Standardized consumption path management method and system based on disease species

Abstract

The invention discloses a standardized consumption path management method and system based on disease categories, and belongs to the technical field of medical equipment and consumable management, and the method comprises the steps: obtaining related data of a target operation case, constructing an operation characteristic parameter set, and building a typical clinical path knowledge base matched with the corresponding disease category; historical operation video data of a target doctor are analyzed, a deviation value sequence of the actual consumable usage amount and a standard consumable baseline value is counted, and a dynamic skill score value of the target doctor is calculated in combination with the complication occurrence rate and the intraoperative operation fluency so as to correct an allowable fluctuation interval of the standard consumable baseline value; and dynamically associating the corrected interval with the knowledge base to form a personalized consumption guidance scheme. According to the method, a multi-modal data analysis and dynamic skill evaluation algorithm is adopted, a standard consumable baseline interval can be automatically corrected, a personalized consumption guide scheme can be generated, accurate balance between consumption and doctor skill adaptability is achieved, resource waste is reduced, and operation safety is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical device and consumable management, and particularly to a method and system for standardized consumption path management based on disease types. Background Art

[0002] Currently, the management of consumable use in clinical surgeries generally faces the problems of lack of standards and insufficient dynamic adaptation. Under the traditional model, medical institutions usually set fixed consumable consumption baseline values according to disease types, but fail to make dynamic adjustments in combination with doctors' operation skills, actual intraoperative scenarios, and postoperative recovery data. Due to the lack of integrated analysis of multi-dimensional data, the operation differences among different doctors under the same disease type are likely to lead to waste or overuse of consumables, and it is difficult to accurately ensure surgical efficiency and patient safety.

[0003] Existing solutions mostly adopt static clinical path templates, classify surgical disease types based on the International Classification of Diseases codes and match preset consumable lists, and some systems rely on manual experience to conduct post hoc audits of abnormal consumption situations. Although such methods can achieve coarse-grained standardized management, they do not build the cross-modal correlation analysis ability of surgical video temporal parsing, electronic medical record semantic extraction, and actual consumable consumption. Especially when dealing with complex surgical variation scenarios, it is impossible to real-time identify the potential impact of intraoperative operation deviations on consumable requirements, resulting in the guidance plan lagging behind the actual clinical progress.

[0004] The core drawbacks of traditional management methods are mainly reflected in the lack of a dynamic adaptation mechanism and a personalized evaluation system. The fixed consumable baseline value does not consider the dynamic changes in doctors' skill levels and cannot form an elastic fluctuation range based on historical operation quality, which may not only limit the clinical decision-making space of high-level physicians but also be difficult to effectively constrain beginners. The manual experience-based audit process is difficult to scale and process multi-source heterogeneous data, and the efficiency of the correlation analysis between abnormal consumable consumption and postoperative complications is low, making it difficult to achieve the goal of refined management of medical resources. Summary of the Invention

[0005] To solve the above problems, the present invention provides a method and system for standardized consumption path management based on disease types, which adopt multi-modal data analysis and dynamic skill evaluation algorithms, can automatically correct the standard consumable baseline interval and generate personalized consumption guidance plans, achieve an accurate balance between consumption and doctors' skill adaptability, reduce resource waste, and improve surgical safety.

[0006] The above objectives can be achieved through the following solutions: The disease-based standardized consumption path management method includes obtaining the electronic medical record text data, surgical video data, consumable usage records, and postoperative monitoring data of target surgical cases, performing standardized processing on the data to generate a structured data set; identifying the International Classification of Diseases codes from the structured data set, extracting data such as the surgical duration, wound area, and instrument usage categories corresponding to the classification codes to obtain an operation feature parameter set; based on the operation feature parameter set, establishing a typical clinical path knowledge base for the corresponding disease, where the typical clinical path knowledge base contains a standard consumable baseline value and an operation abnormality comparison table; analyzing the historical surgical video data of the target doctor, and statistically obtaining a deviation value sequence of the actual consumable usage amount and the standard consumable baseline value; combining the deviation value sequence, the postoperative complication incidence rate, and the intraoperative operation fluency to generate a three-dimensional skill evaluation matrix, and calculating the dynamic skill score value of the target doctor according to the three-dimensional skill evaluation matrix; based on the dynamic skill score value, correcting the allowable fluctuation range of the standard consumable baseline value; dynamically associating the corrected allowable fluctuation range with the typical clinical path knowledge base to form a personalized consumption guidance plan.

[0007] Optionally, the obtaining the electronic medical record text data, surgical video data, consumable usage records, and postoperative monitoring data of target surgical cases, and performing standardized processing on the data to generate a structured data set includes: retrieving the original medical record text data of the target surgical case from the hospital's electronic medical record system, and extracting a keyword set including surgical stage markers, lesion site descriptions, and medication records; obtaining the surgical video data and performing frame sequence parsing to generate a timestamp association table matching the surgical stage markers; synchronously retrieving the consumable requisition detail records marked with the surgical number in the consumable management system, and mapping the consumable names in the consumable requisition detail records to a preset standardized consumable coding library to generate a unified consumable coding sequence; based on the time range of the timestamp association table, locating the vital sign sequence and drug infusion record associated with the target surgery in the postoperative monitoring data, and performing interpolation filling on the missing time period data; performing field association on the keyword set, consumable coding sequence, vital sign sequence, and drug infusion record, and converting them into a JSON format data stream in a unified time series to generate the structured data set; where each record of the structured data set includes synchronous time series fields such as surgical stage markers, consumable codes, vital sign parameters, and medication records.

[0008] Optionally, identifying the International Classification of Diseases codes from the structured dataset, extracting data such as the operation duration, wound area, and instrument usage category corresponding to the classification codes, and obtaining an operation feature parameter set includes: reading a preset rule mapping table in the International Classification of Diseases code library, where the rule mapping table contains standard clinical pathway parameters corresponding to each International Classification of Diseases code; matching the International Classification of Diseases codes in the structured dataset with the rule mapping table, and selecting the International Classification of Diseases code corresponding to the main lesion site; intercepting the operation duration threshold range, wound grading standard, and instrument type list corresponding to the selected International Classification of Diseases code to generate an operation feature parameter set.

[0009] Optionally, establishing a typical clinical pathway knowledge base for the corresponding disease types includes: scanning all postoperative infection and secondary surgery records in the historical complication case library to generate a high-risk case identification set; locating the preoperative instrument calibration data and intraoperative anesthesia records corresponding to the high-risk case identification set, and extracting abnormal parameters exceeding the preset range; marking the associated positions of the abnormal parameters in the operation abnormality comparison table to form a knowledge base architecture with risk level labels, and obtaining a typical clinical pathway knowledge base.

