Infection nursing recording method, device and system based on intelligent medical treatment

Abstract

The invention relates to the technical field of medical information, in particular to an infection nursing recording method, device and system based on intelligent medical treatment, and the method comprises the steps: obtaining an infection nursing record corresponding to a target patient according to a database and identity information, and generating a data sample set; performing comparative analysis on different data of certain infection nursing in the data sample set, and positioning to obtain an abnormal record data sample; obtaining the change of the abnormal record data sample in time sequence and the dynamic change of the infection time interval of the target patient, and predicting the abnormal record data sample to obtain the recovery trend of the target patient; supplementing the remaining data by combining the relevance between the data dimensions and / or vocabularies with the data sample set and the recovery trend, and recovering the abnormal record data sample to obtain a complete infection nursing record data sample; storing the complete infection nursing record data sample into a database; and the repair result of the abnormal infection nursing record data sample is ensured to be more reliable and real.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to an infection nursing record method, device and system based on smart medical care. Background Art

[0002] With the rapid advancement of information technology, smart medical technology has been fully integrated into the entire medical process from prevention, diagnosis, treatment, nursing to rehabilitation. By applying cutting-edge information technology, this technology not only optimizes the allocation of medical resources and improves the quality and efficiency of medical services, but also realizes the intelligent management of medical processes. In particular, in the process of infection care, smart medical technology can record key data such as the patient's physical condition, nurses' nursing operations, and drug use in real time to accurately and objectively reflect the patient's health status and rehabilitation progress. The application of smart medical systems has greatly promoted the automation and standardization of these recording processes, significantly improved work efficiency and data accuracy, and provided an important reference for doctors to make medical decisions.

[0003] However, in the process of recording infection nursing information, traditional methods mainly rely on sensor collection and manual record uploading. If the nursing staff is busy or there is a technical failure in the electronic recording system, such as sensor abnormalities, input errors, etc., the recorded information may be missing or erroneous, affecting the authenticity of the data samples, which in turn may lead to misjudgment of doctors in formulating medical strategies based on these data samples, affecting the patient's treatment process and threatening the patient's personal safety. Summary of the invention

[0004] In order to solve the technical problem that the existing recording technology causes errors in data samples that affect doctors' formulation of medical strategies, and the authenticity and reliability of data samples are affected, the purpose of the present invention is to provide an infection nursing recording method based on smart medical care, and the technical solutions adopted are as follows: Obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; Compare and analyze different data of a certain infection care in the data sample set to locate and obtain abnormal record data samples; Obtain the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval, and predict the abnormal record data samples to obtain the recovery trend of the target patient; Obtain different types of record data samples through the database, analyze the record data samples and data sample sets to obtain associated data dimensions and / or vocabulary types, combine the data sample sets and recovery trends to supplement the remaining data, and recover the abnormal record data samples to obtain complete infection care record data samples; The complete infection care record data sample is stored in the database.

[0005] Furthermore, the database and the identity information of the target patient are obtained, and the infection nursing record corresponding to the target patient is obtained according to the database and the identity information, and a data sample set is generated, including: By accessing the database, the identity information of the target patient is identified, and the infection care records corresponding to the target patient are extracted from the database according to the identity information. The timestamps and record types corresponding to the infection care records are arranged to generate a data sample set.

[0006] Furthermore, the infection care record includes any data of the target patient's infection diagnosis, care plan, historical care assessment, care execution record, drug use record and vital signs data.

[0007] Furthermore, the different data of a certain infection care in the data sample set are compared and analyzed to locate the abnormal record data samples, including: Through natural language technology The first nursing process Identify and classify entities within a record text; The first Sort the entities in the record text according to the order in the data sample set to obtain the vocabulary vector , repeat the steps to get All word vectors corresponding to the nursing process ; Get the recorded text based on all infection care processes of the target patient and record text The largest common subsequence of , get the record text and record text 1. The relationship between Calculate record text based on the maximum common subsequence The abnormal score value of , the corresponding calculation formula is: ; in, Indicates the recorded text and record text The maximum number of common subsequence elements; Indicates the recorded text The number of elements in that match the association; Repeat the steps to calculate the record text All other types of recorded text The abnormal score value of , Indicates the recorded text In the Abnormal score value of the nursing process; based on Determine whether it exceeds the preset score threshold and locate the abnormal record data sample.

[0008] Further, based on Determine whether the score exceeds the preset threshold and locate the abnormal record data sample, including: When the preset score threshold is exceeded, When The record text belongs to the abnormal record data sample; if Does not exceed the preset score threshold. When The record text does not belong to the abnormal record data sample.

[0009] Furthermore, the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval are obtained, and the abnormal record data samples are predicted to obtain the recovery trend of the target patient, including: Define the abnormal record data sample as Y, The time interval between each nursing process is ; Construct a two-dimensional coordinate system for the time interval Perform curve fitting to calculate the overall time interval change trend of the target patient's infection care , and the interval change rate of abnormal record data sample Y , the corresponding calculation formula is: ; in, Indicates that from End of care to the time interval between the start of care; Indicates that from End of care to the time interval between the start of care; Calculate the overall time interval change trend and interval change rate The mean of the target patient's recovery trend is obtained. , the corresponding calculation formula is: .

[0010] Further, different types of record data samples are obtained through the database, and the record data samples and data sample sets are analyzed to obtain the associated data dimensions and / or vocabulary types, and the remaining data is supplemented by combining the data sample sets and the recovery trend, and the abnormal record data samples are recovered to obtain the complete infection care record data samples, including: Obtained through the database Any recorded data sample of the nursing process, and supplement or replace the corresponding data dimension and / or vocabulary type according to the corresponding relationship with the abnormal recorded data sample; Analyze and obtain associated data dimensions and / or vocabulary types based on record data samples and data sample sets, complete the supplementation and replacement of abnormal record data samples by all associated relationships, and obtain abnormal record data samples to be improved; According to the recovery trend, the abnormal record data samples to be improved are repaired to obtain complete infection care record data samples.

