting a historical diagnosis and treatment path of the target patient from a medical history record of the current diagnosis and treatment process of the target patient;

determining a specific patient from at least one reference patient, wherein the diagnosis and treatment path corresponding to the specific patient is matched with the historical diagnosis and treatment record of the target patient;

according to the diagnosis and treatment path of the specific patient, the diagnosis and treatment path of the target patient in the current diagnosis and treatment process is guided.

In an alternative embodiment, the processor 81, when extracting the historical clinical path of the target patient from the medical history of the current clinical process of the target patient, is configured to:

and according to the specified data structure, performing structured processing on the medical history record of the current diagnosis and treatment process of the target patient so as to extract the historical diagnosis and treatment path of the target patient from the medical history record.

In an alternative embodiment, the specified data structure is a data sequence, and the processor 81 is configured to, when performing a structured processing on the medical history record of the current diagnosis and treatment process of the target patient according to the specified data structure, so as to extract the historical diagnosis and treatment path of the target patient from the medical history record:

extracting event representation data corresponding to at least one diagnosis and treatment event from a medical history record of the current diagnosis and treatment process of a target patient;

and sequencing the event representation data corresponding to at least one diagnosis and treatment event according to the time sequence to obtain an event representation data sequence as a historical diagnosis and treatment path of the target patient.

In an optional embodiment, the processor 81, when extracting event representation data corresponding to at least one diagnosis and treatment event from the medical history of the current diagnosis and treatment process of the target patient, is configured to:

acquiring event description information corresponding to a first diagnosis and treatment event from a medical history record of a current diagnosis and treatment process of a target patient;

wherein the first clinical event is any one of the at least one clinical event.

In an optional embodiment, when the processor 81 acquires event description information corresponding to a first medical event from a medical history of a current medical procedure of a target patient, the processor is configured to:

and if the medical history record is unstructured, identifying an event description text corresponding to the first diagnosis and treatment event from the medical history record of the target patient by using a preset diagnosis and treatment identification model as event description information.

In an optional embodiment, the processor 81, when obtaining the treatment path corresponding to each of the at least one reference patient, is configured to:

extracting a basic diagnosis and treatment path from a latest diagnosis and treatment file of a first reference patient according to a specified data structure, wherein the diagnosis and treatment file comprises a medical history record and a current diagnosis and treatment process record;

extracting at least one diagnosis and treatment data conflicting with the basic diagnosis and treatment path from other hospitalizing files of the first reference patient;

and based on the diagnosis and treatment data, correcting the basic diagnosis and treatment path to obtain a diagnosis and treatment path corresponding to the first reference patient.

In an alternative embodiment, the processor 81, when extracting the base clinical path from the last visit file of the first reference patient according to the specified data structure, is configured to:

In an alternative embodiment, processor 81 is further configured to:

selecting at least one patient to be analyzed matched with the specified disease species from at least one reference patient and the target patient;

clustering at least one patient to be analyzed based on a diagnosis and treatment path corresponding to the patient to be analyzed to obtain at least one patient group;

acquiring diagnosis and treatment effect information corresponding to at least one patient group;

and comparing the diagnosis and treatment effect information corresponding to at least one patient group to determine the diagnosis and treatment effect difference of different diagnosis and treatment paths under the appointed disease category.

Further, as shown in fig. 8, the computing device further includes: communication components 82, power components 83, and the like. Only some of the components are schematically shown in fig. 8, and the computing device is not meant to include only the components shown in fig. 8.

It should be noted that, for the technical details in the embodiments of the computing device, reference may be made to the related descriptions in the embodiments of the diagnosis and treatment path guiding method, which are not described herein for brevity, but this should not cause a loss of the scope of the present application.

The memory of fig. 6-8 described above is used to store computer programs and may be configured to store various other data to support operations on the computing platform. Examples of such data include instructions for any application or method operating on the computing platform, contact data, phonebook data, messages, pictures, videos, and so forth. The memory may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

The communication components of fig. 6-8 described above are configured to facilitate wired or wireless communication between the device in which the communication component is located and other devices. The device where the communication component is located can access a wireless network based on a communication standard, such as a WiFi, a 2G, 3G, 4G/LTE, 5G and other mobile communication networks, or a combination thereof. In an exemplary embodiment, the communication component receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

The power supply components of fig. 6-8 described above provide power to the various components of the device in which the power supply components are located. The power components may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device in which the power component is located.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for clinical pathway variant analysis, comprising:

2. The method of claim 1, wherein the extracting the actual clinical path of the target patient from the target clinical process record of the target patient comprises:

and according to a specified data structure, carrying out structural processing on the target diagnosis and treatment process record so as to extract an actual diagnosis and treatment path of the target patient from the target diagnosis and treatment process record.

