Round method and system based on medical advice execution plan drive

Abstract

The invention belongs to the technical field of medical information, and provides a doctor's advice execution plan driven ward round method and system, and the technical scheme is as follows: before ward round, obtaining historical medical record data based on doctor's advice execution plan driven; obtaining a to-be-executed medical technology project based on the historical medical record data driven by the medical advice execution plan and the constructed medical technology project pushing model; in ward round, according to the current diagnosis information of the patient, in combination with the called target data and the constructed doctor's advice recommendation model, an optimal doctor's advice recommendation scheme is obtained; and after ward round, associating the made medical advice recommendation scheme to a ward round record.

Description

Technical Field

[0001] The present invention belongs to the field of medical information technology, and in particular relates to a ward-round method and system driven by a doctor's order execution plan. Background Art

[0002] The statements in this part only provide background technical information related to the present invention, and do not necessarily constitute prior art.

[0003] In the initial stage of hospital informatization construction, the transformation from traditional paper-based management to informatization has been completed. The digital transformation of hospitals is an important trend in the future medical service field of hospitals. In this process of hospital digital transformation, it is an inevitable choice to meet the increasingly complex and diverse needs of medical services.

[0004] The traditional ward-round method has the following problems:

[0005] 1. Most of the previous ward-round methods construct and assemble the patient's condition according to the information system method, but there is a lack of active linkage between systems. Doctors cannot automatically extract effective information during ward rounds, resulting in doctors being unable to timely understand the changes in the patient's condition and affecting the efficiency of ward rounds.

[0006] 2. Lack the ability to automatically analyze the condition.

[0007] 3. The participation of patients or their family members is low, which is likely to cause the omission of important information on the changes in the patient's condition and is not conducive to treatment. Summary of the Invention

[0008] In order to solve at least one of the above technical problems in the background art, the present invention provides a ward-round method and system driven by a doctor's order execution plan. This method and system collect, organize, and analyze the changes in the patient's condition according to structured data, and automatically classify, store, and retrieve them, and automatically refine and report them. At the same time, it is convenient for doctors to view the patient's historical medical record information at any time, providing a reference for formulating treatment plans, so that doctors have more time and energy to focus on the patient's treatment and psychological state. Through faster and more accurate condition analysis and treatment suggestions, patients can obtain better treatment effects, thereby improving the satisfaction with the hospital. To achieve the above object, the present invention adopts the following technical solutions:

[0009] The first aspect of the present invention provides a ward-round method driven by a doctor's order execution plan, including the following steps:

[0010] Before the ward round, obtain historical medical record data driven by the doctor's order execution plan;

[0011] Based on the historical medical record data driven by the doctor's order execution plan and the constructed medical technology item push model, obtain the medical technology items to be executed; wherein, the construction process of the medical technology item push model includes:

[0012] Build corresponding data analysis models around different specialties and diagnoses of patients, combine historical medical record data, construct a standardized symptom rule engine for patients, match the suspected diagnosis with the standardized symptom rule engine, infer the suspected diagnosis result, and push corresponding medical technology items according to the suspected diagnosis result;

[0013] Construct a data integration model by combining historical medical record data, medical technology item data, doctor's order information, and multi-dimensional unified standards, and call target data according to the integration model;

[0014] During the ward round, call the target data according to the patient's request information, and combine the called target data with the constructed doctor's order plan recommendation model to obtain the recommended optimal doctor's order plan;

[0015] After the ward round, associate the issued doctor's order recommendation plan with the ward round record.

[0016] Furthermore, the obtaining of historical medical record data driven by the doctor's order execution plan includes:

[0017] Based on the admission diagnosis and the patient's condition changes, divide the patient's admission period into different time stages, obtain the ward round data of different time stages, and group them;

[0018] Conduct a matching degree analysis on the ward round data and medical record data of different time stages, sort the analysis results, and convert the sorted data into a speech dictionary;

[0019] Match and associate the speech dictionary with the nursing doctor's order. When issuing a nursing doctor's order, convert the corresponding speech dictionary into a plan drive to obtain the corresponding daily nursing content filled in according to the speech.