[0011] Furthermore, the abnormal record data samples to be improved are repaired according to the recovery trend, including: when When the abnormal record data sample is completed, the remaining data of the abnormal record data sample to be improved is supplemented and replaced according to the record data sample of the same type; when When repairing the abnormal record data sample to be improved, the data sample set is not referred to; when When the data is retrieved, the possibility of supplementing different types of data dimensions and / or vocabulary types in the data sample set is calculated based on the recovery trend; In the calculation data sample set, Possibility to supplement with data dimensions and / or vocabulary types , the corresponding calculation formula is: ; in, Indicates that there is The first data dimension and / or vocabulary type exists The probability of a sequence appearing in the data sample set; Indicates that there is The number of sequences that exist for each data dimension and / or vocabulary type; represents the normalization function; Indicates The matching of the sequences in the abnormal record data sample; based on Determine whether it exceeds the preset possible threshold. When The data dimensions and / or vocabulary types are added to the abnormal record data sample to be improved to obtain a complete infection care record data sample; No need to add data dimensions and / or vocabulary types.

[0012] In order to solve the above technical problems, the present invention provides another technical solution as follows: an infection care recording device based on smart medical care, comprising the following units: A generating unit, used to: obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; A data locating unit is used to compare and analyze different data of a certain infection care in a data sample set, and locate and obtain abnormal record data samples; A data prediction unit is used to obtain the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval, and predict the abnormal record data samples to obtain the recovery trend of the target patient; The data processing unit is used to: obtain different types of record data samples through the database, analyze the record data samples and the data sample set to obtain the associated data dimensions and / or vocabulary types, supplement the remaining data in combination with the data sample set and the recovery trend, and recover the abnormal record data samples to obtain the complete infection care record data samples; The storage unit is used to store the complete infection care record data sample in a database.

[0013] To solve the above technical problems, the present invention provides another technical solution as follows: an infection care record system based on smart medical care, comprising the following modules: The acquisition module is used to: obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; The abnormal data analysis and repair module is used to: compare and analyze different data of a certain infection care in the data sample set, locate and obtain abnormal record data samples; obtain the time series changes of abnormal record data samples and the dynamic changes of the infection time interval of the target patient, and predict the abnormal record data samples to obtain the recovery trend of the target patient; obtain different types of record data samples through the database, analyze the record data samples and the data sample set to obtain the associated data dimensions and / or vocabulary types, combine the data sample set and the recovery trend to supplement the remaining data, and restore the abnormal record data samples to obtain a complete infection care record data sample; The storage module is used to store the complete infection care record data sample in the database.

[0014] The present invention has the following beneficial effects: the present invention collects data, mines data associations between different infection nursing records, combines the differences in associated data and vocabulary between different samples, accurately identifies and locates abnormal infection nursing record data samples, so as to improve the ability to identify abnormal samples, and in the repair process of abnormal record data samples, makes full use of the association between samples, and makes fine adjustments to the repair data based on the recovery trend of the target patient and the historical nursing records, to ensure that the repair results of the abnormal infection nursing record data samples are more reliable and authentic, solves the misjudgment caused by record errors or record loss, and prevents the doctor's treatment strategy from being affected. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A system flow chart of an infection care recording method based on smart medical care provided by one embodiment of the present invention; Figure 2 A flowchart of a method for recording infection care based on smart medical care provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of an infection care recording device based on smart medical care provided by an embodiment of the present invention; Figure 4 A schematic diagram of the structure of an infection care record system based on smart medical care provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of an infection care recording method, device and system based on smart medical care proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following is a detailed description of a method, device and system for recording infection care based on smart medical care provided by the present invention in conjunction with the accompanying drawings.

[0020] When recording information in the infection care process, the traditional method is usually sensor collection, manual record uploading and the like. If the nursing staff is busy and makes mistakes or the electronic record system has a technical failure, this will cause the recorded information to be missing or erroneous, affecting the authenticity of the recorded data, and then causing the doctor to make an erroneous medical strategy based on the recorded information, threatening the patient's safety. The first embodiment of the present invention provides an infection care recording method based on smart medical care. The method aims to mine the correlation between different data, and obtain the differences in corresponding data and vocabulary based on their correlation, so as to locate the data samples with abnormalities for recovery, and improve the reliability and authenticity of the abnormal infection care record data samples. In order to realize the infection care recording method based on smart medical care, an infection care recording device based on smart medical care and an infection care recording system based on smart medical care are provided. The device and the system are essentially a software system, which is composed of modules or units that realize corresponding functions. The specific steps in the method are now introduced in detail.

[0021] Please combine Figure 1 and Figure 2 , which respectively show a system flow chart and a step flow chart of an infection care recording method based on smart medical care provided by an embodiment of the present invention, the method comprising: Step S1: Obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; Step S2: Compare and analyze different data of a certain infection care in the data sample set to locate and obtain abnormal record data samples; Step S3: Obtain the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval, and predict the abnormal record data samples to obtain the recovery trend of the target patient; Step S4: different types of record data samples are obtained through the database, and the record data samples and data sample sets are analyzed to obtain associated data dimensions and / or vocabulary types, and the remaining data is supplemented in combination with the data sample set and the recovery trend, and the abnormal record data samples are recovered to obtain complete infection care record data samples; Step S5: Store the complete infection care record data sample in the database.