[0010] Optionally, statistically calculating the deviation value sequence between the actual usage amount of consumables and the standard consumable baseline value includes: retrieving the consumable requisition records of all surgeries of the same type by the target doctor within a preset first time window in the past, and parsing the actual usage quantity; calibrating the distribution loss quantity and emergency replenishment quantity in the actual usage quantity to obtain a net consumable consumption value sequence; performing a point-by-point difference calculation between the net consumable consumption value sequence and the standard consumable baseline value to generate a deviation value sequence.

[0011] Optionally, generating a three-dimensional skill evaluation matrix by combining the deviation value sequence, postoperative complication incidence rate, and intraoperative operation smoothness includes: decomposing the frame sequence data of the surgical video, identifying the number of times the interval duration of the main surgeon's gesture switching exceeds the preset silent threshold, and calculating the operation smoothness index; summarizing the number of monitor alarms and drug addition records within a preset second time window after the corresponding surgery to quantitatively obtain the complication incidence rate index; constructing a three-dimensional coordinate system with the deviation values in the deviation value sequence, the operation smoothness index, and the complication incidence rate index as axes; using the three-dimensional coordinate system to generate a three-dimensional skill evaluation matrix for the target doctor.

[0012] Optionally, the allowable fluctuation range threshold for correcting the standard consumable baseline value includes: setting a mapping relationship between the dynamic skill score value and the reduction amount of the upper limit of the corresponding consumable usage to obtain a threshold contraction coefficient; when the dynamic skill score value is greater than a preset expert threshold, activating an alternative parameter set for the standard consumable baseline value, where the alternative parameter set is from a high-score physician practice database; using the threshold contraction coefficient and the alternative parameter set to correct the allowable fluctuation range and generate a new allowable fluctuation range with a version identifier.

[0013] Optionally, the formation of the personalized consumption guidance plan includes: embedding the corrected allowable fluctuation range into a preset surgical scheduling system to generate a visual usage progress bar; collecting intraoperative sensing data and associating it with the typical clinical pathway knowledge base, adjusting the allowable fluctuation range when abnormalities occur, and updating the usage progress bar; pushing guidance suggestions including the reasons for adjustment.

[0014] Optionally, the method further includes: monitoring the consumable usage data stream in the consecutive surgical records of the target doctor, comparing the actual usage with the corrected allowable fluctuation range case by case, and calculating the consumption deviation degree; verifying whether the consumption deviation degree is less than a preset deviation threshold and the number of postoperative complication marks is zero; when the number of consecutive times of passing the verification result is greater than a preset number threshold, updating the skill label of the target doctor in the typical clinical pathway knowledge base and synchronizing the new version of the threshold to all associated systems.

[0015] Based on the same inventive concept, the present invention also provides a disease-based standardized consumption path management system, which includes: a data collection module for obtaining the electronic medical record text data, surgical video data, consumable usage records, and postoperative monitoring data of the target surgical case, and performing standardized processing on the data to generate a structured data set; a feature extraction module for identifying the international disease classification code from the structured data set and extracting the data of the surgical duration, wound area, and instrument usage category corresponding to the classification code to obtain an operation feature parameter set; a knowledge base construction module for establishing a typical clinical pathway knowledge base matching the corresponding disease based on the operation feature parameter set; a consumption analysis module for analyzing the historical surgical video data of the target doctor and statistically obtaining a deviation value sequence of the actual consumable usage amount from the standard consumable baseline value; a skill evaluation module for generating a three-dimensional skill evaluation matrix by combining the deviation value sequence, the postoperative complication incidence rate, and the intraoperative operation fluency, and calculating the dynamic skill score value of the target doctor according to the three-dimensional skill evaluation matrix; a dynamic threshold calculation module for correcting the allowable fluctuation range of the standard consumable baseline value based on the dynamic skill score value; and a consumption plan generation module for dynamically associating the corrected allowable fluctuation range with the typical clinical pathway knowledge base to form a personalized consumption guidance plan.

[0016] Compared with the prior art, the present invention has the following advantages: 1. The present invention constructs a high-precision surgical consumables consumption model through time-series alignment and cross-modal analysis of multi-source data; accurately associates video operation clips with consumables collection records, eliminates statistical deviations caused by information islands in traditional management, ensures the scientific nature of benchmark values ​​and the reliability of dynamic corrections, and provides a reliable quantitative basis for hospital resource allocation; 2. The present invention integrates a three-dimensional skill evaluation mechanism to achieve personalized consumption control. It not only monitors the consumable deviation, but also incorporates the operation fluency and complication risk into the evaluation system to form a panoramic portrait of the doctor's skills. Based on this, the consumable limit range is flexibly adjusted, which not only avoids rigid standards restricting the clinical decision-making of high-level doctors, but also restrains the waste of resources caused by non-standard operations; 3. The present invention establishes a dynamic knowledge base update path with closed-loop feedback. By continuously monitoring the deviation of consumption and postoperative results, it automatically verifies the effectiveness of the adjustment strategy. After meeting the standards, the doctor's skill label is upgraded and the threshold standards of the entire hospital are revised simultaneously, forming a self-iterative intelligent management cycle to promote the continuous optimization of clinical pathways.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are 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.

[0019] Figure 1 It is a flowchart of a standardized consumption path management method based on disease types according to an embodiment of the present invention.

[0020] Figure 2 This is a three-dimensional skill evaluation diagram of a surgeon according to an embodiment of the present invention.

[0021] Figure 3 It is a structural diagram of a disease-based standardized consumption path management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0023] Referring to Figure 1 , an embodiment of the present invention proposes a standardized consumption path management method based on disease types. By using multi-modal data analysis and dynamic skill evaluation algorithms, it can automatically correct the standard consumable baseline interval and generate a personalized consumption guidance plan, achieving an accurate balance between consumption and doctor skills, reducing resource waste, and improving surgical safety.

[0024] The specific steps of the method in this embodiment include: Obtain the electronic medical record text data, surgical video data, consumable usage records, and postoperative monitoring data of the target surgical case, and perform standardized processing on the data to generate a structured data set; Identify the International Classification of Diseases (ICD) codes from the structured data set, and extract the data of the surgical duration, wound area, and instrument usage categories corresponding to the classification codes to obtain an operation feature parameter set; Based on the operation feature parameter set, establish a typical clinical path knowledge base corresponding to the disease types, and the typical clinical path knowledge base includes a standard consumable baseline value and an operation abnormality comparison table; Analyze the historical surgical video data of the target doctor, and count the deviation value sequence of the actual consumable usage amount and the standard consumable baseline value; Combine the deviation value sequence, postoperative complication incidence rate, and intraoperative operation smoothness to generate a three-dimensional skill evaluation matrix, and calculate the dynamic skill score value of the target doctor according to the three-dimensional skill evaluation matrix; Based on the dynamic skill score value, correct the allowable fluctuation interval of the standard consumable baseline value; Dynamically associate the corrected allowable fluctuation interval with the typical clinical path knowledge base to form a personalized consumption guidance plan.