[0020] Furthermore, the building of corresponding data analysis models around different specialties and diagnoses of patients includes:

[0021] According to the specialty, divide the parts involved in the specialty in combination with the human body structure diagram, consider the onset time and duration of symptoms, inducing factors, number of attacks, exacerbation time, and presence or absence of accompanying symptoms, and combine the accompanying symptoms with relevant disease characteristics according to the specialty. At the same time, combine single symptoms and accompanying symptoms to infer multiple symptoms, and then take the differences between another symptom and the previous symptom.

[0022] Furthermore, the construction of a data integration model by combining historical medical record data, medical technology item data, doctor's order information, and multi-dimensional unified standards includes:

[0023] Break up the sub-nodes of historical medical record data, doctor's orders, and report content, and based on the established multi-dimensional unified model, establish a new model for the fragmented information, connect them according to the business, and schedule the corresponding target data according to the corresponding business items.

[0024] Further, the historical medical record data includes the current medical history, past medical history, physical examination, and auxiliary examinations in the admission record; the case characteristics, preliminary diagnosis, diagnosis basis, differential diagnosis, and treatment plan in the first course record; and the changes in physical signs, mental status, and symptoms of the corresponding parts of the department specialty in the doctor's ward round record.

[0025] The historical medical order information includes the usage of antibiotic drugs for symptom changes, whether dosage adjustment is required or the drug needs to be changed according to the condition changes, and whether there are corresponding medical orders, examinations, and medical technology orders for the corresponding symptoms.

[0026] The report content corresponding to the medical technology project data includes the key value situation and abnormal value situation in the report.

[0027] Further, the construction process of the medical order plan recommendation model includes:

[0028] Call the target data according to the requested patient information text.

[0029] Establish a vector database, group and store all the patient medical record information of the corresponding patient in different departments in the vector library, mark, encode, and sort them, and establish a unique index around the content.

[0030] Embed the target data into the vector database, perform retrieval through the cosine similarity calculation method, and obtain the top N ward round records with the highest matching degree for the current patient.

[0031] Group and sort the recommended medical orders such as drugs, inspection items, and examination items according to the matching results to obtain the medical order recommendation result.

[0032] Further, after obtaining the optimal medical order recommendation plan, judge the type of the medical order recommendation result. If it is a drug medical order, automatically calculate the corresponding dosage, usage, frequency, and whether it is a long-term plan according to the drug attributes. If it is a non-drug medical order, match the pinyin abbreviation with the medical order keyword, establish a corresponding model, and obtain the keyword corresponding to the non-drug medical order.

[0033] The second aspect of the present invention provides a ward round system driven by a medical order execution plan, including:

[0034] A pre-ward round module, which is configured to: obtain historical medical record data driven by a medical order execution plan;

[0035] Obtain the medical technology projects to be executed based on the historical medical record data driven by the medical order execution plan and the constructed medical technology project push model; wherein, the construction process of the medical technology project push model includes:

[0036] Build corresponding data analysis models around different specialties and diagnoses of patients, combine historical medical record data, construct a standardized symptom rule engine for patients, match the suspected diagnosis with the standardized symptom rule engine, infer the suspected diagnosis result, and push corresponding medical technology items according to the suspected diagnosis result;

[0037] Construct a data integration model by combining historical medical record data, medical technology item data, doctor's order information, and multi-dimensional unified standards, and call target data according to the integration model;

[0038] The in-rounds module is configured to:

[0039] Call target data according to the patient's request information, and combine the called target data with the constructed doctor's order plan recommendation model to obtain the recommended optimal doctor's order plan;

[0040] The post-rounds module is configured to: associate the recommended doctor's order plan to the in-rounds record.

[0041] The third aspect of the present invention provides a computer-readable storage medium.

[0042] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps in a rounds method driven by a doctor's order execution plan as described above.

[0043] The fourth aspect of the present invention provides a computer device.

[0044] A computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in a rounds method driven by a doctor's order execution plan as described above.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] 1. The present invention collects, processes, and uses all information around the patient before, during, and after the doctor's rounds. It collects, organizes, and analyzes the changes in the patient's condition according to structured data, automatically classifies, stores, and retrieves it, and automatically extracts and reports it. At the same time, it is convenient for doctors to view the patient's historical medical record information at any time, providing a reference for formulating treatment plans, enabling doctors to have more time and energy to focus on the patient's treatment and psychological state. Through faster and more accurate condition analysis and treatment suggestions, patients can obtain better treatment effects, thereby enhancing their satisfaction with the hospital.