[0022] Specifically, infection care records provide decision support for doctors and ensure that patients receive appropriate treatment. However, traditional information records mainly rely on sensors to collect data and manually record and upload, which has some practical problems. One is that in an extremely busy environment where nursing staff work, the chaotic and high-pressure working atmosphere may lead to various problems in information recording. For example, due to the focus on urgent treatment and nursing tasks, nursing staff may miss some key infection information; or in the intense work, the same information may be recorded repeatedly, or incorrect records may be made due to distraction. Such situations will affect the accuracy of infection care and thus affect subsequent medical decisions.

[0023] Secondly, when a technical failure occurs in the electronic record part of the smart medical system, such as sensor failure or unstable network connection causing jamming, data collection and uploading may be problematic. Sensor failure may result in some key infection indicators not being accurately recorded; network jamming may result in data not being uploaded in time, or even interruption of the data upload process, resulting in data missing or abnormal records, affecting the integrity and authenticity of the infection nursing records; when doctors formulate medical strategies based on these information records that may contain omissions or errors, it may be misleading, resulting in patients being unable to receive the most accurate diagnosis and treatment; this not only delays the disease, but may also pose a greater risk to the patient's health, so an infection nursing record method based on smart medical care is proposed in an embodiment of the present invention.

[0024] To better explain, smart healthcare refers to a medical model that uses modern information technology, such as the Internet of Things, cloud computing, and big data analysis, to intelligently manage and optimize medical activities, and to realize modern medical technology through electronic means. Specifically, smart healthcare integrates all aspects of medical services, including diagnosis, treatment, rehabilitation, and health management, through high-tech means, making medical services more efficient, accurate, and personalized. The application of smart healthcare can greatly improve the utilization efficiency of medical resources, reduce medical costs, and improve the quality of medical services, while also better meeting the needs of patients.

[0025] Infection nursing records refer to the detailed records of the patient's infection status, treatment process and nursing measures during medical activities, especially for patients with infectious diseases, so that medical staff can fully understand and evaluate the patient's condition, and thus develop more scientific and reasonable treatment and nursing plans; it can be explained that the infection nursing record method based on smart medical care is to combine the two, use the high-tech means of smart medical care, and manage the infection nursing records electronically and intelligently, so as to complete the infection nursing work efficiently and accurately and improve the quality of medical services.

[0026] Preferably, step S1 comprises: By accessing the database, the identity information of the target patient is identified, and the infection care records corresponding to the target patient are extracted from the database according to the identity information. The timestamps and record types corresponding to the infection care records are arranged to generate a data sample set.

[0027] Preferably, the infection care record includes any one of the target patient's infection diagnosis, care plan, historical care assessment, care execution record, medication use record and vital signs data.

[0028] It is stated that the data covered by the infection care record is critical for monitoring the health status of the target patient, evaluating the effectiveness of treatment, and adjusting the care plan.

[0029] Infection diagnosis is the process of determining whether the target patient is infected with a certain pathogen, usually including a comprehensive analysis of the patient's symptoms, signs and laboratory test results in order to accurately determine the cause and type of infection; nursing plans are nursing measures formulated to help the target patient recover, including assessing the patient's condition, determining nursing goals, formulating nursing plans, and implementing these plans; historical nursing assessment is the process of reviewing and analyzing the patient's past nursing situation, which helps doctors and nurses understand the patient's medical history, treatment experience and problems in the nursing process, so as to better formulate current and future nursing plans; nursing execution records are records of specific measures performed by nurses and observed patient conditions during the nursing process, which help doctors understand changes in patients' conditions, evaluate nursing effects, and adjust treatment plans as needed; drug use records are records of the types, doses, time and effects of drugs used by patients, which help doctors and nurses monitor the efficacy and side effects of drugs and adjust drug use plans to ensure patient safety and treatment effects; vital signs data are data recording patients' heart rate, blood pressure, respiratory rate, body temperature and other vital signs, which are used to evaluate patients' conditions and treatment effects.

[0030] It can be explained that a database refers to a warehouse that organizes, stores and manages data according to data structures, providing reliable data support for various applications.

[0031] By directly accessing the data, the identity information of the target patient is identified and obtained, including but not limited to the patient's name, age, gender, ID number, etc., to ensure that the user can accurately find the record of the target patient in the database, and further query and extract the infection care records corresponding to the patient in the database based on the identity information, including but not limited to the target patient's infection diagnosis, nursing plan, historical nursing assessment, nursing execution record, drug use record and vital signs data. To ensure the efficiency and quality of data processing, the extracted infection care records are arranged according to timestamp and record type to form an ordered data sample set, which provides a good foundation for subsequent data analysis and model training.

[0032] Specifically, in real life, authorized access is made to the hospital's medical record and treatment database to obtain the electronic medical record information of a specific patient; then, based on the patient's detailed information, the various records of his / her infection care process are accurately retrieved from the database, including but not limited to infection diagnosis information, nursing plans, historical nursing assessment results, specific records of nursing execution, details of drug use, and vital signs data, etc.; to ensure the accuracy and consistency of the data, all extracted infection care data will be arranged in order according to the logical order of timestamps and record types, thereby completing the comprehensive collection of the patient's historical infection care data to generate a data sample set.

[0033] Preferably, step S2 comprises: Step S21: Use natural language technology to The first nursing process Identify and classify entities within a record text; To better explain, natural language technology refers to a technology that uses knowledge from fields such as computer science, artificial intelligence and statistics to process, understand and generate human natural language (such as Chinese). This technology mainly includes language models, syntactic analysis, semantic understanding, text classification, machine translation and other aspects. Its purpose is to enable computers to better understand human language and realize functions such as human-computer interaction and natural language generation. The core task of natural language technology is to understand specific semantics, and semantic understanding refers to the process by which computers understand and infer the meaning of words, sentences and passages in natural language.

