CN-121999998-A - Data processing method and system for monitoring clinical specialized ability

CN121999998ACN 121999998 ACN121999998 ACN 121999998ACN-121999998-A

Abstract

A data processing method and system for monitoring the capability of clinical specialty relate to the field of big data resource service, the method comprises the steps of receiving business source data and analog source data and generating a time sequence index data set; the method comprises the steps of inputting the initial variable into a time sequence prediction model to generate baseline trend data, receiving intervention parameters representing target intervention actions, adjusting initial variables in a causal deduction model according to the intervention parameters, deducting operation based on the causal deduction model to generate simulation result data, generating current capability scores, baseline prediction scores and simulation prediction scores according to current index values, baseline trend data and the simulation result data in a time sequence index data set, and performing data association on the current capability scores, the baseline prediction scores and the simulation prediction scores to generate a result display chart. By implementing the application, the data processing efficiency can be improved.

Inventors

DU ZHIQIANG
MA XIAOPING
Bie Wenjin
HE REN
AN XU
YANG WEIKANG
LI BING

Assignees

江苏守正耘创大数据科技有限公司

Dates

Publication Date: 20260508
Application Date: 20251229

Claims (10)

1. A data processing method for clinical specialty capability monitoring, applied to a data processing system, said method comprising: Receiving service source data and simulation source data, and extracting and integrating the service source data and the simulation source data by combining historical data to generate a time sequence index data set containing capability indexes on a plurality of time nodes, wherein the simulation source data is substitute data generated by simulation according to the historical source data when a corresponding service system does not acquire the service source data within preset response time; inputting the time sequence index data set into a time sequence prediction model to generate baseline trend data representing the condition that external intervention is not carried out in a preset time period in the future; receiving intervention parameters representing a target intervention action, and adjusting initial variables in a causal deduction model according to the intervention parameters; Performing deduction operation based on the causal deduction model, and generating simulation result data representing each capability index in the future preset time period after the target intervention action is implemented; Generating a current capacity score, a baseline prediction score and a simulation prediction score according to the current index value, the baseline trend data and the simulation result data in the time sequence index data set; and carrying out data association on the current capacity score, the baseline prediction score and the simulation prediction score to generate a result display chart.
2. The method of claim 1, wherein prior to the step of receiving traffic source data and analog source data, extracting and integrating the traffic source data and the analog source data in combination with historical data to generate a time series indicator dataset comprising capability indicators on a plurality of time nodes, the method further comprises: When the historical data sample size of the current specialty is lower than a preset sample threshold, determining a reference specialty with similarity of parameters of the business process and the resource configuration of the current specialty higher than a preset similarity threshold; According to the weight coefficient of the parameter similarity, converting the capacity index of the reference specialty into an index estimated value corresponding to the current specialty; according to the index estimation, performing time alignment and numerical fusion on the historical data of the reference specialty and the historical data of the current specialty to generate capacity expansion historical data; and calculating simulation source data based on the capacity expansion historical data when the service source data is not acquired within the preset response time.
3. The method of claim 1, wherein the step of data correlating the current capability score, the baseline predictive score, and the simulated predictive score to generate a result presentation chart, specifically comprises: Calculating a trend deviation coefficient between the current capacity score and the baseline prediction score, and calculating an intervention effect coefficient between the baseline prediction score and the simulation prediction score; Determining the development situation of the current specialty ability based on the trend deviation coefficient, and determining the expected benefit of the target intervention action based on the intervention effect coefficient; And constructing a result display chart comprising a time axis, a scoring curve and a key node mark according to the development situation and the expected benefits, wherein the key node mark is used for identifying the inflection point position and the change amplitude of the scoring change.
4. The method of claim 1, wherein prior to the step of receiving an intervention parameter indicative of a target intervention action, adjusting an initial variable in a causal deduction model according to the intervention parameter, the method further comprises: extracting a plurality of special records corresponding to the same patient identification in the time sequence index data set, and determining the time overlapping times of each special record in the diagnosis and treatment period of the same patient; when the time overlapping times exceeds a preset time threshold, calculating the index value product sum of each special department in the overlapping period to be used as a special department cooperation influence factor; And adjusting initial variable coefficients of corresponding specialized departments in the causal deduction model according to the specialized department collaboration influence factors to generate a causal deduction model containing collaboration relations.
5. The method according to claim 4, wherein when the time overlap number exceeds a preset number threshold, the method further comprises, after the step of calculating the index value product sum of the respective specialized departments during the overlap period as the specialized department cooperation influencing factor: Recording that the corresponding date is a mutation date when the single-day variation in the time sequence index data set exceeds the sum of the historical average variation and the historical standard deviation; taking a preset time period before and after the mutation date as a mutation period, and calculating the average value ratio of index averages of each special department in the mutation period and a stable period, wherein the stable period is a time except the mutation period; when the average value ratio deviates from a preset ratio range, marking the corresponding special department as a mutation response special department; and calculating the time overlapping times of the mutation response specialty and other specialty in the mutation period, and recalculating the specialty cooperation influence factor.
6. The method of claim 1, wherein after the step of generating a current capacity score, a baseline predictive score, and a simulated predictive score from the current index values in the time series index data set, the baseline trend data, and the simulated result data, the method further comprises: calculating the update frequency of each index in the time sequence index data set in a preset time window, and determining an aging sensitive index when the update frequency exceeds a preset frequency threshold; Extracting a plurality of preset nearest numerical values of the aging sensitive index, and calculating the absolute value of the difference value of the preset nearest numerical values of two adjacent times to be used as a fluctuation coefficient; when the fluctuation coefficient exceeds the preset multiple of the index history average value, the weight of the corresponding index is increased to be the preset multiple of the original weight; re-calculating the weighted current capacity score to generate an aging adjustment score; And comparing the grading difference value of the aging adjustment grading and the current capacity grading, and generating prompt information containing an aging sensitive index when the absolute value of the grading difference value exceeds a preset difference value threshold value.
7. The method of claim 6, wherein after the step of comparing the rating difference between the age adjustment rating and the current capacity rating and generating a cue comprising an age sensitivity indicator when an absolute value of the rating difference exceeds a preset difference threshold, the method further comprises: Extracting historical prediction error data of the aging sensitive index before weight adjustment, calculating a standard deviation ratio and generating a stability coefficient of weight adjustment; When the stability coefficient exceeds a preset stability threshold, decreasing and adjusting the increasing weight of the ageing sensitive index according to a preset attenuation proportion to generate an attenuated weight value; reapply the attenuated weight value to the calculation process of the current capacity score to generate a current capacity score after weight optimization; And updating coefficient parameters of corresponding indexes in the causal deduction model according to the current capability scores after weight optimization.
8. A data processing system comprising one or more processors and memory coupled to the one or more processors, the memory for storing computer program code comprising computer instructions that the one or more processors invoke to cause the data processing system to perform the method of any of claims 1-7.
9. A computer readable storage medium comprising instructions which, when run on a data processing system, cause the data processing system to perform the method of any of claims 1-7.
10. A computer program product, characterized in that the computer program product, when run on a data processing system, causes the data processing system to perform the method according to any of claims 1-7.