[0047] 2. The present invention realizes that patients or their family members can independently complete the filling of daily care content by means of doctor's order execution plan drive, changing the patient's passive reception to the active and highly involved participation of patients or their family members, and enhancing the patient's perception of medical treatment ability.

[0048] 3. The present invention takes doctors as units. After reaching an order of magnitude, the data is compared with the departments corresponding to this specialty to deduce the selection of medical technology projects at the department level. In this way, it can not only ensure the personalization of doctors' diagnosis and treatment and the construction of department-standardized treatment plans, thereby greatly saving the cumbersome cycle of doctors' ward rounds and the confirmation of treatment plans, but also improve the accuracy of treatment plans better by relying on data, and establish a good data foundation for the hospital to train new personnel.

[0049] 4. Aiming at the problem of lagging medical orders inferred due to the relatively serious lag of current symptoms and signs, the drug medical orders and auxiliary examination medical orders are analyzed and inferred through enhanced retrieval algorithms and statistical analysis algorithms. At the same time, different characteristic treatments are carried out for different types of medical orders, which helps to improve the accuracy of issuing medical orders and the efficiency of ward rounds.

[0050] Advantages of additional aspects of the present invention will be partly given in the following description, partly will become obvious from the following description, or will be understood through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings forming a part of this specification are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention.

[0052] Figure 1 is a flowchart of a ward round method driven by a medical order execution plan provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0054] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further descriptions of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0055] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form "illustrative diagram" includes the plural form. In addition, it should also be understood that when the terms "include" and / or "comprise" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0056] Embodiment 1

[0057] As Figure 1As shown in the figure, this embodiment provides a ward-round method driven by a medical order execution plan, including the following steps:

[0058] Step 1: Obtain historical medical record data driven by a medical order execution plan;

[0059] Among them, in Step 1, the historical medical record data driven by the medical order execution plan specifically includes:

[0060] Step 101: Based on the admission diagnosis and the patient's condition changes, divide the patient's admission period into different time stages, obtain the ward-round data of different time stages, and group them;

[0061] In this embodiment, the ward-round data includes the situations that medical staff often ask the patient or the patient's family members during the ward-round, such as body temperature, diet, stool, urine, etc. Grouping is carried out in combination with the patient's physical characteristics, including intake and output, etc.

[0062] Step 102: Conduct a matching degree analysis on the ward-round data and medical record data of different time stages, sort the analysis results, and convert the sorted data into a speech dictionary;

[0063] Combined with the doctor's medical orders issued at different time periods and the data analysis of the patient's condition changes and physical characteristics in medical records such as the admission record and ward-round record, analyze the points with high matching degree, sort them, and then convert this data into a speech dictionary that the patient or the patient's family members can understand.

[0064] Step 103: Match and associate the speech dictionary with the nursing medical orders. When a nursing medical order is issued, convert the corresponding speech dictionary into a plan-driven one, and obtain the corresponding daily nursing content filled according to the speech;

[0065] Step 104: Compare the corresponding daily nursing content filled according to the speech with the set alarm line. If it exceeds the alarm line, dispatch corresponding handling measures; among them, the set alarm line and the corresponding early warning handling measures are included in the knowledge base. When including them in the knowledge base, add corresponding input parameters, including age, gender, main admission diagnosis, threshold content, and early warning handling measures for inclusion and sorting. At the same time, optimize the sorting rules in combination with the usage situation of the patient or the patient's family members, so that the patient or the patient's family members can have more efficient and accurate measures to handle when the patient's physical signs change exceeds the early warning.

[0066] Before each ward-round by the doctor, the patient or the patient's family members fill in the content according to the plan, and then through the MQ message, automatically push the content to the doctor to remind the doctor which patients have registered the patient's physical signs and condition, and at the same time pay special attention to the high-warning information that exceeds the threshold. In this way, the doctor can understand the situation in time, and instead of repeating mechanical inquiries during the ward-round, and can also register according to this information to help the doctor make another treatment plan earlier.

[0067] Based on the basic information about the patient's physical signs obtained through the above medical staff-patient ward rounds, a corresponding threshold warning knowledge base can be established, and the doctor's order plan can drive the patient to participate in the treatment. Instead of only regarding oneself as a patient, the medical staff and the patient are fully tied together, and there will be no delay in treatment due to a large number of patients during the ward rounds, uneven distribution of time, and mutual waiting.