[0034] It can be explained that in the records of a single infection nursing process, there must be a close correlation between the various data. For example, the nursing staff's nursing plan and its implementation records will be closely linked with the patient's medication use records to ensure consistency and accuracy between the data. Therefore, users use comparative analysis methods to deeply explore the different data items in a certain infection nursing, so as to accurately identify and locate any abnormal record data samples to ensure the rigor and effectiveness of nursing work.

[0035] In actual operation, natural language processing (NLP) is used to The first Identify and classify entities (vocabulary, symbols, etc.) within a record text, such as drug names, patient symptoms, physical signs data, etc.; by classifying entities, it is convenient for subsequent data analysis and decision support. This technical means can improve the efficiency of medical document processing and ensure the stability of nursing quality.

[0036] As an optional implementation method, NLP technology generates a large amount of recorded text every day, and it takes a lot of time and computing resources to process a lot of data. There may be certain efficiency issues. To solve this problem, NLP technology has a number of extended technologies that can solve the timeliness problem and improve computing efficiency. For example, parallel processing technology: using graphics processing unit (GPU) acceleration or other parallel processing technologies to process multiple data points at the same time; multi-task learning: training multiple related tasks in a model at the same time can improve the efficiency of the model and reduce training and inference time. In multi-task learning, the model can share representations, that is, while processing one task, the model learns knowledge of other related tasks, which can not only improve the generalization ability of the model, but also improve the processing speed of the model; incremental learning / online learning: using incremental learning or online learning algorithms to allow the model to be gradually updated, that is, the model continuously learns from the data instead of training from scratch every time; the application of these extended technologies makes NLP technology more efficient and real-time when processing large amounts of data.

[0037] In addition, in real life, different medical staff use different expressions in recording methods, that is, there are certain differences in the expressions used in recording medical records or medical orders, which may affect the recognition and classification of natural language technology. To avoid this problem, first, medical staff should be trained in standardized recording so that they can follow certain norms and standards when recording medical records and try to use standardized terms and expressions to solve the problem of identification caused by inconsistent expressions from the source; second, context-related NLP models should be used. These models can understand the actual meaning behind different expressions, such as Transformer-based models (such as BERT, GPT, etc.). These models perform well in understanding the context of language and can effectively identify and process complex relationships and meanings in medical texts; third, entity linking technology should be implemented to associate entities mentioned in medical texts (such as disease names, drug names, etc.) with standardized entities in the knowledge base to reduce ambiguity caused by inconsistent entity names.

[0038] Step S22: Sort the entities in the record text according to the order in the data sample set to obtain the vocabulary vector , repeat the steps to get All word vectors corresponding to the nursing process ; To explain, vocabulary vector refers to mapping words or phrases into numerical vectors of fixed length. These vectors can capture the semantic information and contextual associations of the vocabulary, enabling computers to understand and process natural language. They can be used for many tasks, such as text classification, sentiment analysis, machine translation, and information retrieval. Vocabulary vectors capture the semantic information of vocabulary by learning a large amount of text data. This method is also called word embedding. The core idea is to map each word in the vocabulary into a vector space of fixed dimension. In this space, words with similar semantics will be close to each other's vector representations. This spatial structure enables computers to understand and process natural language by calculating the distance and similarity between vectors.

[0039] Specifically, the The entities in the record text are sorted, and the sorted results will form a vocabulary vector ,Right now Repeat this process to ensure that each entity in the nursing process can be accurately identified and sorted, that is, to obtain the first All word vectors corresponding to the nursing process This setting can ensure that the data better reflects the actual situation and is conducive to the subsequent application of the data.

[0040] Step S23: Based on all infection care processes of the target patient, obtain the recorded text and record text The largest common subsequence of , get the record text and record text The relationship between To explain, the largest common subsequence (LCS) refers to the subsequence with the longest length and whose elements correspond to each other in order among two or more sequences. In layman's terms, suppose there are two sequences, sequence A and sequence B. By comparing the elements of sequence A and sequence B, we can find the common elements between the two, and then arrange these common elements in the order in which they appear in the original sequences. The resulting sequence is the largest common subsequence of sequence A and sequence B.

[0041] The process of obtaining the maximum common subsequence is to analyze the association between different record data samples to obtain the common content between different record data samples, that is, the maximum common subsequence refers to the association relationship between record texts. This setting can more accurately grasp the continuity and changes in the target patient's nursing process.

[0042] Step S24: Calculate the record text according to the maximum common subsequence The abnormal score value of , the corresponding calculation formula is: ; in, Indicates the recorded text and record text The maximum number of common subsequence elements; Indicates the recorded text The number of elements in that match the association; To explain, the purpose of obtaining the abnormal score is to obtain a score that measures the degree of abnormality, that is, to find the record text The parts that are most similar to the normal situation and evaluate the recorded text by calculating the continuity of these similar parts The abnormality of Indicates the recorded text and record text The length of the maximum common subsequence between the two records, that is, the length of the maximum continuous series of characters shared between the two record texts; Indicates that the text is being recorded The number of elements that match the preset association relationship in the record text is used to evaluate the number of parts in the record text that meet specific standards or conditions; the longer the length of the maximum common subsequence, the higher the similarity between the record text and the normal situation; conversely, the shorter the length of the maximum common subsequence, the greater the difference between the record text and the normal situation.