3. The method of claim 2, wherein the specified data structure is a data sequence, and the step of structuring the target procedure record according to the specified data structure to generate the actual treatment path of the target patient comprises:

extracting event representation data corresponding to at least one diagnosis and treatment event from the target diagnosis and treatment process record;

and sequencing the event representation data corresponding to the at least one diagnosis and treatment event according to the time sequence to obtain an event representation data sequence as the actual diagnosis and treatment path.

4. The method according to claim 3, wherein the extracting event characterization data corresponding to each of the at least one clinical event from the target clinical process record comprises:

acquiring event description information corresponding to a first diagnosis and treatment event from the target diagnosis and treatment process record;

identifying at least one clinical entity and an entity relationship between the at least one clinical entity under the event description information;

according to the at least one diagnosis and treatment entity and the entity relationship between the at least one diagnosis and treatment entity, establishing event representation data corresponding to the first diagnosis and treatment event;

wherein the first clinical event is any one of the at least one clinical event.

5. The method of claim 3, wherein the actual clinical path includes at least one time period, the method further comprising:

and according to a preset attribute level relation and at least one attribute to which each of the at least one target diagnosis and treatment event belongs, sequencing event characterization data corresponding to each of the at least one target diagnosis and treatment event to determine an event characterization data subsequence in the target time phase.

6. The method according to claim 1, wherein the determining a target clinical path adapted to the target clinical procedure of the target patient from the at least one clinical path comprises:

acquiring a medical history record of the target patient;

identifying at least one patient information affecting clinical pathway assignment from the target medical procedure record and/or the medical history record;

determining the target clinical pathway from the at least one clinical pathway according to the at least one patient information, wherein the at least one patient information meets the admission condition of the target clinical pathway.

7. The method of claim 6, further comprising:

obtaining a description file of the at least one clinical pathway;

and respectively marking patient information required for accessing the at least one clinical path from the description file of the at least one clinical path as the access condition.

8. The method of claim 2, further comprising:

obtaining a description file of the at least one clinical pathway;

and according to the specified data format, structuring the description file of the at least one clinical path respectively to extract the at least one clinical path from the description file of the at least one clinical path.

9. The method of claim 8, wherein performing a mutation analysis based on the actual clinical pathway and the clinical pathway comprises:

aligning the actual diagnosis and treatment path and the clinical path according to the specified data structure;

comparing the aligned actual diagnosis path with the clinical path to analyze the difference between the actual diagnosis path and the clinical path.

10. The method of claim 1, further comprising:

and calculating cost variation information of the target patient in the target diagnosis and treatment process according to the path variation information of the target patient in the target diagnosis and treatment process.

11. The method of claim 10, wherein the path mutation information comprises one or more of mutation properties, mutation controllability, mutation sources, and mutation factors; the cost variation information comprises one or more of consumable material cost, traditional Chinese medicine cost, western medicine cost, image inspection cost and laboratory inspection cost.

12. The method of claim 1, further comprising:

and visually displaying the path variation information.

13. The method of claim 1, further comprising:

and if the variation degree of the path variation information corresponding to the plurality of patients meets a preset condition, correcting the appointed clinical path according to the path variation information corresponding to the plurality of patients.

14. The method of claim 1, further comprising:

15. A diagnosis and treatment path guiding method is characterized by comprising the following steps:

16. The method of claim 15, wherein the extracting the historical clinical path of the target patient from the historical record of the current clinical process of the target patient comprises:

and according to a specified data structure, carrying out structural processing on the medical history record of the current diagnosis and treatment process of the target patient so as to extract the historical diagnosis and treatment path of the target patient from the medical history record.

