Intelligent operation clinical path method and platform

Abstract

The invention belongs to the technical field of data processing, and particularly relates to an intelligent operation clinical path method and platform. According to the method, medical diagnosis information is divided into a plurality of paths through national standard disease category library information, and each path constructs charging item information based on historical cases and expert consultation information and comprises charging data; the hospital information system preprocesses the patient medical diagnosis information, matches the patient medical diagnosis information to charging item information, and imports charging data; setting a benchmark diagnosis and treatment coefficient under each medical diagnosis path, and constructing the benchmark diagnosis and treatment coefficient; through comparison of actual charging items of the patient, calculating the compliance rate of the item diagnosis and treatment coefficient, and summarizing; performing abnormity judgment and adjustment on the medical diagnosis path according to the item diagnosis and treatment coefficient standard reaching rate CM; if no abnormity exists, a medical diagnosis report is generated, an expert formulates consultation information according to the report, and therefore a hospital dynamically optimizes the cost structure in the diagnosis and treatment process.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of data processing, and particularly relates to a smart operation clinical pathway method and platform. BACKGROUND

[0002] With the gradual deepening of the medical insurance payment system reform in China, the medical service payment method gradually changes from the traditional project payment to the more scientific and reasonable disease type score payment. Among them, the main disease type basis is the national standard disease type library; the national standard disease type library information refers to the collection of various disease information constructed according to the standard formulated at the national level. It aims to unify the key information such as disease related classification, diagnosis and coding, and lay an important foundation for standardized management of medical services. However, in the actual medical diagnosis and hospital operation process, the disease type score payment standard disease type classification is too general, and it is difficult to accurately reflect the actual medical demand.

[0003] In order to make up for the technical defects of the above medical insurance payment mode, the existing technology generally adopts the construction of a clinical pathway management smart diagnosis platform to optimize the allocation of medical resources and cost management of the hospital. The existing clinical pathway management scheme mainly standardizes the clinical medical behavior, and clearly defines the standard diagnosis and treatment process of each disease, trying to achieve medical cost control. However, the current clinical pathway management technology still has the following serious technical defects in the application of medical insurance score payment standard: Firstly, the existing clinical pathway management mode generally stays in simply listing diagnosis and treatment projects, and cannot accurately distinguish the reasonable use amount and cost of various medical diagnosis and treatment projects under the medical insurance score payment system. This extensive management method cannot accurately reflect whether each diagnosis and treatment project meets the national standard disease type library information, and cannot accurately match the disease data in the national standard disease type library, resulting in low accuracy of hospital medical cost control, and often exceeding the budget or failing to meet the medical insurance payment requirements in actual operation. Secondly, the existing clinical pathway management technology only relies on static cost control scheme and post-analysis, and cannot realize real-time monitoring and active adjustment of diagnosis and treatment project cost use, nor can it effectively evaluate the instant medical diagnosis and treatment cost expenditure with indicators, so that the hospital cannot timely find and correct unreasonable charging phenomenon in the diagnosis and treatment process, and cannot dynamically optimize the cost structure.

[0004] Therefore, it is urgent to improve the existing clinical pathway diagnosis analysis method and smart diagnosis platform to solve the defects of the foregoing technology. SUMMARY

[0005] One of the purposes of the present application is to provide a medical insurance version of clinical pathway diagnosis analysis method which can match the national standard disease type library information, and at the same time, real-time monitor and actively adjust the use of diagnosis and treatment project cost, so as to make the hospital timely find and correct the unreasonable charging phenomenon and dynamically optimize the cost structure in the diagnosis and treatment process.

[0006] In order to achieve the above-mentioned purposes of the application, the following technical solutions are implemented: A smart operation clinical pathway method, comprising the following steps: S101, according to the national standard disease type library information, the medical diagnosis information is divided into multiple medical diagnosis paths, each medical diagnosis path is constructed by the historical disease diagnosis case information and the expert consultation information The charging item information contains the charging data of the charging item in the medical diagnosis information; S201, pre-process the medical diagnosis information of multiple patients in the hospital information system, and automatically match the pre-processed medical diagnosis information with the diagnosis basic information, and import the charging data of each charging item in the matched medical diagnosis information into the charging item information; S301, according to the expert consultation information, set the benchmark diagnosis and treatment coefficient under each medical diagnosis path , based on the benchmark diagnosis and treatment coefficient , construct the benchmark diagnosis and treatment cost ; S401, in each medical diagnosis path, compare the actual charging item usage amount R real,j of the patient with the benchmark diagnosis and treatment cost to obtain the project diagnosis and treatment coefficient standard item a j , and summarize the project diagnosis and treatment coefficient standard item a j in each medical diagnosis path to obtain the project diagnosis and treatment coefficient standard rate ; S501, according to the project diagnosis and treatment coefficient standard rate in each medical diagnosis path, the path abnormality of the medical diagnosis path is judged, and the benchmark diagnosis and treatment coefficient of the path abnormality of the medical diagnosis path is adjusted; When each medical diagnosis path does not appear path abnormality, a medical diagnosis report is generated; the medical diagnosis report includes a data report and an effective income report, and the effective income report contains the summary data of the effective income of each medical diagnosis path, and the expert formulates the expert consultation information through the medical diagnosis report.

[0007] The above technical solutions have the following technical effects: According to the medical insurance national payment standard, the medical diagnosis information is further optimized and divided, and a plurality of (drug path, technique path, consumption path, nursing path, medical path and daily path, etc.) medical diagnosis paths are established. Through the historical case data and expert consultation information, the charging item information of the medical diagnosis path is defined item by item, so as to realize the accurate mapping of the medical diagnosis data and the national standard disease library information, and ensure the accurate matching of the national standard disease library information and the actual medical expenses of the hospital. In addition, the automatic data preprocessing technology is adopted, the medical diagnosis information of the patient in the hospital information system (HIS) is standardized, and is automatically matched to the plurality of medical diagnosis paths constructed, effectively solving the problems of weak data correlation and insufficient real-time performance, so that the hospital can timely and accurately master the cost control situation within the medical insurance payment limit.