[0043] Step S25: Repeat the steps to calculate the recorded text All other types of recorded text The abnormal score value of , Indicates the recorded text In the Abnormal score value of the nursing process; Explanation: Repeat the steps to calculate the abnormal score values ​​of different types of record texts to comprehensively evaluate the nursing process, and calculate the average of the abnormal score values ​​to reflect the record text. The relative abnormality degree with other recorded texts is used to conduct quantitative analysis of nursing work.

[0044] Step S26: Based on Determine whether it exceeds the preset score threshold and locate the abnormal record data sample.

[0045] It is explained that a given preset score threshold can more accurately locate abnormal recorded data samples to effectively identify data samples that do not conform to the normal range.

[0046] Preferably, step S26 includes: if When the score exceeds the preset threshold, When The record text belongs to the abnormal record data sample; if Does not exceed the preset score threshold. When The record text does not belong to the abnormal record data sample.

[0047] It can be explained that, according to the actual situation and the setting of the demand, the preset score threshold is 0.4. The abnormal record data sample refers to the data point that is different from the majority of the data in the data sample set and may contain errors or abnormal information. By classifying the data with a given preset score threshold, it is easy to identify and process abnormal records, thereby improving the data quality and the accuracy of the analysis results.

[0048] Preferably, step S3 includes: It is explained that after locating the abnormal record data samples, the fluctuation characteristics of this type of data in the time series and the dynamic change rules of the target patient's infection interval are analyzed to accurately predict the abnormal record data samples, thereby realizing the correction and optimization of the abnormal record data samples; and the recovery trend of the target patient is obtained in order to obtain more accurate and reliable prediction results.

[0049] Step S31: Define the abnormal record data sample as Y, The time interval between each nursing process is ;

[0050] It should be noted that the time interval for the target patient to receive infection care is usually determined by the doctor after a comprehensive assessment of multiple factors such as the target patient's infection type, severity, and overall health status, and this care time interval will be adjusted in a timely manner according to the target patient's disease response and treatment progress; when the infection is more serious and the target patient's health is poor, in order to control the disease more quickly, the care time interval will usually be shortened; conversely, when the infection is mild and the target patient's health is better, the doctor may set a longer care time interval to reduce the impact on the patient's daily life. Therefore, by observing the changes in the target patient's care time interval, it is possible to determine whether the target patient's condition is improving or worsening.

[0051] Step S32: Construct a two-dimensional coordinate system for the time interval Perform curve fitting to calculate the overall time interval change trend of the target patient's infection care , and abnormal record data samples The interval change rate , the corresponding calculation formula is: ; in, Indicates that from End of care to the time interval between the start of care; Indicates that from End of care to the time interval between the start of care; To explain, curve fitting obtains the changing trend. Curve fitting is to establish a mathematical model to approximate the changing relationship between data. Its purpose is to find a curve that can reflect the changing trend of data as accurately as possible. Obtaining the interval change rate can understand the changing speed of the time interval, so as to better monitor and adjust the nursing strategy during the nursing process, and ensure that the target patient receives timely and effective infection care. In the nursing process, the time interval between each nursing is of great significance to the recovery and treatment effect of the target patient. Specifically, and Can be used to assess the rate and continuity of infection care to ensure maximum treatment effectiveness and target patient comfort.

[0052] Step S33: Calculate the overall time interval change trend and interval change rate The mean of the target patient's recovery trend is obtained. , the corresponding calculation formula is: ; It is explained that the recovery trend refers to the dynamic process of the target patient's condition gradually improving, which is calculated by calculating the change trend of the overall time interval and interval change rate to obtain a quantitative value that can reflect the recovery trend of the target patient. .

[0053] Preferably, step S4 comprises: It is explained that there are certain correlation data dimensions or key words between different types of record data samples obtained in the database and the data sample set. Based on this, combined with the data sample set and recovery trend, the remaining data is supplemented to completely restore the abnormal record data samples and improve data quality and availability.

[0054] Step S41: Get the first step through the database Any recorded data sample of the nursing process, and supplement or replace the corresponding data dimension and / or vocabulary type according to the corresponding relationship with the abnormal recorded data sample; To explain, data dimension refers to the attributes of various aspects contained in the data, which can be numerical, text or any other form. Data dimensions can be used to analyze and understand data from more angles; vocabulary type refers to the type and nature of vocabulary in the recorded data sample, which is divided into nouns, verbs, adjectives and other types. Each type has a specific usage and meaning. Through the analysis and processing of vocabulary types, we can better understand the meaning of the data and extract key information.

[0055] Specifically, to implement step S41, the following steps are included: First, we use the N-gram model to analyze the probability of word combinations of similar historical samples of abnormal record data samples, and obtain the probability of occurrence of different word types and data sequences in historical samples. ; Next, according to the probability of occurrence of different vocabulary types and data sequences in historical samples and the data dimensions and / or vocabulary types that need to be supplemented, and finding relevant vocabulary in the abnormal record data samples; Finally, the location where the data dimension and / or vocabulary type is added is determined according to the sequence order of the supplementary vocabulary and its related vocabulary; if the relevant data dimension and / or vocabulary type is not found in the abnormal record data sample, the data dimension and / or vocabulary type is added to the end of the abnormal record data sample.

[0056] To explain, the N-gram model is a language model based on statistics. Its core idea is that the probability distribution of a sentence or word sequence can be estimated by the historical frequency of its constituent elements. It can be explained that the N-gram model assumes that the first n The probability of an element (word or character) depends only on the previous n-1 It is related to the elements in the sequence but not to the elements farther away in the sequence, which can effectively capture the local dependencies in the language and provide probability estimates for tasks such as machine translation, text generation, and information retrieval.