[0025] Specifically, a structured data set is constructed by integrating electronic medical records, surgical videos, and consumable records. The operation feature parameters are matched by combining the International Classification of Diseases (ICD) codes. A clinical path knowledge base including standard baselines and abnormality comparisons is established. The doctor's operation level is quantified using deviation sequences and three-dimensional skill matrices, and the consumable fluctuation interval is dynamically corrected to finally generate a consumption plan adapted to individual skills.

[0026] Optionally, obtaining the electronic medical record text data, surgical video data, consumable usage records, and postoperative monitoring data of the target surgical case, and performing standardized processing on the data to generate a structured data set includes: Retrieving the original medical record text data of the target surgical case from the hospital's electronic medical record system, and extracting a keyword set including surgical stage markers, lesion site descriptions, and medication records; Specifically, after accessing the hospital electronic medical record database through the API interface, a natural language processing algorithm is used to parse the surgical report text. Regular expressions are used to capture the surgical process paragraphs starting with "Stage X", and the incision position and lesion tissue description are marked. At the same time, sentences containing the generic name of the drug are extracted to form a keyword triple. Among them, the electronic medical record system is a digital management platform for medical institutions to record patient diagnosis and treatment information; the surgical stage marker is an identifier for surgical process steps divided based on specific grammar rules; the lesion site description is the anatomical location information of the lesion marked by pathological terms; the keyword set is a standardized term combination extracted from unstructured text.

[0027] Obtaining the surgical video data and performing frame sequence parsing to generate a timestamp association table matching the surgical stage markers; Specifically, after using the OpenCV library to read the surgical video file, it is decoded into a time-series image sequence at a rate of 30 frames per second. The types of surgical instruments and operation actions in each frame are detected through a convolutional neural network, and in combination with the surgical stage markers, a mapping relationship table between the start frame number of the stage and the system time is established. Among them, frame sequence parsing is a technical process of decomposing a continuous video stream into discrete time-point images; the timestamp association table is a structured data table recording the correspondence between the start time of the surgical stage and the video frame number.

[0028] Synchronously retrieving the consumable usage details record marked with the surgical number in the consumable management system, and mapping the consumable names in the consumable usage details record to a preset standardized consumable coding library to generate a unified consumable coding sequence; Specifically, according to the consumable batch number in the operating room RFID scan record, the procurement name of the corresponding item is retrieved in the material management database. Using a coding matching algorithm based on a tree structure, the commodity name is converted into a term in the International Medical Device Nomenclature, and a coding string conforming to ISO 15225 is output. Among them, the standardized consumable coding library is a set of standardized codes conforming to the international medical material classification system; the consumable coding sequence is a queue of unique identifiers sorted by usage time.

[0029] Based on the time range of the timestamp association table, locating the vital sign sequence and drug infusion record associated with the target surgery in the postoperative monitoring data, and performing interpolation filling on the missing time period data; Specifically, using the start and end times of the surgery as anchor points, data within 24 hours after the surgery is screened in the intensive care system. For the missing time periods of the invasive blood pressure sampling points generated by the electrocardiogram monitor, the cubic spline interpolation method is used to generate continuous waveforms.

[0030] Associate the keyword set, consumable coding sequence, vital sign sequence, and drug infusion record by fields, and convert them into a JSON format data stream in a unified time series to generate the structured data set; Among them, each record of the structured data set includes synchronous time series fields of surgical stage markers, consumable coding, vital sign parameters, and medication records.

[0031] Specifically, establish a unified time axis in seconds, and align the discrete events of each data source according to the occurrence time. Use a nested JSON structure to organize the data, with the primary key being the timestamp, and the sub-keys including fields such as surgical stage, consumable use, and physiological parameters. The data sampling rate is uniformly 1Hz. Among them, the unified time series is a processing method for multi-source data to be aligned based on absolute time; the JSON format data stream is a data transmission form that represents structured data using specific syntax specifications.

[0032] Exemplarily, assume a case of cholecystectomy. First, extract 5 steps including "excising the gallbladder bed" as stage markers from the electronic medical record; secondly, identify the usage period of the hemostatic forceps through video analysis and align it with the stage markers; then map the "disposable electrocoagulation forceps" to the A7352 coding; the postoperative blood pressure data is missing at 15:00, and it is interpolated to get 113 / 74 mmHg at 15:00 through 112 / 75 mmHg at 14:58 and 115 / 73 mmHg at 15:02 adjacent to it; finally, generate a JSON data set containing 2000 time points, and each record includes the stage status and corresponding consumables and vital signs. This method constructs a high-precision surgical consumable usage analysis system through multi-modal data fusion and intelligent analysis. Establish a unified data structure based on spatio-temporal association, enabling a multi-dimensional mapping relationship among doctors' operation behaviors, consumable consumption, and patients' physiological indicators. While reducing the workload of manual collation, it eliminates the semantic ambiguity of different data sources and ensures the input quality of the analysis model. Use dynamic interpolation technology to handle data missing problems, ensuring temporal integrity and avoiding introducing statistical biases. The finally generated structured data set is compatible with various machine learning algorithms, providing a standardized data basis for subsequent intelligent decision-making.

[0033] Optionally, identify the International Classification of Diseases (ICD) coding from the structured data set, and extract data such as the surgical duration, wound area, and instrument usage category corresponding to the classification coding to obtain an operation feature parameter set including: Read the preset rule mapping table in the International Classification of Diseases coding library, and the rule mapping table contains standard clinical path parameters corresponding to each International Classification of Diseases coding; Specifically, retrieve the mapping relation table of the International Classification of Diseases (ICD) coding version from the standardized coding library built into the medical information system. The ICD coding version can be ICD-11. This table defines the association rules between each disease code and the key surgical parameters. For example, the code K35.80 corresponds to the standard duration range, wound surface grading index, and list of essential instruments for cholecystectomy; the code K35.80 is acute cholecystitis with cholelithiasis, and the standard duration range for cholecystectomy can be 90 - 120 minutes; the wound surface grading index can be that the exposed area of the gallbladder bed is ≥ 3 cm², and the list of instruments can be an electrocautery, laparoscopic grasper, and hemostatic clip. Among them, the International Classification of Diseases coding library is a standardized code system for diseases and surgical operations released by the World Health Organization; the rule mapping table is a structured query table storing the association relationship between disease codes and surgical path parameters; the standard clinical path parameters are the standardized operation indicators to be followed during the surgery for a certain type of disease.