[0008] In addition, for the medical diagnosis path of the path exception, the application sets a dynamic benchmark diagnosis and treatment coefficient through expert consultation information, compares the actual diagnosis and treatment item usage amount with the benchmark diagnosis and treatment cost item by item, obtains the item diagnosis and treatment coefficient standard item in real time, calculates the diagnosis and treatment coefficient standard rate, quickly finds the cost exception and diagnosis and treatment path exception, and timely initiates intervention and adjustment, so as to ensure that the medical cost control is more refined and accurate. Effectively avoid waste of medical resources and decline of medical service quality, and comprehensively solve the problems of insufficient cost control accuracy and poor real-time response ability in the application of medical insurance score payment mode.

[0009] Further, each charging item information is associated with diagnosis basic information and medical insurance payment attribute information; the medical insurance payment attribute information sets the effective income proportion of each charging item according to the medical insurance national payment standard of disease score payment.

[0010] As a further improvement of the intelligent operation clinical path method of the application, the effective income of each medical diagnosis path is calculated as follows: Wherein, the effective income represents the result of the summary data, The first medical diagnosis path is counted into the effective income in the medical insurance national payment standard of disease score payment. The calculation method of each medical diagnosis path is as follows: Wherein, is the benchmark diagnosis and treatment cost of the jth charging item in the medical diagnosis path, is the medical insurance payment attribute information of the jth charging item.

[0011] ​As a further improvement of the intelligent operation clinical pathway method of the present application, when the medical diagnosis process of the patient includes multiple stages, the summary data calculation method of the effective income of each medical diagnosis pathway included in the effective income report is: Wherein, L is the number of stages included in the medical diagnosis process.

[0012] As a further improvement of the intelligent operation clinical pathway method of the present application, the calculation method of the benchmark diagnosis and treatment cost is: Wherein, is the standard cost of the jth charge item at each use, is the expected number of uses or use amount of the jth charge item; The project diagnosis and treatment coefficient meets: Wherein, m is the number of charge items containing data in the medical diagnosis information in the medical diagnosis pathway.

[0013] As a further improvement of the intelligent operation clinical pathway method of the present application, the medical diagnosis pathway includes drug path, technique path, consumption path, nursing path, medical path and daily path; The drug path includes charge data of drug characteristics in the medical diagnosis information; The technique path includes charge item information of medical technique inspection characteristics in the medical diagnosis information; The consumption path includes charge item information of consumable characteristics in the medical diagnosis information; The nursing path includes charge item information of nursing service characteristics in the medical diagnosis information; The medical path includes charge item information of physician operation characteristics in the medical diagnosis information; The daily path includes charge item information of daily characteristics of hospitalization in the medical diagnosis information.

[0014] As a further improvement of the intelligent operation clinical pathway method of the present application, the method for matching each charge data in the medical diagnosis information to the charge item information in different medical diagnosis pathways is: S11, set a medical diagnosis pathway template, the medical diagnosis pathway template includes a feature vector of drug characteristics in the drug path, a feature vector of medical technique inspection characteristics in the technique path, a feature vector of consumable characteristics in the consumption path, a feature vector of nursing service characteristics in the nursing path, a feature vector of physician operation characteristics in the medical path, and a feature vector of daily characteristics of hospitalization in the daily path; ​S21, the feature vector of each charge item in the medical diagnosis information is calculated with the feature vector of each medical diagnosis path in the medical diagnosis path template; S31, each charge item in the medical diagnosis information is imported into the charge item information of the medical diagnosis path with the maximum similarity calculation value; S41, for the charge items with the error between the similarity calculation results of the charge items and the feature vectors of multiple medical diagnosis paths less than the second threshold value, the charge items are manually imported into the charge item information of the medical diagnosis path by the administrator.

[0015] As a further improvement of the intelligent operation clinical path method of the present application, the formula for similarity calculation is: wherein, is the feature vector of the charge item j; is the feature vector of each medical diagnosis path in the medical diagnosis path template; k is the number of feature vectors in the medical diagnosis path, t1 is the feature vector of the drug characteristics in the drug path, t2 is the feature vector of the medical examination characteristics in the technique path, t3 is the feature vector of the consumable characteristics in the consumption path, t4 is the feature vector of the nursing service characteristics in the nursing path, t5 is the feature vector of the physician operation characteristics in the physician path, and t6 is the feature vector of the daily characteristics in the daily path.

[0016] As a further improvement of the intelligent operation clinical path method of the present application, the method for preprocessing the medical diagnosis information is: S1, the medical diagnosis information in the hospital information system is constituted into an original data set, the medical diagnosis information of each record in the original data set is marked as X, and the original data set is imported into the platform data processing module by the interface or manual uploading mode; wherein, the original data set is expressed as: {X1, X 2,.., X N}, and N is the number of medical diagnosis information; S2, the abnormal value judgment formula for each data in the original data set is: wherein, Z is the abnormal value of the medical diagnosis information in the original data set; μ is the average value of each medical diagnosis information in the original data set; is the average deviation amplitude between each medical diagnosis information in the original data set and the average value μ; S3, the medical diagnosis information with |z|>3 is marked as abnormal, and the missing value filling or manual filling is performed on the medical diagnosis information, and the missing value filling formula is: M is the number of non-missing records in the medical diagnosis information, medical diagnosis information data after missing value supplement, representing non-missing data in the medical diagnosis information.

[0017] The second object of the present application is to provide a medical insurance version of a clinical pathway diagnosis intelligent diagnosis platform that can match the national standard disease type library information while performing real-time monitoring and active adjustment of diagnosis and treatment project costs, in view of the deficiencies of the prior art.

[0018] In order to achieve the above technical purpose, the present application implements the following technical scheme: The medical insurance version of the clinical pathway diagnosis intelligent diagnosis platform of the present application comprises a kind of intelligent operation clinical pathway method as shown in any of the above.