17. The method according to claim 16, wherein the specified data structure adopts a data sequence, and the step of structuring the medical history record of the current medical procedure of the target patient according to the specified data structure to extract the historical medical path of the target patient from the medical history record comprises:

extracting event representation data corresponding to at least one diagnosis and treatment event from the medical history record of the current diagnosis and treatment process of the target patient;

and sequencing the event representation data corresponding to the at least one diagnosis and treatment event according to the time sequence to obtain an event representation data sequence as a historical diagnosis and treatment path of the target patient.

18. The method according to claim 17, wherein the extracting event characterization data corresponding to at least one clinical event from the medical history of the current clinical process of the target patient comprises:

acquiring event description information corresponding to a first diagnosis and treatment event from a medical history record of the current diagnosis and treatment process of the target patient;

wherein the first clinical event is any one of the at least one clinical event.

19. The method according to claim 18, wherein the obtaining event description information corresponding to a first clinical event from the medical history of the current clinical process of the target patient comprises:

and if the medical history record is unstructured, identifying an event description text corresponding to a first diagnosis and treatment event from the medical history record of the target patient by using a preset diagnosis and treatment identification model as the event description information.

20. The method according to claim 15, wherein the obtaining of the diagnosis and treatment path corresponding to each of the at least one reference patient comprises:

extracting at least one diagnosis and treatment data conflicting with the basic diagnosis and treatment path from other medical files of the first reference patient;

and correcting the basic diagnosis and treatment path based on the diagnosis and treatment data to obtain a diagnosis and treatment path corresponding to the first reference patient.

21. The method of claim 20, wherein said extracting a base clinical path from the first reference patient's most recent visit archive according to the specified data structure comprises:

according to the appointed data structure, extracting a first diagnosis and treatment path for representing a historical diagnosis and treatment program from a medical history record in the last hospitalization file of the first reference patient;

according to the appointed data structure, extracting a second diagnosis and treatment path for representing a current diagnosis and treatment program from a current diagnosis and treatment process record in the latest hospitalizing file of the first reference patient;

22. The method of claim 15, further comprising:

selecting at least one patient to be analyzed matched with the specified disease species from the at least one reference patient and the target patient;

clustering the at least one patient to be analyzed based on the diagnosis and treatment path corresponding to the at least one patient to be analyzed to obtain at least one patient group;

acquiring diagnosis and treatment effect information corresponding to the at least one patient group;

and comparing the diagnosis and treatment effect information corresponding to the at least one patient group to determine the diagnosis and treatment effect difference of different diagnosis and treatment paths under the specified disease category.

23. A method for analyzing path variation, comprising:

determining a standard processing path adapted to the target event;

24. The method according to claim 23, wherein the extracting an actual processing path corresponding to a target event from a processing record of the target event comprises:

and according to a specified data structure, carrying out structuring processing on the processing procedure record so as to extract the actual processing path from the processing procedure record.

25. The method of claim 24, wherein the specified data structure is in a data sequence, and wherein structuring the process record according to the specified data structure to extract the actual processing path from the process record comprises:

and sequencing the link representation data corresponding to the at least one processing link according to the time sequence to obtain a link representation data sequence as the actual processing path.

26. The method of claim 24, further comprising:

obtaining a description file of the standard processing path;

27. The method of claim 26, wherein performing a path variation analysis based on the actual processing path and the standard processing path comprises:

comparing the aligned actual processing path with the standard processing path to analyze the difference between the actual processing path and the standard processing path.

28. A computing device comprising a memory and a processor;

the memory is to store one or more computer instructions;

29. A computing device comprising a memory and a processor;

the memory is to store one or more computer instructions;

30. A computing device comprising a memory and a processor;

the memory is to store one or more computer instructions;

determining a standard processing path adapted to the target event;

31. A computer-readable storage medium storing computer instructions which, when executed by one or more processors, cause the one or more processors to perform the clinical pathway variation analysis method of any one of claims 1-14, the clinical pathway guidance method of any one of claims 15-22, or the pathway variation analysis method of any one of claims 23-27.

32. A computer program product comprising computer programs/instructions, wherein the computer programs/instructions, when executed by a processor, cause the processor to implement the clinical pathway variation analysis method of any one of claims 1-14, the clinical pathway guidance method of any one of claims 15-22, or the pathway variation analysis method of any one of claims 23-27.