[0057] Understandably, the N-gram model is used to conduct in-depth word combination probability analysis on similar historical samples in the abnormal record data samples to obtain the probability of different word types and data sequences appearing in historical samples. P , providing basic data for subsequent analysis and processing; according to the probability of occurrence of different vocabulary types and data sequences in historical samples P , as well as the data dimensions and / or vocabulary types that need to be supplemented, the abnormal record data samples are searched and analyzed in detail; finally, the adding position of the data dimensions and / or vocabulary types is determined according to the sequence order of the supplementary vocabulary and its related vocabulary. This setting can ensure that the added data dimensions and / or vocabulary types can be effectively integrated into the entire data sequence; in addition, if the relevant data dimensions and / or vocabulary types are not searched in the abnormal record data sample, these data dimensions and / or vocabulary types are supplemented to the end of the abnormal record data sample; the whole method can improve the understanding and processing capabilities of the abnormal record data samples, so as to better serve the data analysis and processing work.

[0058] Step S42: Analyze and obtain associated data dimensions and / or vocabulary types based on the record data samples and the data sample set, complete the supplementation and replacement of the abnormal record data samples by all associated relationships, and obtain the abnormal record data samples to be improved; It is explained that supplementing and replacing the abnormal record data samples based on the association relationship can fill in and correct the missing and erroneous information in the abnormal record data samples, and completing the supplement of all association relationships makes the abnormal record data samples more complete and accurate, providing a solid foundation for subsequent data analysis and processing.

[0059] Step S43: Repair the abnormal record data sample to be improved according to the recovery trend to obtain a complete infection care record data sample.

[0060] Explanation is given, and some different strategies are adopted according to the abnormal record data samples and recovery trends to be improved, so that they can be handled more appropriately, and finally a comprehensive and accurate infection nursing record data sample is formed to ensure the accuracy of the information and provide strong support for infection nursing work.

[0061] Preferably, in step S43, repairing the abnormal record data sample to be improved according to the recovery trend includes: It can be explained that, according to the different current recovery trends of the target patients, it is necessary to restore the abnormal record data samples according to different situations to improve the accuracy of the data.

[0062] Specifically, when When the abnormal record data sample is completed, the remaining data of the abnormal record data sample to be improved is supplemented and replaced according to the record data sample of the same type; To explain, That is, when the recovery trend of the target patient is good, the data sample set is carefully analyzed and compared, and the remaining data of the abnormal record data sample to be improved is supplemented and replaced according to the valid record data samples of the same type, ensuring that the supplemented and replaced data samples are consistent with the type and characteristics of the original data samples, thereby enhancing the value of the record data samples.

[0063] when When repairing the abnormal record data sample to be improved, the data sample set is not referred to; Understandably, when That is, when the condition of the target patient deteriorates, the nursing plan changes greatly. Without referring to the data sample set, the perfect abnormal record data sample is directly repaired to better meet the needs of the patient, improve the quality of care, and prevent the target patient's condition from further deteriorating.

[0064] when When the data is retrieved, the possibility of supplementing different types of data dimensions and / or vocabulary types in the data sample set is calculated based on the recovery trend; It can be explained that the probability of occurrence of different vocabulary types and data sequences in historical samples of the same type of historical samples combined with abnormal record data samples , and the recovery trend of the target patient, calculate the Possibility to supplement with data dimensions and / or vocabulary types , when the probability of a vocabulary type and / or data dimension in the historical samples The larger it is, the greater the possibility of supplementing it; if the elements in the vocabulary type and / or data dimension match some elements in the abnormal data sample, it will be supplemented.

[0065] Specifically, in the calculation data sample set, Possibility to supplement with data dimensions and / or vocabulary types , the corresponding calculation formula is: ; in, Indicates that there is The first data dimension and / or vocabulary type exists The probability of a sequence appearing in the data sample set; Indicates that there is The number of sequences that exist for each data dimension and / or vocabulary type; represents the normalization function; Indicates initial possibilities for supplementation with data dimensions and / or vocabulary types; Indicates The matching status of the sequences in the abnormal record data samples.

[0066] It can be explained that represents the penalty term for the initial likelihood; Indicates The matching of the sequences in the abnormal data samples, except for the When there is an exact match outside of the vocabulary type and / or data dimension, , otherwise .

[0067] It is explained that the penalty term of the initial possibility refers to a penalty measure taken for the data with a low possibility initially given in the calculation formula. It mainly penalizes the data to different degrees according to the initial possibility of the data, so as to ensure that the data can pay more attention to the data with a higher initial possibility during the training process, thereby improving the accuracy and stability of the calculation. In this embodiment, the penalty term is used to reduce the possibility of the sequence that does not match the abnormal record data sample, that is, if the matching degree of the sequence in the abnormal record data sample is not high, the initial possibility is penalized, thereby reducing its overall possibility.

[0068] If a sequence except the The vocabulary types and / or data dimensions match exactly, then , indicating that the sequence completely matches the abnormal record data sample; conversely, if the sequence matches the abnormal record data sample in other aspects, then , indicating that the sequence partially matches the abnormal record data sample.

[0069] based on Determine whether it exceeds the preset possible threshold. When The data dimensions and / or vocabulary types are added to the abnormal record data sample to be improved to obtain a complete infection care record data sample; No need to add data dimensions and / or vocabulary types.

[0070] To explain, the preset possible threshold is 0.5. When The data dimensions and / or vocabulary types are supplemented to obtain a more complete and comprehensive infection care record data sample; At this time, the abnormality of the abnormal record data sample is low and does not exceed the given preset possible threshold, so there is no need to supplement the data dimensions and / or vocabulary types, indicating that the data anomaly at this time is normal and does not require additional processing, so as to more efficiently utilize resources, avoid unnecessary processing of relatively normal data, and improve work efficiency.