[0019] The above technical scheme produces the following technical effects: The medical insurance version of the clinical pathway diagnosis intelligent diagnosis platform of the present application realizes fine management and real-time monitoring of medical diagnosis information by integrating the above analysis method. The intelligent diagnosis platform can automatically process medical diagnosis data in the hospital information system, match it to the preset medical diagnosis path, and dynamically adjust the benchmark diagnosis and treatment coefficient according to the expert consultation information. This not only improves the precision of medical cost control, but also ensures the accurate matching of national standard disease type library information and actual medical costs of the hospital. In addition, the system can evaluate the project diagnosis and treatment coefficient compliance rate in real time, timely discover and correct unreasonable charging phenomena, thereby optimizing the cost structure and avoiding waste of medical resources. BRIEF DESCRIPTION OF DRAWINGS

[0020] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings: Figure 1 is a workflow diagram of example 1 in the present application; Figure 2 is a workflow diagram of example 3 in the present application; Figure 3 is a workflow diagram of example 4 in the present application. DETAILED DESCRIPTION

[0021] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application. Unless otherwise defined, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs. The terms used in the specification of the present application are only for the purpose of describing the specific embodiments of the present application, and are not intended to limit the present application.

[0022] In order to facilitate the understanding of the technical solutions of the present application, the present application will explain the following terms: National standard disease library information (Medical Insurance Payment Standard): a standardized payment method formulated by the medical insurance management department, used to standardize the reimbursement amount of medical expenses, including payment by disease, payment by project, payment by disease point value (DIP) and other modes.

[0023] Medical diagnosis path (Clinical Pathway): a standardized medical process formulated by medical institutions to standardize medical services according to disease diagnosis and treatment routines and national standard disease library information, including clear diagnosis and treatment projects, medical service content, cost standards and implementation steps, aiming to reduce medical costs and ensure medical quality.

[0024] Benchmark diagnosis and treatment coefficient: a reasonable cost reference index formulated by hospitals through historical data and expert consultations, used to evaluate whether the cost of each diagnosis and treatment project meets the established standard. If the coefficient is met, it is considered reasonable (qualified), otherwise it is considered unreasonable (unqualified).

[0025] Project diagnosis and treatment coefficient compliance rate: after comparing the cost use of each charge item in the patient's diagnosis and treatment process with the benchmark diagnosis and treatment coefficient, the proportion of items meeting the benchmark coefficient is calculated, which is the project diagnosis and treatment coefficient compliance rate.

[0026] Pathway anomaly analysis: according to the diagnosis and treatment coefficient compliance, automatically analyze whether the diagnosis and treatment path exists cost over-standard or abnormal use, and identify and feedback unqualified charge items, so as to adjust the medical path and optimize the medical resource allocation.

[0027] Effective Revenue: The actual revenue a hospital receives from medical insurance departments or patients during actual medical treatment, based on medical insurance payment rules and the hospital's operational strategies. Calculation methods typically involve weighted adjustments for different categories such as service revenue, treatment revenue, examination revenue, and revenue from traditional Chinese medicine and herbal remedies.

[0028] Mapping Relationship: The data association between medical diagnostic information and medical pathways. It automatically maps different diagnostic and treatment features (drugs, medical technologies, consumables, etc.) to the corresponding medical pathways to achieve accurate matching between data and pathways.

[0029] HIS (Hospital Information System): A HIS is a comprehensive information system used for clinical diagnosis and treatment, medical management, financial management, and administrative management in hospitals. It typically includes modules such as patient basic information, outpatient registration, inpatient management, drug management, medical billing, financial statistics, and decision support.

[0030] It is known that the disease-based value-based payment standard has the following significant drawbacks in actual medical diagnosis and hospital operation: The disease classification is too broad and fails to accurately reflect actual medical needs. Current disease-based payment standards often employ broad disease classifications, neglecting individual patient differences and the actual circumstances during diagnosis and treatment. While disease treatment involves significant variations in costs, the broad classifications of the medical insurance payment system fail to effectively differentiate the actual needs of different patients, leading to substantial discrepancies in hospital cost accounting and medical insurance settlement.

[0031] There is a mismatch between the information in the national standard disease database and the actual medical treatment costs: The standardized value-based payment model adopted by medical insurance can easily lead to overspending on some medical items or underspending on others. Due to the medical insurance payment limit, hospitals cannot accurately adjust the diagnosis and treatment items and cost structure, which may result in an imbalance in the allocation of medical resources, and may even induce hospitals to over-treat or reduce the quality of medical services for specific diseases.

[0032] To address the technical shortcomings of the aforementioned medical insurance payment models, current technologies generally employ the construction of intelligent diagnostic platforms for clinical pathway management to optimize hospital resource allocation and cost management. Existing clinical pathway management solutions primarily aim to control medical costs by standardizing clinical medical practices and defining standard treatment procedures for each disease. However, current clinical pathway management technology still suffers from the following serious technical deficiencies in its application to medical insurance point-based payment standards: 1) Clinical pathway management is rough and cannot control medical expenses in detail: The existing clinical pathway management mode generally stays in simply listing diagnosis and treatment projects, and cannot accurately distinguish the reasonable use amount and cost of various medical diagnosis and treatment projects under the medical insurance score payment system. This extensive management method cannot accurately reflect whether each diagnosis and treatment project meets the national standard disease library information, resulting in low accuracy of hospital medical expense control, and actual operation often exceeds the budget or cannot meet the medical insurance payment requirements.

[0033] 2) Lack of real-time and effective diagnosis and treatment project cost evaluation and diagnosis and treatment coefficient standard monitoring mechanism: The existing clinical pathway management technology only relies on static cost control scheme and post-analysis, and cannot realize real-time monitoring and active adjustment of diagnosis and treatment project cost use, nor can it effectively evaluate the medical diagnosis and treatment cost expenditure in real time, so that the hospital cannot timely find and correct unreasonable charging phenomenon in the diagnosis and treatment process, and cannot dynamically optimize the cost structure.

[0034] 3) Insufficient data matching capability, serious disconnection between diagnosis information and medical insurance standards: The actual medical diagnosis information of the hospital and the national standard disease library information have poor data correlation, and the existing technology cannot efficiently realize accurate matching of medical diagnosis data and national standard disease library information, so that the cost in the actual diagnosis and treatment process cannot be effectively mapped to the medical insurance payment score standard, and the hospital cannot accurately master the use of medical insurance reimbursement amount, thereby causing serious problems such as medical insurance settlement error and hospital economic benefit decline.