[0034] Match the International Classification of Diseases coding in the structured dataset with the rule mapping table, and select the International Classification of Diseases coding corresponding to the main lesion site. Specifically, for the multiple disease codes included in the structured dataset, screen the core diseases through the lesion priority algorithm. The disease codes are such as the combined code K35.80 and R10.4 abdominal pain. The algorithm sorts according to the keyword frequency of the disease course description, the pathological examination conclusion, and the mass significance in the imaging report. For example, when the CT report shows thickening of the gallbladder wall and impacted stones, the K35.80 code is preferentially matched while ignoring the secondary codes. Among them, the main lesion site is the core disease diagnosis target that the patient currently needs surgical treatment for; the structured dataset is the time-series associated multi-source dataset generated in claim 2; the matching is to screen the most suitable coding mapping relationship through feature similarity calculation.

[0035] Intercept the surgical duration threshold range, wound surface grading standard, and list of instrument types corresponding to the selected International Classification of Diseases coding, and generate an operation feature parameter set.

[0036] Specifically, according to the main code output in sub-step 2, extract the corresponding three-dimensional parameter group (surgical duration threshold range, wound surface grading standard, list of instrument types) from the rule mapping table. The finally generated parameter set will be used to construct the clinical path knowledge base. Among them, the surgical duration threshold range is the reasonable interval of the surgical operation time restricted by medical consensus, the minimum to maximum surgical time defined according to the disease stage; the wound surface grading standard is the medical specification for quantitative assessment of the tissue damage degree, based on the grading table of the International Wound Healing Society. For example, cholecystectomy corresponds to a grade II wound surface; the list of instrument types is the list of tools that must be used or recommended according to the surgical type, the mandatory and optional instrument lists in the standardized classification library.

[0037] Exemplarily, for a patient with cholecystitis, the structured dataset contains the diagnostic codes K35.80, which is acute cholecystitis, R10.4, which is abdominal pain, and E66.9, which is obesity; the main lesion screening algorithm is based on the stone diameter and white blood cell count in the gallbladder ultrasound image. The stone diameter is 12 mm, and the white blood cell count is 15×10 9 / L. It is determined that the main code is K35.80; the parameters corresponding to this code are extracted from the rule mapping table, and the surgical duration threshold range is obtained as 90 - 120 minutes, the wound grading standard is grade III of the gallbladder bed, electrocoagulation hemostasis is required, and the instrument list includes a laparoscope, 5 mm Trocar, and Hem-o-lok clips. Then, an operation feature parameter set is generated and passed into the knowledge base construction module. By accurately matching the main code with the parameter mapping rule, the interference of secondary diseases on the surgical plan is effectively excluded. The standardized parameter group provides a reliable basis for the subsequent calculation of the consumable baseline and eliminates the risk of incorrect instrument configuration caused by misselection of codes.

[0038] Optionally, the establishment of a typical clinical path knowledge base for the corresponding disease types includes: Scanning all postoperative infection and secondary surgery records in the historical complication case database to generate a high-risk case identification set; Specifically, the system traverses the hospital complication database, filters cases with incision infection, organ failure, or secondary surgery within 30 days after surgery. A hash index table is established based on the surgical number field in the case metadata, and through time window filtering (such as only retaining data in the past 5 years), a set containing the unique identifiers of high-risk cases is finally output. Among them, the historical complication case database is a structured database storing postoperative adverse event records; the high-risk case identification set is a set of case numbers with clear complication marks; the secondary surgery record is an archival record that requires re-intervention due to the failure of the first surgery or complications.

[0039] Locate the preoperative instrument calibration data and intraoperative anesthesia records corresponding to the high-risk case identification set, and extract abnormal parameters outside the preset range; Specifically, the preoperative instrument calibration data is a quantitative indicator for the accuracy verification of surgical equipment before use; the intraoperative anesthesia record is the drug dosage and physiological parameters output in real time by the anesthesia machine and monitor; the preset range is the safety interval of equipment or physiological indicators defined by medical norms. According to the unique number of cases in the high-risk identification set, cross-system associate the preoperative equipment calibration log with the vital sign records of the anesthesia system. The preoperative equipment calibration log is such as the calibration value of the laparoscope pressure sensor, and the vital sign records of the anesthesia system are such as the dosage of anesthetic and the fluctuation curve of blood oxygen saturation. Compare the equipment calibration error threshold and the standard range of anesthesia parameters, and mark continuously exceeding data points. The equipment calibration error threshold can be ±5% of the allowable error of the pressure sensor, and the standard range of anesthesia parameters can be a propofol maintenance dose of 4 - 12 mg / kg / h.

[0040] The associated positions of the abnormal parameters in the operation abnormality comparison table are marked to form a knowledge base architecture with risk level labels, thereby obtaining a typical clinical pathway knowledge base.

[0041] Specifically, the operation abnormality comparison table is a classification table for standardizing the types of deviations during the surgical process; the risk level label is a priority identifier set according to the degree of impact of the abnormality; and the knowledge base architecture is a storage model that includes index structures and data relationships. Abnormal parameters are classified into predefined operation abnormality comparison table columns by type, such as "device calibration abnormality" and "anesthesia depth abnormality". Three levels of risk labels are divided according to the frequency of abnormalities and the severity of complications. High risk refers to more than three calibration deviations in the same case and complications require ICU intervention. Medium risk refers to a single anesthetic dose exceeding the standard but not causing serious adverse events. Low risk refers to occasional equipment errors but are discovered and corrected during surgery. Finally, a hierarchical knowledge base with disease code as the primary key is generated, which supports rapid retrieval by risk level.

[0042] For example, the colon resection case database was analyzed, and 5 cases of postoperative abdominal infection were found by scanning. Their numbers were added to the high-risk identification set. It was found that the calibration pressure value of the colonoscope instrument before surgery in 3 cases was 7.5kPa, and the standard range was 6.5-7.0kPa. In the anesthesia records, the blood oxygen saturation of 2 cases was continuously lower than 90%. The pressure exceeding the standard was marked as "device calibration abnormality-high risk", and the blood oxygen abnormality was marked as "anesthesia management-medium risk", and stored in association with the ICD code K55.9 (colon disease). By quantitatively marking the correlation between preoperative equipment abnormalities and intraoperative anesthesia omissions, it helps physicians to check the equipment status and adjust the anesthesia plan before subsequent operations. The risk level classification enables the knowledge base to prioritize the early warning information of high-probability hidden dangers and reduce the occurrence of avoidable complications.

[0043] Optionally, the deviation value sequence of the actual usage of statistical consumables and the standard consumables baseline value includes: Retrieve the consumables collection records of all the same type of surgeries by the target doctor within the preset first time window in the past, and analyze the actual usage quantity; Specifically, the preset first time window is the length of the review period set during statistical analysis; the consumables collection record is the electronic ledger data of the collection and return of medical devices before and after the operation; the actual number of uses is the number of non-reusable devices actually consumed during the operation. The in-and-out records in the consumables management system are associated with the operation number, and similar operations performed by the target doctor in a certain period of time in the past are screened out. The certain period of time can be the last 6 months, and the same operation can be a cholecystectomy. Regular expressions are used to match barcode or RFID tag data to identify consumable items that have been claimed but not returned by the operating room, and the actual consumption quantity of each operation is counted.