[0071] To illustrate, in step S5, the complete infection care record data sample is stored in the database.

[0072] Storing a sample of detailed infection care record data in a database ensures that all necessary information is properly maintained so that all relevant personnel can access the latest and accurate information.

[0073] As an optional implementation, before storing the complete infection care record data sample in the database, an impact assessment is performed on the infection care record data sample, that is, based on the detailed recovery status of each target patient, the potential impact of the data repair process on the target patient's nursing regimen and treatment plan is carefully considered. This setting is used to ensure that the data repair work will not have an adverse effect on the treatment and care of the target patient, and to appropriately adjust the repair strategy to ensure that the patient receives the most appropriate treatment and care.

[0074] At the same time, when storing, the complete infected nursing record data samples after repair are marked, and the repair process and final results are recorded, including but not limited to key information such as the specific repair measures taken, the reasons for the repair, and the content of the repaired data, so as to track and understand the repair history of the data and reduce the possibility of similar problems occurring again; by conducting detailed impact assessments and sufficient marking records on the repaired infected nursing record data samples, the efficient management and use of data are ensured, thereby improving the quality of medical services and patient satisfaction.

[0075] Please combine Figure 1 and Figure 3 , which respectively show a system flow chart of an infection care recording method based on smart medical care provided by an embodiment of the present invention and a structural schematic diagram of an infection care recording device based on smart medical care provided by an embodiment of the present invention, including the following units: A generating unit, used to: obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; A data locating unit is used to compare and analyze different data of a certain infection care in a data sample set, and locate and obtain abnormal record data samples; A data prediction unit is used to obtain the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval, and predict the abnormal record data samples to obtain the recovery trend of the target patient; The data processing unit is used to: obtain different types of record data samples through the database, analyze the record data samples and the data sample set to obtain the associated data dimensions and / or vocabulary types, supplement the remaining data in combination with the data sample set and the recovery trend, and recover the abnormal record data samples to obtain the complete infection care record data samples; The storage unit is used to store the complete infection care record data sample in a database.

[0076] It is explained that the generation unit is mainly responsible for creating new data patterns or samples; the data location unit is used to quickly retrieve and locate the required data from a large data set; the data prediction unit is used to predict established trends or events through data sets; the data processing unit is responsible for performing corresponding conversions, supplements and other processing operations on the data; and the storage unit is responsible for long-term or short-term storage of data.

[0077] Specifically, first, a generation unit is used to obtain the infection care record of the target patient to generate a data sample set; a data positioning unit is used to compare and analyze each infection care record in the data sample set, and locate the abnormal record data sample through a high-precision algorithm; the data prediction unit analyzes the changes in the abnormal record data sample in the time series, and calculates and obtains the recovery trend in combination with the dynamic changes in the infection time interval of the target patient; the data processing unit uses different types of record data samples to mine the correlation between the data to integrate and process the analysis results and the data sample set, and processes the abnormal record data sample according to the recovery trend to obtain a complete infection care record data sample; and then the complete infection care record data sample is safely stored back in the database through the storage unit; the device in this embodiment uses multiple functional units to work together to improve the accuracy and efficiency of the infection care of the target patient, and improve the reliability and authenticity of the final repair of the abnormal record data sample.

[0078] Please combine Figure 1 and Figure 4 , which respectively show a system flow chart of an infection nursing record method based on smart medical care provided by an embodiment of the present invention and a structural diagram of an infection nursing record system based on smart medical care provided by an embodiment of the present invention, including the following modules: The acquisition module is used to: obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; The abnormal data analysis and repair module is used to: compare and analyze different data of a certain infection care in the data sample set, locate and obtain abnormal record data samples; obtain the time series changes of abnormal record data samples and the dynamic changes of the infection time interval of the target patient, and predict the abnormal record data samples to obtain the recovery trend of the target patient; obtain different types of record data samples through the database, analyze the record data samples and the data sample set to obtain the associated data dimensions and / or vocabulary types, combine the data sample set and the recovery trend to supplement the remaining data, and restore the abnormal record data samples to obtain a complete infection care record data sample; The storage module is used to store the complete infection care record data sample in the database.

[0079] It can be understood that when the module of the infection care record system based on smart medical care is in operation, it is necessary to utilize an infection care record method based on smart medical care provided in the first embodiment. Therefore, whether the acquisition module, the abnormal data analysis and repair module and the storage module are integrated or different hardware is configured to produce functions similar to the effects achieved by the present invention, they all fall within the scope of protection of the present invention.

[0080] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for recording infection care based on smart medical care, characterized in that: The method comprises: Obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; Compare and analyze different data of a certain infection care in the data sample set to locate and obtain abnormal record data samples; Obtain the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval, and predict the abnormal record data samples to obtain the recovery trend of the target patient; Obtain different types of record data samples through the database, analyze the record data samples and data sample sets to obtain associated data dimensions and / or vocabulary types, combine the data sample sets and recovery trends to supplement the remaining data, and recover the abnormal record data samples to obtain complete infection care record data samples; The complete infection care record data sample is stored in the database.

2. The infection nursing recording method based on smart medical care according to claim 1, characterized in that: Obtain the database and the identity information of the target patient, obtain the infection care record corresponding to the target patient based on the database and the identity information, and generate a data sample set, including: By accessing the database, the identity information of the target patient is identified, and the infection care records corresponding to the target patient are extracted from the database according to the identity information. The timestamps and record types corresponding to the infection care records are arranged to generate a data sample set.