[0035] Based on this, the present application generates the invention starting point to solve the above technical defects, and designs the corresponding technical scheme.

[0036] The present application will be further described in detail below in combination with specific embodiments, but the embodiments of the present application are not limited thereto.

[0037] Embodiment 1 As shown in the Figure 1 , in order to solve the problem of mismatching between the national standard disease library information and the actual medical diagnosis and treatment cost in the prior art, the present application provides a more detailed medical insurance version clinical pathway diagnosis analysis method. Specifically, the medical insurance version clinical pathway diagnosis analysis method of the present application includes the following steps: S101, dividing medical diagnosis information into a plurality of medical diagnosis paths according to the medical insurance national payment disease score payment standard, constructing charging project information for each medical diagnosis path through historical disease diagnosis case information and expert consultation information, and the charging project information contains charging data of charging projects in the medical diagnosis information; S201. Preprocess the medical diagnosis information of multiple patients in the hospital information system and automatically match the preprocessed medical diagnosis information with the basic diagnosis information. Import the billing data of each billing item in the matched medical diagnosis information into the billing item information. S301. Set benchmark treatment coefficients for each medical diagnostic pathway based on expert consultation information. Based on benchmark treatment coefficient Building benchmark treatment costs ; S401. Under each medical diagnostic pathway, based on the actual usage R of the j-th chargeable item by the patient. real,j Compared to benchmark treatment costs By comparing the results, we can obtain the standard item a for the treatment coefficient of the project. j By analyzing the treatment coefficient of each project under each pathway, the target items a j The project's treatment coefficient compliance rate was obtained by summarizing the results. ; S501, Based on the compliance rate of the treatment coefficient for each item in the medical diagnostic pathway. The medical diagnostic pathway is anomaly identified, and a benchmark treatment coefficient is calculated for the abnormal medical diagnostic pathways. Make adjustments; Once no path anomalies are found in any of the medical diagnostic pathways, a medical diagnostic report is generated. The medical diagnostic report includes a data report and an effective revenue report. The effective revenue report contains summary data of the effective revenue of each medical diagnostic pathway. Experts use the medical diagnostic report to formulate expert consultation information.

[0038] Specifically, this application, through step S101, systematically divides complex medical diagnostic information into multiple specific medical diagnostic pathways based on the disease-based value payment standard in the national medical insurance payment system. This step ensures the accuracy and relevance of subsequent analysis because each medical diagnostic pathway is based on a deep integration of historical disease diagnosis case information and expert consultation information, thus including detailed billing information. This application also includes preprocessing the complex medical diagnostic information in the hospital information system. This process includes data cleaning and format standardization to ensure the accuracy and usability of the information. The preprocessed medical diagnostic information is then automatically matched with basic diagnostic information, achieving precise information matching. The key to this step is that it can effectively associate the patient's actual medical diagnostic information with the preset medical diagnostic pathways, laying a solid foundation for subsequent cost comparison and evaluation.

[0039] Specifically, the information in the national standard disease database of this application mainly refers to: ICD-9, ICD-10, and the information of ICD-11, which will be implemented in the future.

[0040] ICD-9 is the ninth revision of the International Classification of Diseases, a disease classification standard published by the World Health Organization. It systematically classifies and codes various diseases and has been widely used in the global medical field for a period of time. Through ICD-9, medical institutions can standardize disease records, facilitating medical statistics, scientific research, and medical cost accounting. However, with the development of medicine and the deepening of disease understanding, its limitations gradually emerge. ICD-10 is the tenth revision of the International Classification of Diseases, an update based on ICD-9. It further improves the disease classification system, adds new disease types and codes, and better meets the needs of medical development. ICD-10 has also played an important role in global medical work, and China has also explicitly required medical institutions to use it in medical record writing to standardize disease classification management.

[0041] Further, with the continuous discovery of new diseases in the medical field, the understanding of known diseases is also deepening. The old classification standard may not accurately cover new disease types or be precise enough in classifying certain diseases. For example, with the development of genetic diagnosis technology, some disease characteristics based on genes need to be more accurately reflected in the classification standard, so the standard needs to be updated to accommodate the growth of medical knowledge. For example, the promotion of the Chinese version of ICD-11 helps various levels and types of medical institutions to achieve consistency in medical record writing, disease classification coding, and other aspects, improving the overall quality and comparability of medical services. ICD-11 is the eleventh revision of the International Classification of Diseases, the latest disease classification version published by the World Health Organization. It is the result of the continuous accumulation of medical knowledge and changes in the disease spectrum, and it makes more scientific and detailed adjustments to disease classification. China organized relevant experts to compile it into the Chinese version of ICD-11, which is more in line with China's medical reality and helps to improve the standardization level of China's medical services. However, there is no data platform and diagnostic method referring to ICD-11 in the existing technology, which is also the protection scope of the present application.

[0042] Further, the present application sets a benchmark diagnosis and treatment coefficient for each medical diagnosis path based on expert consultation information. This coefficient, as an important reference for cost reasonableness, is crucial for evaluating the use of medical expenses. Based on the benchmark diagnosis and treatment coefficient, the present application further constructs a benchmark diagnosis and treatment cost, providing a clear benchmark for subsequent comparison and monitoring.

[0043] Further, the actual use amount of each medical diagnosis path is compared and analyzed in detail. Through comparison with the benchmark diagnosis and treatment cost, it can be accurately identified which items meet the requirements of the benchmark diagnosis and treatment coefficient and which items have non-compliance problems. Through the collection and arrangement of these information, the item diagnosis and treatment coefficient compliance rate is further obtained, which provides an important basis for subsequent path abnormality judgment. Among them, according to the item diagnosis and treatment coefficient compliance rate in each medical diagnosis path, comprehensive path abnormality judgment is performed on multiple medical diagnosis paths to ensure the rationality and accuracy of medical expenses.