[0068] Step 2: Based on the historical medical record data driven by the doctor's order execution plan and the constructed medical technology item push model, obtain the medical technology items to be executed; wherein, the construction process of the medical technology item push model includes:

[0069] Establish corresponding data analysis models around different specialties and diagnoses of the patient, combine the patient's medical record data, construct a patient standardized symptom rule engine, infer suspected diagnoses based on standard symptoms, and push corresponding medical technology items according to the suspected diagnosis results;

[0070] In Step 2, divide the departments into professional groups, analyze the process of doctors in the group asking the patient's symptoms during the ward rounds, and find that the process of asking symptoms and the results obtained are similar. Therefore, establish a rule engine for the process of doctors asking the patient's symptoms and transforming from fuzzy symptoms to standardized symptoms. The rule engine is established based on the detailed process of different doctors asking the patient's symptoms in the same specialty, and the entire process is sorted step by step.

[0071] Specifically, the specific construction process of the medical technology item push model includes:

[0072] Step 201: Establish corresponding standardized data analysis models around different specialties and diagnoses of the patient, specifically including:

[0073] According to the specialty, divide the parts involved in the specialty in combination with the human body structure diagram, and then establish the onset time and duration of symptoms, inducing factors, number of attacks, time of aggravation, presence or absence of accompanying symptoms, and take relevant onset characteristics of the accompanying symptoms in combination with the specialty. At the same time, combine single symptoms and accompanying symptoms to infer multiple symptoms, and then take the differences between another symptom and the previous symptom, so as to realize the construction of multiple diagnoses and recommendations for multiple patients;

[0074] Step 202: Based on the constructed standardized analysis data model, combine the patient's medical record data to construct a patient standardized symptom rule engine;

[0075] Record the specific content of the doctor's asking process, and at the same time combine the clinical guidelines of this specialty as a criterion for judgment, and record it after the number of times of using the basis in each asking in the asking content exceeds a calculated value, and then sort the content;

[0076] In this embodiment, only the first 5 digits of data are retained after sorting. To ensure that the newly entered data can be pushed to the front, the newly entered data is extended and 5 digits of data are retained. In this way, there are 10 digits in a group. In this way, the process of doctors asking about symptoms is continuously optimized based on the standardized symptom engine. Efficiency and accuracy of inquiry;

[0077] Step 203: Match the suspected diagnosis with the optimized patient standardized symptom rule engine to infer the suspected diagnosis result;

[0078] Standardized symptoms are associated with the suspected diagnoses selected by doctors. Symptoms selected by doctors of the same specialty based on the patients are recorded, and the corresponding suspected diagnoses are recorded. Records exceeding an agreed value (the value is adjusted based on subsequent data analysis) are included in the database. After that, each time a value is included, it is judged against the previous value, and the values ​​that exceed are moved to the front. As the data is included in the database, the top 20 data in the database are compared and optimized with other hospitals of similar size in the region on a hospital basis, so as to optimize the suspected diagnoses corresponding to the standard diagnoses. At the same time, this part of the data is organized by the medical department and then released to departments of the same specialty for use as the new hospital goes online.

[0079] Step 204: combining the suspected diagnosis result and the medical technology items associated with the suspected diagnosis result to obtain corresponding medical technology items to be selected;

[0080] In combination with the inferred suspected diagnosis, the medical technology projects such as examinations, tests, electrocardiograms, etc. that are frequently used in this specialty are established in a similar process to the suspected diagnosis establishment process. The difference is that in this embodiment, the doctor is first taken as the unit, and after reaching an order of magnitude, the data is compared with the department corresponding to this specialty to infer the department-level medical technology project selection. This can not only ensure the personalization of the doctor's diagnosis and treatment and the construction of standardized treatment plans in the department, thereby greatly saving the doctor's ward rounds and the tedious cycle of treatment plan confirmation, but also better improve the accuracy of the treatment plan by speaking with data, and establish a good data foundation for the hospital to train new people.

[0081] Before making rounds, doctors need to obtain a large amount of patient information, including the patient's admission time, medical record information on changes in condition, including current medical history, past medical history, physical examination in the admission record, and changes in condition in the medical history record, combined with the content of the medical technical report, including key values, abnormal values ​​and other information corresponding to the doctor's attention report, as well as medical order information; at the same time, the key changes in the patient's condition must be marked so that targeted inquiries can be made during rounds.