[0044] Calibrate the distribution loss and emergency replenishment quantity in the actual usage quantity to obtain a net consumable consumption value sequence; Specifically, the distribution loss quantity is the quantity of consumables that cannot be used due to accidents during transportation; the emergency replenishment quantity is the quantity of consumables temporarily applied for additional during the operation; the net consumable consumption value sequence is a set of real consumption data after eliminating external interferences. Extract the distribution loss records of consumables delivered to the operating room from the logistics system (such as the quantity of ineffective items caused by damaged outer packaging), and retrieve the additional replenishment quantity of consumables temporarily added during the operation from the emergency material application form. Construct a net consumable calculation formula as follows: In the formula, is the net consumable consumption value, is the original issued quantity, is the distribution loss quantity, is the emergency replenishment quantity; calculate for each operation and generate a time-series net consumable consumption value sequence.

[0045] Perform a point-by-point difference calculation between the net consumable consumption value sequence and the standard consumable baseline value to generate a deviation value sequence.

[0046] Specifically, the standard consumable baseline value is the reasonable consumable usage for a single operation preset in the typical clinical pathway knowledge base; the point-by-point difference calculation is a one-to-one subtraction operation between the actual value and the reference value in the same time dimension; the deviation value sequence is a set of quantitative indicators reflecting the degree of deviation of actual usage from standardization. Align the net consumable consumption value with the standard baseline value of this disease type in the knowledge base according to the operation time sequence, and calculate the deviation value for each data point: In the formula, is the deviation value of the th operation, is the net consumable consumption value of the th operation, is the standard baseline value of the disease type corresponding to the th operation; all deviation values are sorted by time to form a deviation value sequence.

[0047] Exemplarily, count 10 cases of laparoscopic cholecystectomy performed by a certain doctor in the recent 3 months, retrieve his consumable issue records, and the original data example shows that for the 7th operation, 12 hemostatic clips were issued, 3 were additionally applied during the operation, and 1 was lost during distribution; calibrate and calculate the net consumable consumption value pieces. The standard baseline is 10 hemostatic clips required for each cholecystectomy, then the deviation value is 4 pieces. After calculating for 10 operations, the sequence [+4, +1, -2, +3,...] is obtained, which reflects the usage fluctuation of hemostatic clips by this doctor. By eliminating the interference factors of logistics loss and emergency addition, the consumable management level of the doctor's actual operation is accurately quantified. The deviation sequence provides an objective data basis for subsequent skill evaluation and avoids misjudging the operation standardization due to accidental factors.

[0048] Optionally, generating a three-dimensional skill evaluation matrix by combining the deviation value sequence, the incidence rate of postoperative complications, and the intraoperative operation fluency includes: Decompose the frame sequence data of the surgical video, identify the number of times that the interval duration of the main surgeon's gesture switching exceeds the preset silence threshold, and calculate the operation fluency index; Specifically, the frame sequence data is a set of continuous static images after the surgical video is decomposed; the gesture switching interval duration is the time difference required for the conversion of different operation actions; the silence threshold is the time critical value for determining the operation interruption; the operation fluency index is a numerical value quantifying the continuity of the surgeon's operation. The surgical video is disassembled into an image sequence at a rate of 30 frames per second, and the YOLO algorithm is used to detect the change of the main surgeon's hand posture. The stagnation interval between instrument switches is recorded through timestamps, such as the conversion time from a needle holder to an electrocautery, and the number of silent pauses exceeding 2 seconds during a single surgery is counted. For calculating the fluency index , there is: In the formula, is the number of times exceeding the super-silence threshold, is the total number of instrument switches, and the fluency index The closer it is to 1, the higher the operation fluency.

[0049] Summarize the number of monitor alarms and the drug addition records within the preset second time window after the operation for the corresponding cases, and quantitatively obtain the incidence rate of complications index; Specifically, the preset second time window is the time range for postoperative monitoring (such as 48 hours); the number of monitor alarms is the trigger times when the device detects abnormal physiological parameters; the drug addition record is the record of emergency remedial measures for postoperative drug use exceeding the original plan; the incidence rate of complications index is the quantitative value of the patient's postoperative recovery risk. Extract monitor alarm events from the ICU monitoring data within 48 hours after the patient's operation, such as heart rate < 50 beats per minute, systolic blood pressure < 90 mmHg; and additional drug use records, such as dopamine addition. For the incidence rate of complications index , there is: In the formula, is the number of alarms, is the number of drug additions, is the monitoring duration, with the unit of hour.

[0050] Construct a three-dimensional coordinate system with the deviation values in the deviation value sequence, the operation fluency index, and the incidence rate of complications index as axes; Specifically, such as Figure 2The three-dimensional skill evaluation graph of the surgeon generated from the 150 times of data statistics processes the multiple surgical data of the target doctor, calculates the deviation value, statistically analyzes the fluency index of each surgery, calculates the complication incidence index, uses the deviation value as the X-axis, the fluency index as the Y-axis, and the complication incidence index as the Z-axis to establish a right-handed coordinate system. The deviation value reflects the stability of consumable use, the fluency index reflects the continuity of operation, and the complication incidence index reflects the postoperative risk.

[0051] Generate a three-dimensional skill evaluation matrix of the target doctor using the three-dimensional coordinate system.

[0052] Specifically, the three-dimensional skill evaluation matrix is a distribution statistical model of the doctor's operation mode in a multi-dimensional space. Normalize the deviation value, fluency index, and complication incidence index using Min-Max normalization to eliminate the dimension difference. For the dynamic skill evaluation value , there is: In the formula, is the weight coefficient of the deviation value, is the weight coefficient of the fluency index, is the weight coefficient of the complication incidence index, is the normalized deviation value, is the normalized fluency index, is the normalized complication incidence index.

[0053] Exemplarily, analyze 20 cholecystectomy operations of a hepatobiliary surgeon. Calculate that the operation fluency index is 0.8; the average alarm is 2 times and additional drugs are used 1 time within 48 hours after the operation, and the monitoring duration is 24 hours, then the complication incidence is 0.125; the standard deviation of the hemostatic clip usage is calculated to be 2.3; map the doctor's data to 3D coordinate points (X = 2.3, Y = 0.8, Z = 0.125) in the coordinate system. After clustering with historical data, it is found that the skill points are located in the "high stability - medium fluency - low complication" area. Intuitively display the strengths and weaknesses of the doctor's skills through three-dimensional indicators. The surgical team can customize training plans for high consumable deviation or low fluency nodes to shorten the skill improvement cycle. The three-dimensional matrix also provides a unified evaluation framework for comparing the capabilities of multiple doctors and optimizes the surgical scheduling decision.