[0044] Further, the technical scheme of the present application also includes the judgment of path abnormality and the adjustment of the medical diagnosis path of path abnormality. Specifically, the principle of the present application for judging path abnormality is based on whether the item diagnosis and treatment coefficient compliance rate in each path of the present application meets a certain standard. If it does not meet the corresponding standard, it is judged that the path is abnormal. For the medical diagnosis path of path abnormality, in the specific implementation process, the benchmark diagnosis and treatment coefficient of the non-compliant charging item in the path needs to be dynamically adjusted to optimize the cost control and eliminate the abnormality.

[0045] Among them, the adjustment formula of the benchmark diagnosis and treatment coefficient is: The adjusted benchmark diagnosis and treatment coefficient of the charging item j is ; Ereal, the actual cost of the charging item j (the actual use amount of the current patient); The calculation method of the benchmark diagnosis and treatment cost of the present application is: Specifically, in the above calculation formula , represents the unit charging standard of the charging item j, that is, the standard cost generated by each use of the item. For example, an electrocardiogram may be designated as ￥200 / time.

[0046] represents the expected use frequency or quantity of the charging item j in the standard clinical path. For example, if the standard requires that an electrocardiogram be performed once per patient, then =1; if a certain nursing service requires to be performed once a day for 7 days of hospitalization, then =7.

[0047] Benchmarking coefficient of item j, i.e. the use intensity or standard that the item should reach in the standard path, determined by expert consultation and historical data. If the patient should fully meet the requirements of a certain service under normal circumstances, then Usually 1; but other values can also be set according to the characteristics of the item. j Used to calculate the total cost that the charge item j should generate if it is fully executed according to the standard. Its purpose is to provide a baseline for subsequent comparison with actual data, and then to judge whether the item charge is reasonable.

[0048] Further, it should be noted that α is the adjustment intensity coefficient (0 < α ≤ 1), which is set by experts to control the sensitivity of the adjustment range. In the specific implementation process, the process of expert consultation and expert setting of the adjustment intensity system involved in the present application can be realized by importing the medical diagnosis report including the data report and the effective income report generated by the present application into the AI agent or different types of large models through the interface, and the suggestions given by the AI agent or the large model are used as the result of the expert consultation combined with the expert setting of the adjustment mechanics system.

[0049] In the specific implementation process, it is assumed that the charge item is "antibiotic drugs", and the original benchmarking coefficient of diagnosis and treatment = 1.0, i.e. 100% of the standard dosage. The benchmarking cost: = 100 yuan.

[0050] The actual cost Ereal=130 yuan, which exceeds the benchmarking cost by 30%. If the adjustment coefficient α = 0.5 (moderate adjustment intensity) is selected; Further, substituting the above data into the formula gives: That is, the adjusted benchmarking coefficient of diagnosis and treatment is 1.15, i.e. the benchmarking cost is increased by 15%, so that the actual cost and the adjusted cost standard are closer, and similar situations in the future can be brought into a reasonable range.

[0051] Further, the verification is performed again: After adjusting the benchmarking coefficient, whether the single item meets the standard is calculated again: The adjusted benchmarking cost is: The adjusted benchmarking cost meets its requirements, then no further adjustment is made, and a j = 1 (the item meets the standard).

[0052] Further, the item diagnosis and treatment coefficient meets the standard rate satisfies: Wherein, m is the number of charge items containing data in the medical diagnosis information in the medical diagnosis path.

[0053] Further, it needs to be explained that the data report in the medical diagnosis report records the use of charge items, actual cost, benchmark cost, project diagnosis and treatment coefficient reaching standard and other information under each medical diagnosis path, providing comprehensive data analysis support for experts. Experts can understand the composition and distribution of medical expenses in depth through careful examination of these data reports, so as to formulate more scientific and reasonable expert consultation information and adjustment strategies. In addition, after generating the medical diagnosis report, the technical scheme of the present application also provides a feedback mechanism for expert consultation information. Experts adjust and optimize key parameters such as benchmark diagnosis and treatment coefficient and adjustment intensity coefficient according to the data and analysis results in the medical diagnosis report, so as to ensure the rationality and accuracy of medical expenses. These adjusted parameters will be applied again in the subsequent medical diagnosis information analysis and cost control process, forming a closed-loop intelligent diagnosis platform for continuous improvement and optimization.

[0054] In summary, the technical scheme of the present application solves the problem of mismatch between medical insurance payment standard and actual medical treatment cost in the prior art by constructing a refined medical insurance version of clinical path diagnosis analysis method. Through a series of steps such as setting benchmark diagnosis and treatment coefficient, constructing benchmark diagnosis and treatment cost, analyzing project diagnosis and treatment coefficient reaching standard rate and path abnormality judgment, the present application realizes accurate control and optimization of medical expenses. At the same time, through the output forms such as data report and effective income report, the present application provides comprehensive data analysis and decision support for hospital managers and experts.

[0055] Embodiment 2 Different from embodiment 1: in order to further improve the calculation accuracy of the summary data of the effective income of each medical diagnosis path of the present application, further, in the present application, each charge item information is associated with diagnosis basic information and medical insurance payment attribute information; the medical insurance payment attribute information sets the effective income proportion of each charge item according to the medical insurance national payment standard of payment according to disease type score.

[0056] Further, the summary data calculation method of the effective income of each medical diagnosis path is: Wherein, the effective income represents the result of the summary data, The medical diagnosis path is counted into the effective income in the medical insurance national payment standard of payment according to disease type score; The calculation method of each medical diagnosis path is: Wherein, Benchmark diagnosis and treatment fee for the jth charge item in the medical diagnosis path, represents the medical insurance payment attribute information of the jth charge item. It is worth noting that, represents the proportion of the cost of the charge item that can be included in the "effective income" of the hospital according to the national medical insurance (DIP) rules. For example, examination items are generally included in the effective income at 40%, treatment items are included at 100%, and drugs and consumables are generally not included (weight 0%). The parameter reflects the calculation rules when medical insurance is paid, and ensures that the final calculation of effective income meets the national standard disease library information. By converting the benchmark fee , the income contribution that meets the medical insurance payment accounting requirements is obtained.