[0082] In the past, doctors needed to log in to different systems to view relevant information about patients. For example, they needed to check the medical records in the EMR system, check the doctor's orders in the HIS, and check the medical reports in the LIS and PACS systems. In this embodiment, a patient-centered, cloud-health integrated digital ward design concept is adopted to share data such as medical records, doctor's orders, and medical reports in real time, and the services are interconnected. Doctors can view the patient's outpatient medical records, inpatient electronic medical records, and examination and test results and reports from different dimensions to achieve data interconnection.

[0083] Step 3: Combine historical medical record data, medical technology project data, doctor's order information and multi-dimensional unified standards to build a data integration model, and call the target data according to the integration model;

[0084] Break up the sub-nodes of historical medical records, doctor's orders, reports, etc., and build a new model for the fragmented information based on the established multi-dimensional unified model. Then, connect the business in series and dispatch the corresponding target data according to the corresponding business projects.

[0085] By obtaining the scheduling target data through the data integration model, doctors can schedule the desired content before ward rounds. At the same time, the model changes according to changes in the patient's condition. In this way, doctors can form a line around the patient's condition and focus points during ward rounds, providing doctors with the data they need at any time.

[0086] Among them, the medical record content in the historical data includes the current medical history, past history, physical examination, and auxiliary examinations in the admission record; the case characteristics, preliminary diagnosis, diagnostic basis, differential diagnosis, and treatment plan in the first medical record; and the changes in physical signs, mental state, and symptoms in the corresponding parts of the department in the physician's ward rounds record.

[0087] The content of medical orders in historical data includes the use of antibiotics in response to changes in symptoms, whether the dosage needs to be adjusted or the medicine needs to be changed according to changes in the condition, and whether medical orders, examinations, tests and other medical technical orders corresponding to the corresponding symptoms have been issued.

[0088] The historical data of TCM technical instructions correspond to the report content, including key values ​​and abnormal values ​​in the report. The so-called key values ​​are associated and maintained with the test values ​​based on the profession.

[0089] Among them, the multi-dimensional unified standard model includes: unified hospital departments, unified hospital medical staff, unified standard dictionary, unified service dictionary, unified patient index, and unified resources. The specific contents are as follows:

[0090] Unification of hospital departments: The ability to link business operations across different institutions is the basis for achieving business collaboration.

[0091] Unification of hospital medical staff: Establishing a unified master index for hospital staff, hospital staff working in different institutions can achieve information association of hospital staff, and ultimately provide important data basis for the hospital's refined personnel management.

[0092] Standard dictionary unification: The standard dictionaries used by the hospital are unified. Everyone can refer to the required dictionary according to their needs, and can also adjust the personalized variable attributes of some dictionaries.

[0093] Unification of service dictionaries: Based on the unified hospital, department, personnel and standard dictionary described above, medical services can be unified and medical collaboration between hospital departments can be facilitated.

[0094] The patient index is unified. For patients whose files are created in the unified cloud health system, a unique primary key index will be established for the patient, which unifies the patient's entry and realizes the uniqueness of patient information from a business perspective.

[0095] Resource unification: Establishing a unified resource index for medicines, materials, equipment, etc. used in the hospital can open up resource channels between different departments and establish resource unification.

[0096] After segmenting the historical data, doctors divide the data into different stages according to their specialties and the time of admission of patients. At each stage, they accumulate and mark the contents of interest according to the business process, set keywords, and then sort them. In this way, the data with high usage of tags is pushed out. Then, the keywords and voice are adapted and trained. In this way, the desired content can be quickly found and read according to the keywords during the ward rounds. At the same time, after the model is established, the work before the ward rounds can be optimized to the ward rounds, which can save doctors time in the preparation stage before the ward rounds.

[0097] By passing in a piece of text from the corresponding medical record of the patient to be queried, the similarity with several texts in the vector database is calculated to find the corresponding medical orders. In the past, medical orders for items such as drugs and auxiliary examinations were inferred by combining symptoms, signs, and information such as age, diagnosis, gender, surgery, and auxiliary examinations and tests. However, after analysis, it was found that the symptoms and signs currently provided are relatively seriously lagging, resulting in a lag in the medical orders inferred from the query.