[0054] Optionally, the allowable fluctuation interval threshold for correcting the standard consumable baseline value includes: Set the mapping relationship between the dynamic skill score value and the reduction amount of the corresponding consumable usage upper limit to obtain the threshold shrinkage coefficient; Specifically, the threshold shrinkage coefficient is a proportional parameter of the adjustment range allowed for the consumables; the mapping relationship is the functional correspondence rule between the score and the adjustment amount. A linear regression model between the dynamic skill score value and the upper limit adjustment amount of the consumables is established, and the range of the dynamic skill score value is 0 - 100. The calculation formula for the threshold shrinkage coefficient is defined as: In the formula, is the qualified threshold, is the highest skill threshold, is the maximum shrinkage ratio. For example, when the dynamic skill score value is 80, the calculated threshold shrinkage coefficient is 0.175.

[0055] When the dynamic skill score value is greater than the preset expert threshold, an alternative parameter group for the standard consumable baseline value is activated, and the alternative parameter group comes from the high - scoring physician practice database; Specifically, the expert threshold is the critical score for determining that the clinical operation reaches the expert level; the alternative parameter group is a set of statistical parameters that optimally reflect the actual consumption of high - level physicians; the high - scoring physician practice database is a standardized warehouse storing the surgical practice data of excellent physicians.

[0056] The preset expert threshold is 90. When the doctor's dynamic skill score value is greater than or equal to 90 for three consecutive times, the median value and standard deviation of the consumable usage records are automatically retrieved from the high - scoring physician database as the alternative parameter group. The alternative parameter group includes the new baseline value and the fluctuation range.

[0057] Using the threshold shrinkage coefficient and the alternative parameter group, the allowed fluctuation interval is corrected to generate a new allowed fluctuation interval with a version identifier.

[0058] Specifically, the version identifier is a unique code for distinguishing data of different revision batches; the allowed fluctuation interval is the upper and lower limit values of the compliance range of the consumable usage. The original allowed fluctuation interval is The calculation formula for the corrected interval is: In the formula, is the original standard consumable baseline value, is the threshold of the original allowed fluctuation interval, is the standard consumable baseline value of the alternative parameter group, is the threshold of the allowed fluctuation interval of the alternative parameter group, is the preset expert threshold. The new interval is appended with a version number identifier, such as V2.1.3, indicating the update date and the applicable physician category.

[0059] Exemplarily, for a spinal fusion consumable management scenario, a certain doctor's dynamic skill score value is 85, and the expert threshold is 90, which fails to meet the standard. The original consumable baseline is 6 titanium nails ± 1, and the calculated shrinkage coefficient is 0.193. Therefore, the new allowable fluctuation range is [5.193, 6.807] nails. Another doctor's dynamic skill score value is 92, activating the alternative parameter group. The median number of titanium nails used by the high-scoring doctor is 5, the threshold of the allowable fluctuation range is 0.5, the new range is [4.5, 5.5] nails, and the version identifier is updated to V2.1.3_20240515. The differentiated interval adjustment strategy can not only restrict ordinary doctors from excessive consumption but also release the innovation space for highly skilled physicians. The version identifier mechanism ensures the traceability of revision records and avoids compliance risks caused by clinical misuse of old version standards.

[0060] Optionally, the forming of the personalized consumption guidance plan includes: Embedding the corrected allowable fluctuation range into a preset surgical scheduling system to generate a visual usage progress bar; Specifically, the allowable fluctuation range is the upper and lower limits of the consumable usage compliance range adjusted after doctor skill evaluation and historical data analysis, which is used to restrict the use of consumables during the operation. The surgical scheduling system is a management platform for medical institutions to plan the allocation of operating room resources and personnel deployment, including scheduling elements such as time, personnel, and equipment. The usage progress bar is a visual control that dynamically displays the consumable usage process in a graphical way. The allowable fluctuation range of consumable usage adjusted through dynamic skill scoring is written into the hospital surgical scheduling system database through an application programming interface. In the system operation interface, a dynamic progress bar component based on web technology is generated for each scheduled operation, and this component real-time displays the percentage of the usage amount of each consumable type in the current case in the corrected range. When the actual consumption during the operation approaches the interval boundary, the color of the progress bar gradually changes from green to red.

[0061] Collect intraoperative sensing data and associate it with the typical clinical pathway knowledge base, adjust the allowable fluctuation range when abnormalities occur, and update the usage progress bar; Specifically, the intraoperative sensing data is real-time monitoring data generated by medical devices during the operation, including physiological signals and the number of times of instrument use, etc. The typical clinical pathway knowledge base is a database storing standardized surgical procedures and consumable reference values, including complication association rules. An anomaly is a deviation beyond the threshold between the actual parameter and the preset standard. During the operation, streaming data from a vital sign monitor and an intelligent instrument counting module are collected in real time. The intraoperative blood loss and the number of instrument activations are compared and analyzed with the thresholds stored in the clinical pathway knowledge base. When the blood loss exceeds 20% of the standard value for 5 consecutive minutes or the usage rate of hemostatic clips exceeds the benchmark by 50%, the interval adjustment algorithm is automatically triggered. The adjusted new interval is immediately pushed to the scheduling system interface via the WebSocket protocol, triggering the re-rendering and numerical update of the progress bar component.

[0062] Push guidance suggestions including the reasons for adjustment.

[0063] Specifically, the guidance suggestions are operational countermeasures for intraoperative anomalies, including the basis for specific parameter adjustment and recommended operation steps. After the allowable fluctuation interval is adjusted, the system automatically generates a natural language description text containing the reasons for adjustment and recommended operations. A structured notification message is sent to the mobile terminal of the surgeon through the hospital's internal messaging platform, such as "Due to the intraoperative blood loss reaching 150 ml / 10 min (benchmark 120 ml), the upper limit of the hemostatic material has been increased to 5 units. It is recommended to give priority to using bipolar electrocoagulation for bleeding points."

[0064] Exemplarily, during a laparoscopic cholecystectomy in the hepatobiliary surgery department of a certain hospital, it was detected that the intraoperative blood loss of the patient reached 80 ml / 5 min during the laparotomy stage, exceeding the threshold of 50 ml / 5 min corresponding to the allowable interval of the current version. The postoperative complication probability model of similar cases in the clinical pathway knowledge base was automatically retrieved, and it was determined that the upper limit of the allowable amount of hemostatic material needed to be increased. The allowable interval of the original 3 absorbable hemostatic gauzes was adjusted to 5 pieces, and at the same time, guidance information "The amount of bleeding from the wound surface is abnormal. It is recommended to use hemostatic gauze for positioning and compression and apply for blood preparation" was pushed to the doctor. This adjustment enables the doctor to obtain additional hemostatic consumables in compliance without interrupting the surgical process, and the system can monitor in real time whether the actual usage exceeds the new threshold through the updated progress bar. This method can flexibly adjust the consumable limit on the premise of ensuring surgical safety, avoiding the risk of insufficient resource supply due to unexpected situations, and strengthening the standardization of consumption through a dynamic feedback mechanism, forming an operation closed-loop.