[0057] Further, when the medical diagnosis process of the patient includes multiple stages, the summary data calculation method of the effective income of each medical diagnosis path included in the effective income report is: Where L is the number of stages included in the medical diagnosis process.

[0058] Other than example 1, this embodiment will not be described again.

[0059] Example 3 As Figure 2 shown, different from example 1: in order to further improve the matching degree of the technical scheme of the present application with the disease type score payment standard, further, the medical diagnosis path includes drug path, technique path, consumption path, nursing path, medical path and day path; The drug path includes the charge data of the drug characteristics in the medical diagnosis information; The technique path includes the charge item information of the medical examination characteristics in the medical diagnosis information; The consumption path includes the charge item information of the consumable characteristics in the medical diagnosis information; The nursing path includes the charge item information of the nursing service characteristics in the medical diagnosis information; The medical path includes the charge item information of the physician operation characteristics in the medical diagnosis information; The day path includes the charge item information of the daily characteristics of the hospitalization in the medical diagnosis information.

[0060] In the specific implementation process, assuming that the hospital formulates a clinical path for the "coronary heart disease bypass surgery" disease type, which needs to cover each key link of the patient during hospitalization. In order to fine management, we will split this path into the following six sub-paths (six paths): Drug path: involves all drug-related costs, such as antibiotics, analgesics, prophylactic drugs, etc.

[0061] Drug path: related to drug expenses, such as drug costs, etc.

[0062] Consumable path: related to disposable consumable and medical instrument expenses, such as surgical kit, implantation equipment, etc.

[0063] Nursing path: related to nursing service expenses, such as nursing fee, bed fee, rehabilitation physiotherapy, etc.

[0064] Doctor path: related to doctor operation and treatment expenses, such as operation fee, anesthesia fee, operation operation fee, etc.

[0065] Day path: related to daily expenses of hospitalization, such as hospital management fee, diagnosis fee, etc.

[0066] Therefore, the present application further constructs the following mapping relationship (here is an example to illustrate, and does not represent that the present application is limited to the following contents): Item code Item name Category Description Budget cost P001 Aspirin Drug For anti-platelet therapy ￥20 / visit P002 Antibiotic Drug To prevent surgical infection ￥20 / visit P010 Electrocardiogram Examination For cardiac function evaluation ￥200 / visit P011 Chest CT Examination For cardiovascular disease diagnosis ￥200 / visit P020 Surgical kit Consumable Disposable surgical consumables ￥1000 / visit P021 Implant device Consumable Artificial joint ￥1000 / visit P030 Surgical fee Treatment Operation fee of attending physician ￥50000 / case P031 Anesthesia fee Treatment Anesthesia service cost ￥1000 / day P040 Nursing fee Nursing Inpatient nursing service cost ￥500 / day Table 1 In the above table, the medical diagnosis information of the present application is an example, and after the pre-processing of such data and the automatic matching of the pre-processed data with each medical path of the present application, the following is obtained: Six categories Map items Example item name Benchmark diagnosis and treatment coefficient requirement Budget cost Effective income percentage (medical insurance included) Drug path Drug Aspirin Once a day; charge reaches preset dosage ￥20 / visit 0% (not included in DIP) Technical path Examination Electrocardiogram Once per case ￥200 / visit 40% Consumable path Consumable Implant device Fixed usage amount per case ￥1000 / visit 0% Medical path Treatment Surgical fee Once per case ￥50000 / case 100% Nursing path Nursing Nursing fee Once a day nursing ￥1000 / day 100% Day path Hospitalization day Hospital management fee Standard hospitalization days (e.g. 7 days) ￥500 / day 100% Drug path Drug Antibiotic Fixed usage amount per case ￥1000 / visit 0% Technical path Examination Chest CT Once per case ￥200 / visit 40% Medical path Treatment Anesthesia fee Once per case ￥50000 / case 100% Consumable path Consumable Surgical kit Fixed usage amount per case ￥1000 / visit 0% Table 2 As can be seen from Table 2, specifically, the present application constructs six sub-paths (six paths) in advance before importing the HIS system data, which is based on the clinical path theory and the cost composition, and further subdivides the traditional clinical path according to the treatment process corresponding to the disease and the medical insurance payment rules. The data imported by the present application is processed as a charge item data dictionary (charge item), each item has a unique code, name, and belongs to a category (drug, examination, consumable, treatment, nursing, hospitalization), and thus the above charge item is imported into the corresponding charge item information, and the medical insurance attribute of each charge item can be further obtained. Each charge item information is associated with diagnosis basic information and medical insurance payment attribute information; the medical insurance payment attribute information sets the effective income proportion of each charge item according to the medical insurance national payment standard of payment according to disease score.

[0067] Further, the method for matching each charge data in the medical diagnosis information to the charge item information in different medical diagnosis paths is as follows: S11, set a medical diagnosis path template, the medical diagnosis path template includes a feature vector of drug characteristics in a drug path, a feature vector of medical examination characteristics in a technique path, a feature vector of consumable characteristics in a consumable path, a feature vector of nursing service characteristics in a nursing path, a feature vector of doctor operation characteristics in a doctor path, and a feature vector of daily characteristics in a daily path; S21, the feature vector of each charge item in the medical diagnosis information is calculated with the feature vector of each medical diagnosis path in the medical diagnosis path template; S31, each charge item in the medical diagnosis information is imported into the charge item information of the medical diagnosis path with the maximum similarity calculation value; S41, for the charge items with the error between the similarity calculation results of the charge items and the feature vectors of multiple medical diagnosis paths being less than the second threshold value, the charge items are manually imported into the charge item information of the medical diagnosis path by an administrator.

[0068] Notably, in the above technical solutions, for each charge item in the medical diagnosis information, the system will extract its key features to form a feature vector. Assuming that the feature vector of the charge item j is denoted as: where each component represents the value of the item on a certain characteristic (feature dimension). These characteristics are obtained through structured data or natural language processing (such as word segmentation and keyword extraction of medical order text), and are standardized to numerical indicators through preprocessing, thereby forming the feature vector of the charge item.