[0098] In this embodiment, enhanced retrieval algorithms and statistical analysis algorithms are used to analyze and infer drug orders and auxiliary examination orders, making it easier for doctors to prescribe them.

[0099] Step 4: Call the target data according to the patient's request information, and combine the called target data with the constructed medical advice plan recommendation model to obtain the recommended optimal medical advice plan;

[0100] The present invention collects the patient's condition in real time, and then uses digital splitting and combining operations to help doctors analyze changes in the patient's condition, thereby forming auxiliary recommended treatment measures to assist doctors in making medical decisions. At the same time, by establishing a digital link, the information channel between doctors and patients is opened up, and a doctor management assistant is provided to help patients answer their condition and relieve their emotions, while also reporting the patient's condition to the doctor in a timely manner.

[0101] In step 4, the target data is called according to the patient's request information, and the optimal recommended medical advice plan is obtained by combining the called target data with the constructed medical advice plan recommendation model, which specifically includes:

[0102] Step 401: calling target data according to the requested patient information text; specifically including patient gender, age, chief complaint, and admission diagnosis information;

[0103] Step 402: Establish a vector database with N dimensions, adjust it according to the query accuracy, and collect all the patient medical records of different departments corresponding to the corresponding patients, mainly collecting admission records, first course of illness records, third-level physician rounds records, etc., and put them into the vector database for group storage, and mark the coding sorting to establish a unique index around the content;

[0104] Step 402: embed the target data into the vector database, and perform retrieval through the cosine similarity calculation method to obtain the top N ward round records that match the current patient;

[0105] Among them, first match the most recent ward round record, whether it is the director, attending physician, or resident doctor, and query the 100 results with the highest matching degree in the entire hospital through the vector database. As follows: Enhance the search to increase the weight analysis, for example, Test_List{visit record: 0.3; diagnosis: 0.3; gender: 0.2; age: 0.2};

[0106] Step 403: grouping and sorting the corresponding medical orders of the recommended drugs, test items, examination items, etc. according to the matching results to obtain medical order recommendation results;

[0107] Step 404: determine the type of the doctor's order recommendation result. If it is a drug order, automatically calculate the corresponding dosage, usage, frequency and whether it is a long-term plan based on the drug attributes;

[0108] During the ward rounds, the doctor will inquire about changes in the patient's condition that he or she is concerned about based on the preparations made before the rounds. During the inquiry process, most of the content is recorded in paper form, which is divided into medication orders, including the drug name, dosage, frequency, and whether it is long-term; and medical technical orders, such as lung CT, head CTA, blood tests, urine insertion, routine ECG, dynamic ECG and other information; functional orders, such as discharge tomorrow and taking medication when discharged; and entrusting orders, such as asking the patient to drink more water, pay attention to turning over, observe changes in body temperature, etc., as well as whether a consultation application is needed.

[0109] By analyzing the medication orders in the paper ward rounds records collected by doctors, we found that the drugs recorded by doctors were all abbreviations, either the brand name or the first four names of the drugs, which is very similar to the pinyin codes used by doctors when writing orders in the EMR system. In this way, the pinyin codes entered in the ERM and the finally selected drugs are used to establish a matching association model, which divides the groups into groups based on professions, and the pinyin codes can be matched to the keywords in the drugs. Subsequently, the keywords are used as the basis of the model for continuous training and sorting. At the same time, horizontal dimensions are established in the model to correspond to the drug's attributes such as dosage, usage, frequency, and whether it is long-term.

[0110] Step 4041: First, the drug dictionary is used as the basic data for training, and then compared with the historical data for analysis. If the corresponding drug attributes under the specialty change, the corresponding rankings of the respective attributes are established according to the corresponding attribute classifications of the drugs;

[0111] Step 4042: The drug keywords are then matched with the voice, and the drug is found through the voice keywords. The corresponding dosage, usage, frequency and whether it is long-term are automatically calculated in combination with the drug attribute recommendation, unless the doctor adds dosage, frequency or usage or whether it is long-term medical order attributes when speaking the drug keyword, and then the voice broadcast is made. Whether to save and sign is confirmed based on the doctor's voice broadcast.

[0112] Step 4043: The accuracy of the speech recognition keywords is also recorded in the system for marking, and the speech recognition keywords are also included in the model vocabulary. Finally, the prescription information is automatically synchronized to the medical record.