[0065] Optionally, the method further includes: Monitor the consumable usage data stream in the consecutive surgical records of the target doctor, compare the actual usage with the corrected allowable fluctuation interval case by case, and calculate the consumption deviation degree; Specifically, the allowable fluctuation range is the compliance range of consumable use dynamically adjusted according to the doctor's historical skill score. The consumption deviation degree is a quantitative index for the deviation degree between the actual consumption and the standardized benchmark. The consumable use data stream is the detailed information of consumable consumption continuously recorded according to the operation timestamp. All the same-type surgical cases completed by the target doctor within a specific time period are collected in real time through the hospital material management system, and the actual consumption quantity of key consumables in each operation is extracted. The actual usage is compared case by case with the allowable fluctuation range in effect at that time, and the range includes the upper and lower limits of the consumable usage dynamically adjusted for this doctor. Calculate the degree of deviation of the usage of each consumable in each operation from the benchmark value to form a group of deviation degree values arranged in chronological order. When the actual consumable usage exceeds the allowable fluctuation range, the deviation degree is calculated using the percentage difference method: In the formula, is the consumption deviation degree, is the actual consumable usage, is the median value of the corrected allowable interval.

[0066] Verify whether the consumption deviation degree is less than the preset deviation threshold and the number of postoperative complication markers is zero; Specifically, the deviation threshold is the maximum percentage range of allowable consumption deviation. The number of postoperative complication markers is the number of standardized diagnostic codes of postoperative adverse events in the database. The preset standard is that the consumable deviation does not exceed the standard and there are no postoperative complications. Set the deviation threshold for each consumable. For example, the deviation threshold for hemostatic materials is ±15%. Check in turn whether the consumable deviation degrees of each case in the continuous surgical records are all lower than the threshold, and at the same time retrieve the electronic medical record data within 30 days after the operation of the patient to confirm that there are no complication markers such as incision infection or secondary surgery in all cases. If it is found that the deviation degree of any case exceeds the standard or there is a complication marker, it is determined that this verification fails.

[0067] When the number of consecutive times of passing the verification results is greater than the preset number threshold, update the skill label of the target doctor in the typical clinical pathway knowledge base and synchronize the new version threshold to all associated systems.

[0068] Specifically, the skill label is a classification identifier reflecting the doctor's consumable management ability. The new version threshold is the consumable use limit standard adjusted to apply to this doctor. The typical clinical pathway knowledge base is a database storing standardized surgical procedures and consumption parameters. Set the number threshold of consecutive times of passing the verification to 5 times. When the consumption deviation degrees of 5 consecutive operations of the target doctor meet the requirements and there is no complication record, the system automatically modifies the skill label level in the knowledge base, and synchronizes the latest consumable fluctuation range of this doctor as the new standard to associated platforms such as the operation scheduling system and the consumable approval system. The label update adopts a binary version number increment mechanism to ensure the traceability of historical operations.

[0069] For example, a gastrointestinal surgeon at a tertiary hospital completed 6 consecutive laparoscopic radical resections for colorectal cancer. The monitoring system recorded the use of 12, 13, 11, 10, 14, and 12 staplers in each operation. The corrected allowable fluctuation range is 10-14, and the corresponding median value is 12. The deviations of each case were calculated as 0%, +8.3%, -8.3%, -16.7%, +16.7%, and 0%, respectively, of which the 4th and 5th exceeded the ±15% threshold. Only when the verification was triggered for the 6th operation, it was found that the number of consecutive compliances was 1, and only the 6th was satisfied, so it was determined that the skill label would not be updated. The reliability of skill label updates is ensured by strictly verifying the continuous compliance of consumables use. The blocking mechanism when the standard is not met avoids the risk of misjudgment, and the coordinated update after continuous compliance strengthens the consistency of consumables management throughout the hospital.

[0070] Based on the same inventive concept, Figure 3 As shown, the present invention also provides a standardized consumption path management system based on disease types, the system comprising: The data collection module is used to obtain the electronic medical record text data, surgical video data, consumables usage records and postoperative monitoring data of the target surgical case, and standardize the data to generate a structured data set; A feature extraction module is used to identify the International Classification of Diseases code from the structured data set, extract the data of the operation duration, wound area, and instrument use category corresponding to the classification code, and obtain an operation feature parameter set; A knowledge base construction module is used to establish a typical clinical pathway knowledge base matching the corresponding disease based on the operation characteristic parameter set; a consumption analysis module is used to analyze the historical surgical video data of the target doctor and to count the deviation value sequence between the actual consumption of consumables and the baseline value of the standard consumables; A skill evaluation module, for generating a three-dimensional skill evaluation matrix based on the deviation value sequence, the incidence of postoperative complications, and the fluency of intraoperative operation, and calculating the dynamic skill score of the target doctor according to the three-dimensional skill evaluation matrix; A dynamic threshold calculation module is used to correct the allowable fluctuation range of the baseline value of the standard consumables based on the dynamic skill score value; a consumption plan generation module is used to dynamically associate the corrected allowable fluctuation range with the typical clinical pathway knowledge base to form a personalized consumption guidance plan.

[0071] It should be noted that the electrical connection between the above-mentioned units does not necessarily mean direct connection of the lines, and the indirect connection mode can be applied to the embodiments of the present invention as long as the purpose of the present invention is achieved. The above is only an exemplary embodiment of the present invention and cannot be used to limit the scope of the present invention.

[0072] That is, any equivalent changes and modifications made in accordance with the teachings of the present invention still fall within the scope covered by the present invention. Those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and the practice of the disclosure. This application is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include the common general knowledge or conventional technical means in the technical field not recorded in the present invention.

Claims

1. A standardized consumption path management method based on disease types, characterized in that: The method comprises: Obtain the electronic medical record text data, surgical video data, consumables usage records and postoperative monitoring data of the target surgical cases, and standardize the data to generate a structured data set; Identifying the International Classification of Diseases code from the structured data set, extracting the data of the operation duration, wound area, and instrument use category corresponding to the classification code, and obtaining an operation characteristic parameter set; Based on the operation characteristic parameter set, a typical clinical pathway knowledge base matching the corresponding disease is established, wherein the typical clinical pathway knowledge base includes a comparison table of standard consumables baseline values ​​and operation abnormalities; Analyze the historical surgical video data of the target doctor and count the deviation value sequence between the actual usage of consumables and the baseline value of the standard consumables; Combining the deviation value sequence, the incidence of postoperative complications and the fluency of intraoperative operation, a three-dimensional skill evaluation matrix is ​​generated, and the dynamic skill score of the target doctor is calculated according to the three-dimensional skill evaluation matrix; Based on the dynamic skill score value, modify the allowable fluctuation range of the standard consumables baseline value; The revised allowable fluctuation range is dynamically associated with the typical clinical pathway knowledge base to form a personalized consumption guidance plan.