Description

Data processing method and system for monitoring clinical specialized ability Technical Field The application relates to the field of large data resource service, in particular to a data processing method and system for monitoring clinical specialized ability. Background In modern hospital fine management, the construction of a real-time and dynamic monitoring system for the clinical specialty is a key link for supporting scientific decisions and continuously improving medical quality. The system needs to integrate data from a plurality of heterogeneous business systems such as a Hospital Information System (HIS), an electronic medical record system (EMR), a Laboratory Information System (LIS) and the like, so as to quantitatively evaluate the capacity performance of each department in the dimensions such as resource allocation, medical service, quality safety and the like, and visually present the capacity performance in the form of a data large screen and the like. In the related art, a data warehouse-based timing aggregation and presentation system is constructed. According to the scheme, a data Extraction (ETL) tool is configured, a scheduling task is executed according to a preset time period, and original data of required indexes are pulled from databases of all source service systems. After being cleaned and converted, the data are loaded into an index fact table of a central data warehouse. The visual monitoring platform at the front end directly inquires the data warehouse, renders the latest index data after aggregation into a chart, and generates periodic capability reports of special departments of the whole hospital. However, the index data must wait for all source data constituting the index to be successfully acquired before initiating the calculation. Because the construction years and technical architecture of informatization subsystems relied by departments of a hospital are different, and the generation and archiving periods of business data, such as detection reports, are different, when any source system (such as an LIS system for delaying data uploading due to maintenance) has data supply delay or interruption, the index calculation flow depending on the data is blocked, so that delay exists in data processing, and the subsequent capability report display is affected. Disclosure of Invention The application provides a data processing method and a system for monitoring the capability of clinical specialty, which are used for improving the data processing efficiency. The application provides a data processing method for monitoring clinical specialty capability, which is applied to a data processing system and comprises the steps of receiving service source data and analog source data, combining historical data to extract and integrate the service source data and the analog source data to generate a time sequence index data set containing capability indexes on a plurality of time nodes, wherein the analog source data is substitute data generated by the corresponding service system according to the analog of the historical source data when the service source data is not acquired within preset response time, inputting the time sequence index data set into a time sequence prediction model to generate baseline trend data representing the condition that the service source data is not subjected to external intervention within a preset time period, receiving intervention parameters representing a target intervention action, adjusting initial variables in a causal deduction model according to the intervention parameters, performing deduction operation based on the causal deduction model to generate simulation result data representing each capability index in the preset time period in the future, generating a current capability score, a simulation prediction score and a simulation result score according to the current index value, the baseline trend data and the simulation result data in the time sequence index data set, and performing correlation score display of the current capability score, the baseline prediction score and the simulation result score. In the above embodiment, when the service source data is delayed or missing, the data processing system generates the substitute data by using the history data, so that the continuity of the data stream is ensured, and the overall data processing flow blockage caused by the single data source problem is avoided. Meanwhile, by combining time sequence prediction and causal deduction, the method not only can display the current capability state, but also can predict the natural development trend and simulate the effect of specific intervention measures, thereby improving the real-time performance of the capability monitoring of clinical specialization. In combination with some embodiments of the first aspect, in some embodiments, before the step of receiving the service source data and the analog source data, extracting and integrating the service s