[0069] Further, the present application predefines diagnosis path templates, each of which represents a typical diagnosis and treatment path corresponding to different cost categories. k is the number of feature vectors in the medical diagnosis path, t1 is the feature vector of the drug feature in the drug path, t2 is the feature vector of the medical examination feature in the skill path, t3 is the feature vector of the consumable feature in the consumption path, t4 is the feature vector of the nursing service feature in the nursing path, t5 is the feature vector of the physician operation feature in the physician path, and t6 is the feature vector of the daily feature in the daily path. Among them, each t is a vector with the same dimension as the feature vector of the charge item, reflecting the typical attributes of the path.

[0070] Further, in order to automatically classify a charge item j into the corresponding path, the present application uses similarity calculation, and the formula is: wherein, is the feature vector of the charge item j; is the feature vector of each medical diagnosis path in the medical diagnosis path template; In the specific implementation process, if the charge item P001 (such as aspirin cost information) obtains the vector after feature extraction: t1=(1.0,0.0,0.0,0.0,0.0,0.0). And other path templates, such as skill path: t2 = (0.0, 1.0, 0.0, 0.0, 0.0, 0.0) The result of the cosine similarity calculation is: And the similarity with other templates t2, t 3, will be much lower. Therefore, the platform classifies P001 into the drug path.

[0071] The same as Embodiment 1, this embodiment will not be repeated.

[0072] Embodiment 4 As shown in Figure 3 , different from Embodiment 1, in order to further improve the completeness and accuracy of the medical diagnosis information of the present application, further, the method for preprocessing the medical diagnosis information is: S1, the medical diagnosis information in the hospital information system constitutes an original data set, and the medical diagnosis information of each record in the original data set is marked as X. The original data set is imported into the platform data processing module through the interface or manual uploading mode; Wherein, the original data set is expressed as: {X1, X 2,... X N}, N is the number of medical diagnosis information; S2, for each data in the original data set, the abnormal value judgment formula is: Wherein, Z is the abnormal value of the medical diagnosis information in the original data set; μ is the average value of each medical diagnosis information in the original data set; is the average deviation amplitude between each medical diagnosis information in the original data set and the average value μ; S3, the medical diagnosis information with |z|>3 is marked as abnormal, and the missing value filling or manual filling is carried out on the medical diagnosis information, and the missing value filling formula is: Wherein, M is the number of non-missing records in the medical diagnosis information, is the data of the medical diagnosis information after missing value filling, indicates the non-missing data in the medical diagnosis information.

[0073] In the implementation process, the mean value as the central tendency in the above loss filling formula is obtained by calculating the average value of the existing data, and the central tendency of the field can be obtained, that is, the most likely value of the missing data. Under the condition that the data obeys the normal distribution, the premise assumption of the mean filling is that the data approximately obeys the normal distribution under a large sample, and at this time the mean value can better reflect the overall level. If the data has serious skewness or outliers, data cleaning or other filling methods (such as median filling) need to be used first. The mean filling method is simple and easy to implement, and is suitable for the case where the missing proportion is low. But the disadvantage is that it may underestimate the variance of the data, so it may need to be corrected in subsequent statistical analysis.

[0074] Further, for normally distributed data, most data values are concentrated around the mean, and data exceeding 3 standard deviations has a very low probability of occurrence (about 0.3%), so the condition of |z|>3 is used to judge the abnormal value. When the absolute value of the zzz value of a data point in the actual data is greater than 3, it means that it deviates too much compared with other data, so it is marked as an abnormal value.

[0075] Specifically, if some data does not obey the normal distribution, the technical personnel need to manually assist in removing or supplementing the abnormal values of the data.

[0076] Other than Embodiment 1, this embodiment will not be described again.

[0077] Embodiment 5 Different from Embodiment 1: In order to further improve the applicability of the technical scheme of the application, further, the application also designs a smart operation clinical path platform, which includes any of the above smart operation clinical path methods.

[0078] Based on the above technical scheme, the main functional modules of the smart diagnosis platform include a data import module, a preprocessing module, a path matching module, and a report generation module. The data import module is responsible for importing medical diagnosis information in the hospital information system (HIS) into the smart diagnosis platform, providing basic data for subsequent analysis. The preprocessing module cleans and standardizes the imported data, including operations such as abnormal value judgment and missing value filling, to ensure the accuracy and integrity of the data. The path matching module is the core part of the smart diagnosis platform, which uses pre-defined medical diagnosis path templates and similarity calculation algorithms to automatically classify each charge item in the medical diagnosis information into the corresponding medical diagnosis path. Finally, the report generation module generates a report containing the effective income summary data of each medical diagnosis path based on the matching results, providing decision support for hospital managers.

[0079] In the implementation process, the intelligent diagnosis platform can also be seamlessly connected with the HIS system of the hospital to realize real-time synchronization and updating of data. At the same time, the intelligent diagnosis platform also provides a friendly user interface to facilitate users to import, query and analyze data. In addition, the intelligent diagnosis platform also supports storage and management of historical data to facilitate users to conduct long-term data analysis and trend prediction.

[0080] When all medical diagnosis paths do not have path abnormalities, the intelligent diagnosis platform generates a detailed medical diagnosis report. This report not only contains a data report, but also provides an effective income report, providing comprehensive cost analysis and optimization suggestions for hospitals and experts.

[0081] Other than the same as example 1, this embodiment is not described.

[0082] Those skilled in the art should understand that the embodiments of the present application can provide a method, an intelligent diagnosis platform or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer usable program code.

[0083] The present application is described with reference to flowcharts and / or block diagrams of methods, devices (intelligent diagnosis platform) and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be realized 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 device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0084] These computer program instructions can also be stored in a computer readable memory that can guide the computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer readable memory produce a product including instruction devices, which implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one flow or multiple flows and / or blocks Figure 1 The functions specified in one flow or multiple flows and / or blocks

[0085] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate computer-implemented processes, so that the instructions executed on the computer or other programmable data processing devices provide operational steps for implementing the functions specified in the flowchart Figure 1 one flow or multiple flows and / or blocks ​ one block or multiple blocks.