3. The infection nursing recording method based on smart medical care according to claim 2, characterized in that: The infection care record includes any data of the target patient's infection diagnosis, care plan, historical care assessment, care execution record, medication use record and vital signs data.

4. The infection nursing recording method based on smart medical care according to claim 1, characterized in that: Compare and analyze different data of a certain infection care in the data sample set, and locate the abnormal record data samples, including: Through natural language technology The first nursing process Identify and classify entities within a record text; The first Sort the entities in the record text according to the order in the data sample set to obtain the vocabulary vector , repeat the steps to get All word vectors corresponding to the nursing process ; Get the recorded text based on all infection care processes of the target patient and record text The largest common subsequence of , get the record text and record text The relationship between Calculate record text based on the maximum common subsequence The abnormal score value of , the corresponding calculation formula is: ； in, Indicates the recorded text and record text The maximum number of common subsequence elements; Indicates the recorded text The number of elements in that match the association; Repeat the steps to calculate the record text All other types of recorded text The abnormal score value of , Indicates the recorded text In the Abnormal score value of the nursing process; based on Determine whether it exceeds the preset score threshold and locate the abnormal record data sample.

5. The infection nursing recording method based on smart medical care according to claim 4, characterized in that: based on Determine whether the score exceeds the preset threshold and locate the abnormal record data sample, including: When the score exceeds the preset threshold, When The record text belongs to the abnormal record data sample; if Does not exceed the preset score threshold. When The record text does not belong to the abnormal record data sample.

6. The infection nursing recording method based on smart medical care according to claim 1, characterized in that: Obtain the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval, and predict the abnormal record data samples to obtain the recovery trend of the target patient, including: Define the abnormal record data sample as Y, The time interval between each nursing process is ; Construct a two-dimensional coordinate system for the time interval Perform curve fitting to calculate the overall time interval change trend of the target patient's infection care , and the interval change rate of abnormal record data sample Y , the corresponding calculation formula is: ； in, Indicates that from End of care to the time interval between the start of care; Indicates that from End of care to the time interval between the start of care; Calculate the overall time interval change trend and interval change rate The mean of the target patient's recovery trend is obtained. , the corresponding calculation formula is: .

7. The infection nursing recording method based on smart medical care according to claim 6, characterized in that: Different types of record data samples are obtained through the database, and the record data samples and data sample sets are analyzed to obtain the associated data dimensions and / or vocabulary types. The remaining data is supplemented by combining the data sample set and the recovery trend, and the abnormal record data samples are recovered to obtain the complete infection care record data samples, including: Obtained through the database Any recorded data sample of the nursing process, and supplement or replace the corresponding data dimension and / or vocabulary type according to the corresponding relationship with the abnormal recorded data sample; Analyze and obtain associated data dimensions and / or vocabulary types based on record data samples and data sample sets, complete the supplementation and replacement of abnormal record data samples by all associated relationships, and obtain abnormal record data samples to be improved; According to the recovery trend, the abnormal record data samples to be improved are repaired to obtain complete infection care record data samples.

8. The infection nursing recording method based on smart medical care according to claim 7, characterized in that: Repair the abnormal record data samples to be improved according to the recovery trend, including: when When the abnormal record data sample is completed, the remaining data of the abnormal record data sample to be improved is supplemented and replaced according to the record data sample of the same type; when When repairing the abnormal record data sample to be improved, the data sample set is not referred to; when When the data is retrieved, the possibility of supplementing different types of data dimensions and / or vocabulary types in the data sample set is calculated based on the recovery trend; In the calculation data sample set, Possibility to supplement with data dimensions and / or vocabulary types , the corresponding calculation formula is: ； in, Indicates that there is The first data dimension and / or vocabulary type exists The probability of a sequence appearing in the data sample set; Indicates that there is The number of sequences that exist for each data dimension and / or vocabulary type; represents the normalization function; Indicates The matching of the sequences in the abnormal record data sample; based on Determine whether it exceeds the preset possible threshold. When The data dimensions and / or vocabulary types are added to the abnormal record data sample to be improved to obtain a complete infection care record data sample; No need to add data dimensions and / or vocabulary types.

9. An infection care recording device based on smart medical care, characterized in that: The following units are included: A generating unit, used to: obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; A data locating unit is used to compare and analyze different data of a certain infection care in a data sample set, and locate and obtain abnormal record data samples; A data prediction unit is used to obtain the time series changes of abnormal record data samples and the dynamic changes of the target patient's infection time interval, and predict the abnormal record data samples to obtain the recovery trend of the target patient; The data processing unit is used to: obtain different types of record data samples through the database, analyze the record data samples and the data sample set to obtain the associated data dimensions and / or vocabulary types, supplement the remaining data in combination with the data sample set and the recovery trend, and recover the abnormal record data samples to obtain the complete infection care record data samples; The storage unit is used to store the complete infection care record data sample in a database.

10. An infection nursing record system based on smart medical care, characterized by: Includes the following modules: The acquisition module is used to: obtain the database and the identity information of the target patient, obtain the infection nursing record corresponding to the target patient according to the database and the identity information, and generate a data sample set; The abnormal data analysis and repair module is used to: compare and analyze different data of a certain infection care in the data sample set, locate and obtain abnormal record data samples; obtain the time series changes of abnormal record data samples and the dynamic changes of the infection time interval of the target patient, and predict the abnormal record data samples to obtain the recovery trend of the target patient; obtain different types of record data samples through the database, analyze the record data samples and the data sample set to obtain the associated data dimensions and / or vocabulary types, combine the data sample set and the recovery trend to supplement the remaining data, and restore the abnormal record data samples to obtain a complete infection care record data sample; The storage module is used to store the complete infection care record data sample in the database.

Images