2. The standardized consumption path management method based on disease type according to claim 1 is characterized in that: The step of obtaining electronic medical record text data, surgical video data, consumables usage records, and postoperative monitoring data of the target surgical case, and performing standardization processing on the data to generate a structured data set includes: The original medical record text data of the target surgical case was retrieved from the hospital's electronic medical record system, and a set of keywords including surgical stage marks, lesion site descriptions, and medication records were extracted; Acquire surgical video data and perform frame sequence analysis to generate a timestamp association table that matches the surgical stage markers; Synchronously retrieve the consumables collection details record marked with the surgical number in the consumables management system, and map the consumables name in the consumables collection details record to a preset standardized consumables coding library to generate a unified consumables coding sequence; Based on the time range of the timestamp association table, locate the vital sign sequence and drug infusion records associated with the target surgery in the postoperative monitoring data, and perform interpolation filling on the missing time period data; Associating the keyword set, consumables coding sequence, vital signs sequence and drug infusion record by fields, converting them into a JSON format data stream with a unified time sequence, and generating the structured data set; Each record of the structured data set includes synchronous timing fields of surgical stage mark, consumables code, vital sign parameters and medication record.

3. The standardized consumption path management method based on disease type according to claim 1 is characterized in that: The step of identifying the International Classification of Diseases code from the structured data set, extracting the data of the operation duration, wound area, and instrument use category corresponding to the classification code, and obtaining the operation characteristic parameter set includes: Reading a preset rule mapping table in the International Classification of Diseases code library, wherein the rule mapping table contains standard clinical pathway parameters corresponding to each International Classification of Diseases code; Matching the International Classification of Diseases code in the structured data set with the rule mapping table, and selecting the International Classification of Diseases code corresponding to the main lesion site; The operation duration threshold range, wound grading standard and instrument type list corresponding to the selected International Classification of Disease codes were intercepted to generate an operation characteristic parameter set.

4. The standardized consumption path management method based on disease type according to claim 1 is characterized in that: The typical clinical pathway knowledge base for matching corresponding diseases is established as follows: Scan all postoperative infection and secondary surgery records in the historical complication case database to generate a high-risk case identification set; Locating the preoperative instrument calibration data and intraoperative anesthesia records corresponding to the high-risk case identification set, and extracting abnormal parameters that exceed a preset range; The associated positions of the abnormal parameters in the operation abnormality comparison table are marked to form a knowledge base architecture with risk level labels, thereby obtaining a typical clinical pathway knowledge base.

5. The standardized consumption path management method based on disease type according to claim 1 is characterized in that: The deviation value sequence of the actual usage of statistical consumables and the baseline value of the standard consumables includes: Retrieve the consumables collection records of all the same type of surgeries by the target doctor within the preset first time window in the past, and analyze the actual usage quantity; Calibrate the distribution loss amount and emergency replenishment amount in the actual usage quantity to obtain a net consumables consumption value sequence; Perform point-by-point difference calculation between the net consumables consumption value sequence and the standard consumables baseline value to generate a deviation value sequence.

6. The standardized consumption path management method based on disease type according to claim 1 is characterized in that: The three-dimensional skill evaluation matrix generated by combining the deviation value sequence, the incidence of postoperative complications and the fluency of intraoperative operation includes: Decompose the frame sequence data of the surgical video, identify the number of times the interval between the surgeon's gesture switching exceeds the preset silence threshold, and calculate the operation fluency index; Summarize the number of monitor alarms and drug supplementation records of the corresponding cases within the preset second time window after surgery to quantify the complication incidence index; constructing a three-dimensional coordinate system with the deviation value in the deviation value sequence, the operation fluency index, and the complication incidence index as axes; The three-dimensional coordinate system is used to generate a three-dimensional skill evaluation matrix of the target doctor.

7. The standardized consumption path management method based on disease type according to claim 1 is characterized in that: The allowable fluctuation range threshold for correcting the baseline value of the standard consumables includes: A mapping relationship between the dynamic skill score value and the corresponding upper limit reduction amount of consumables usage is set to obtain a threshold reduction coefficient; When the dynamic skill score value is greater than a preset expert threshold, activating a replacement parameter set for a standard consumable baseline value, the replacement parameter set is from a high-scoring physician practice database; The threshold shrinkage coefficient and the replacement parameter group are used to modify the allowable fluctuation range and generate a new allowable fluctuation range with a version identifier.

8. The standardized consumption path management method based on disease type according to claim 7 is characterized in that: The forming of a personalized consumption guidance plan includes: Embed the corrected allowable fluctuation range into the preset surgical scheduling system to generate a visual usage progress bar; Collecting intraoperative sensor data and associating it with the typical clinical pathway knowledge base, adjusting the allowable fluctuation range when an abnormality occurs, and updating the dosage progress bar; Push guidance suggestions including reasons for adjustments.

9. The standardized consumption path management method based on disease type according to claim 8 is characterized in that: The method further comprises: Monitor the consumables usage data stream in the target doctor's continuous surgical records, compare the actual usage with the corrected allowable fluctuation range case by case, and calculate the consumption deviation; Verify whether the consumption deviations are all less than a preset deviation threshold and the postoperative complication marker number is zero; When the verification result meets the standard for more than a preset number of times in a row, the skill label of the target doctor in the typical clinical pathway knowledge base is updated, and the new version of the threshold is synchronized to all related systems.

10. A standardized consumption path management system based on disease types, applied to a standardized consumption path management method based on disease types as claimed in any one of claims 1 to 9, characterized in that: The system comprises: The data collection module is used to obtain the electronic medical record text data, surgical video data, consumables usage records and postoperative monitoring data of the target surgical case, and standardize the data to generate a structured data set; A feature extraction module is used to identify the International Classification of Diseases code from the structured data set, extract the data of the operation duration, wound area, and instrument use category corresponding to the classification code, and obtain an operation feature parameter set; A knowledge base construction module is used to establish a typical clinical pathway knowledge base matching the corresponding disease based on the operation characteristic parameter set; The consumption analysis module is used to analyze the historical surgical video data of the target doctor and to count the deviation value sequence between the actual consumption of consumables and the baseline value of the standard consumables; A skill evaluation module, for generating a three-dimensional skill evaluation matrix based on the deviation value sequence, the incidence of postoperative complications, and the fluency of intraoperative operation, and calculating the dynamic skill score of the target doctor according to the three-dimensional skill evaluation matrix; A dynamic threshold calculation module, used to correct the allowable fluctuation range of the standard consumables baseline value based on the dynamic skill score value; The consumption plan generation module is used to dynamically associate the corrected allowable fluctuation range with the typical clinical pathway knowledge base to form a personalized consumption guidance plan.

Images