[0086] The above only is the preferred embodiment of the present application, and is not used to limit the present application, and for the person skilled in the art, the present application can have various changes and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A smart operation clinical pathway method, characterized in that, It comprises the following steps: S101, according to the national standard disease library information, the medical diagnosis information is divided into a plurality of medical diagnosis paths, each of the medical diagnosis path is constructed by historical disease diagnosis case information and expert consultation information Charge item information, the charge item information contains the charge data of the charge item in the medical diagnosis information; S201, the medical diagnosis information of a plurality of patients in the hospital information system is preprocessed, and the preprocessed medical diagnosis information is automatically matched with the diagnosis basic information, and the charge data of each charge item in the matched medical diagnosis information is imported into the charge item information; S301、According to the expert consultation information, set a benchmark diagnosis and treatment coefficient under each medical diagnosis path , based on the benchmark diagnosis and treatment coefficient , construct a benchmark diagnosis and treatment cost ; S401、In each of the medical diagnosis path, through the patient j actual charge item use R real,j With the benchmark diagnosis and treatment costs Comparison of project diagnosis and treatment coefficient standard a j , through the project diagnosis and treatment coefficient standard a j In each of the medical diagnosis path, the project diagnosis and treatment coefficient standard rate ; S501、According to the item diagnosis and treatment coefficient standard rate in each medical diagnosis path Abnormal judgment is made on the medical diagnosis path, and the benchmark diagnosis and treatment coefficient of the medical diagnosis path with path abnormalities Adjustment; When each of the medical diagnosis path does not appear path abnormality, a medical diagnosis report is generated; the medical diagnosis report includes data report and effective income report, and the effective income report contains the summary data of the effective income of each medical diagnosis path, and the expert formulates expert consultation information through the medical diagnosis report. 2.The intelligent operation clinical pathway method of claim 1, wherein, Each of the charge item information is associated with diagnosis basic information and medical insurance payment attribute information; the medical insurance payment attribute information sets the effective income proportion of each charge item according to the medical insurance national payment standard of payment according to disease type. 3.The intelligent operation clinical pathway method of claim 2, wherein, The summary data calculation method of the effective income of each medical diagnosis path is: wherein the effective revenue represents a result of the aggregated data, The first medical diagnosis path is counted into the effective revenue in the medical insurance country payment by disease type score payment standard. Each of the medical diagnostic pathways is calculated as: wherein, the benchmark diagnosis and treatment cost of the charge item in the medical diagnosis path, is the medical insurance payment attribute information representing the jth charge item. 4.The intelligent operation clinical pathway method of claim 1, wherein, The medical diagnosis path includes drug path, skill path, consumption path, nursing path, medical path and daily path; The drug path contains the charge data of the drug characteristics in the medical diagnosis information; The skill path contains the charge item information of the medical skill inspection characteristics in the medical diagnosis information; The consumption path contains the charge item information of the consumption characteristics in the medical diagnosis information; The nursing path contains the charge item information of the nursing service characteristics in the medical diagnosis information; The medical path contains the charge item information of the doctor operation characteristics in the medical diagnosis information; The daily path contains the charge item information of the daily characteristics of hospitalization in the medical diagnosis information.

5. The method of claim 4, wherein, The method for matching each of the charge data in the medical diagnosis information to the charge item information in different medical diagnosis paths is: S11, set up a medical diagnosis path template, the medical diagnosis path template includes the feature vector of the drug characteristics in the drug path, the feature vector of the medical skill inspection characteristics in the skill path, the feature vector of the consumption characteristics in the consumption path, the feature vector of the nursing service characteristics in the nursing path, the feature vector of the doctor operation characteristics in the medical path and the feature vector of the daily characteristics of hospitalization in the daily path; S21, the feature vector of each charge item in the medical diagnosis information is calculated with the feature vector of each medical diagnosis path in the medical diagnosis path template; S31, each charge item in the medical diagnosis information is imported into the charge item information of the medical diagnosis path with the maximum similarity calculation value; S41, for the error between the similarity calculation results of the charge item and the feature vector of a plurality of medical diagnosis paths is less than the second threshold value, the charge item is manually imported into the charge item information of the medical diagnosis path by the administrator.

6. The method of claim 5, wherein, The formula of the similarity calculation is: wherein, is a feature vector of the charge item j; is a feature vector of each medical diagnosis path in the medical diagnosis path template; k is the number of feature vectors in the medical diagnosis path, t1 is a feature vector of the drug characteristics in the drug path, t2 is a feature vector of the medical examination characteristics in the medical technology path, t3 is a feature vector of the consumable characteristics in the consumable path, t4 is a feature vector of the nursing service characteristics in the nursing path, t5 is a feature vector of the physician operation characteristics in the physician path, and t6 is a feature vector of the daily characteristics in the daily path.

7. The method of claim 1, wherein, The method for preprocessing the medical diagnosis information is: S1, the medical diagnosis information in the hospital information system is constituted as an original data set, the medical diagnosis information of each record in the original data set is marked as X, and the original data set is imported into a platform data processing module through an interface or a manual uploading manner; Wherein, the original data set is expressed as: {X1, X 2,.., X N}, N is the number of the medical diagnosis information; S2, for each data in the original data set, an outlier judgment is performed, and the outlier judgment formula is: Where Z represents the medical diagnostic information in the original dataset. The outlier; μ is the average value of each medical diagnostic information in the original dataset; This represents the average deviation between each medical diagnosis information in the original dataset and the average value μ. S3, the medical diagnosis information with |z|>3 is marked as an exception, and the medical diagnosis information is filled with missing values or manually filled, and the missing value filling formula is: Wherein, M is the non-missing record number in the medical diagnosis information, is the medical diagnosis information The data after the missing value supplement, represents the non-missing data in the medical diagnosis information.

8. A smart operation clinical pathway platform, characterized in that, A smart operation clinical path method comprising any one of claims 1-7.

Images