[0113] If the medical order recommendation result is a non-drug order, it will be matched with the medical order keyword through the pinyin code, just like drugs, and a corresponding model will be established. It will be continuously trained and then adapted to the voice to facilitate doctors to directly read out the recommended keywords.

[0114] Step 405: During the ward round, the doctor conducts a physical examination based on the changes in the patient's condition, establishes a keyword library for physical examination, includes the keywords for physical examination, compares the keywords with professional clinical guidelines to check if they meet the guideline requirements, can initiate a consultation in the corresponding department, calculates the consultation department if it meets the requirements, combines with the voice assistant, and notifies the doctor through voice broadcast whether a consultation application needs to be initiated for the department accepting the corresponding symptoms. At the same time, a multi-disciplinary consultation is initiated based on the amount of symptom content, and the doctor is allowed to select and confirm according to the urgency of the consultation application.

[0115] Step 5: After the ward round, in the past, the doctor used to write medical orders in the system based on the content on the record sheet, and then restore the scene through the medical orders to recall the changes in the patient's condition, and then complete the writing of the ward round record.

[0116] In this embodiment, based on the medical orders and the physical examination keywords, the symptom information corresponding to the included keywords is analyzed during the doctor's communication, and then the symptom information is included in the ward round record, so as to avoid the doctor supplementing the medical care and medical records later.

[0117] Embodiment Two

[0118] This embodiment provides a ward round system driven by a medical order execution plan, including:

[0119] A pre-ward round module, which is configured to: obtain historical medical record data driven by a medical order execution plan;

[0120] Based on the historical medical record data driven by the medical order execution plan and the constructed medical technology project push model, obtain the medical technology projects to be executed; wherein, the construction process of the medical technology project push model includes:

[0121] Establish corresponding data analysis models around different specialties and diagnoses of the patient, combine with historical medical record data, construct a standardized symptom rule engine for the patient, match the suspected diagnosis with the standardized symptom rule engine, infer the suspected diagnosis result, and push the corresponding medical technology projects according to the suspected diagnosis result;

[0122] Construct a data integration model by combining historical medical record data, medical technology project data, medical order information and multi-dimensional unified standards, and call the target data according to the integration model;

[0123] A mid-ward round module, which is configured to:

[0124] Call the target data according to the patient's request information, and combine the called target data with the constructed medical order plan recommendation model to obtain the recommended optimal medical order plan;

[0125] A post-ward round module, which is configured to: associate the recommended medical order plan with the ward round record.

[0126] Embodiment Three

[0127] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps in a ward-round method driven by a doctor's order execution plan as described above.

[0128] Embodiment 4

[0129] This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in a ward-round method driven by a doctor's order execution plan as described above.

[0130] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories and optical memories, etc.) containing computer-usable program code.

[0131] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0132] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means, and the instruction means implements the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0133] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the functions in the flowFigure 1 one process or multiple processes and / or boxes Figure 1 steps of functions specified in one box or multiple boxes.

[0134] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0135] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A ward round method based on doctor's order execution plan drive, characterized in that: The steps include: Before ward rounds, obtain historical medical record data driven by the doctor's order execution plan; Based on the historical medical record data driven by the medical order execution plan and the constructed medical technology project push model, the medical technology project to be executed is obtained; wherein the construction process of the medical technology project push model includes: Establish corresponding data analysis models around different specialties and diagnoses of patients, build a standardized symptom rule engine for patients based on historical medical records, match suspected diagnoses with the standardized symptom rule engine, infer suspected diagnosis results, and push corresponding medical technology projects based on suspected diagnosis results; Combine historical medical record data, medical technology project data, doctor's order information and multi-dimensional unified standards to build a data integration model, and call the target data according to the integration model; During the ward rounds, the target data is called according to the patient's request information, and the optimal recommended medical order plan is obtained by combining the called target data with the constructed medical order plan recommendation model; After the ward round, the recommended plan of medical orders issued will be linked to the ward round record.

2. A ward round method based on doctor's order execution plan drive as claimed in claim 1, characterized in that: The acquisition of historical medical record data driven by the doctor's order execution plan includes: Based on the admission diagnosis and the patient's condition changes, the patient's admission period is divided into different time stages, and the ward rounds data of different time stages are obtained and grouped; Perform matching analysis on ward rounds data and medical records data at different time stages, sort the analysis results, and convert the sorted data into a speech dictionary; Match and associate the speech dictionary with the nursing instructions. When issuing nursing instructions, convert the corresponding speech dictionary into a plan-driven program to obtain the corresponding daily care content filled in according to the speech.

3. A ward round method based on doctor's order execution plan drive as claimed in claim 1, characterized in that: The corresponding data analysis models are established around different specialties and different diagnoses of patients, including: Based on the specialty, the parts involved in the specialty are divided in combination with the human body structure diagram, taking into account the onset time and duration of symptoms, inducing factors, number of onsets, aggravation time, and the presence or absence of accompanying symptoms. In combination with the specialty, the accompanying symptoms are associated with relevant onset characteristics. At the same time, multiple symptoms are deduce by combining a single symptom and accompanying symptoms, and then taking the difference between another symptom and the previous symptom.

4. A ward round method based on doctor's order execution plan drive as claimed in claim 1, characterized in that: The data integration model is constructed by combining historical medical record data, medical technology project data, doctor's order information and multi-dimensional unified standards, including: Break up the sub-nodes of historical medical records, medical orders and report contents, and build a new model for the fragmented information based on the established multi-dimensional unified model. Connect the businesses in series and schedule the corresponding target data according to the corresponding business projects.

5. A ward round method based on doctor's order execution plan drive as claimed in claim 1, characterized in that: The historical medical record data include the current medical history, past medical history, physical examination, and auxiliary examinations in the admission record; the case characteristics, preliminary diagnosis, diagnostic basis, differential diagnosis, and treatment plan in the first course of illness record; and the changes in physical signs, mental status, and symptoms of the corresponding parts of the department in the doctor's ward rounds record; Historical medical order information includes the use of antibiotics in response to symptom changes, whether the dosage needs to be adjusted or the medicine needs to be changed according to the change in the condition, and whether medical orders, examinations, and technical orders corresponding to the corresponding symptoms have been issued; The medical technology project data corresponds to the report content, including the key values ​​and abnormal values ​​in the report.

6. A ward round method based on doctor's order execution plan drive as claimed in claim 1, characterized in that: The construction process of the medical advice plan recommendation model includes: Calling target data according to the text requesting patient information; Establish a vector database, group and store all the patient records of different departments corresponding to the patients in the vector database, mark and code them, and establish a unique index around the content; The target data is embedded into the vector database, and the cosine similarity calculation method is used to retrieve the top N ward round records that match the current patient. Based on the matching results, the recommended drugs, test items, examination items and other corresponding medical orders are grouped and sorted to obtain the medical order recommendation results.

7. A ward round method based on doctor's order execution plan drive as claimed in claim 1, characterized in that: After obtaining the optimal medical order recommendation plan, determine the type of medical order recommendation result. If it is a drug order, automatically calculate the corresponding dosage, usage, frequency and whether it is a long-term plan based on the drug attributes. If it is a non-drug order, match the pinyin code with the medical order keyword, establish a corresponding model, and obtain the keyword corresponding to the non-drug order.

8. A ward rounds system driven by a doctor's order execution plan, characterized in that: include: A pre-ward rounds module is configured to: obtain historical medical record data driven by a doctor's order execution plan; Based on the historical medical record data driven by the medical order execution plan and the constructed medical technology project push model, the medical technology project to be executed is obtained; wherein the construction process of the medical technology project push model includes: Establish corresponding data analysis models around different specialties and diagnoses of patients, build a standardized symptom rule engine for patients based on historical medical records, match suspected diagnoses with the standardized symptom rule engine, infer suspected diagnosis results, and push corresponding medical technology projects based on suspected diagnosis results; Combine historical medical record data, medical technology project data, doctor's order information and multi-dimensional unified standards to build a data integration model, and call the target data according to the integration model; The ward rounds module is configured as follows: Calling target data according to the patient's request information, combining the called target data with the constructed medical advice plan recommendation model, and obtaining the recommended optimal medical advice plan; The post-ward round module is configured to associate the issued medical order recommendation plan with the ward round record.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in a ward round method driven by a doctor's order execution plan as described in any one of claims 1 to 7 are implemented.

10. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the ward round method driven by a doctor's order execution plan as described in any one of claims 1 to 7 are implemented.

Images