Intelligent hospital information supervision method and system based on clustering processing
By adopting clustering-based processing methods in hospital information supervision, dynamically adjusting the data acquisition frequency and performing information balance optimization, combining local density estimation and adaptive optimization, the problems of data redundancy or insufficient and lack of real-time abnormal detection in the prior art are solved, and the accuracy and efficiency of hospital information supervision are improved.
Patent Information
- Application Number
- CN202510286276.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-12
- Publication Date
- 2025-04-11
AI Technical Summary
The existing hospital information supervision methods have redundant or insufficient data, which cannot monitor the hospital's operating status in all aspects, and cannot adapt to data changes in the hospital's dynamic operations, resulting in poor accuracy of information supervision and lack of real-time dynamic abnormality detection and processing mechanisms, which affects the hospital's operating efficiency.
The intelligent hospital information supervision method based on clustering processing is adopted, and data collection is collected based on department priorities and traffic changes, and information balance optimization and local information distribution adjustment mechanism are introduced to ensure the comprehensive capture of multi-dimensional data; at the same time, local density estimation, density measurement and error adjustment parameters are introduced to calculate abnormal scores, and adaptive optimization is carried out to adapt to data characteristics changes.
It improves the accuracy and efficiency of hospital information supervision, ensures real-time monitoring of high-load and high-priority departments, enhances resource allocation efficiency and patient management capabilities, and improves the accuracy of abnormal detection and hospital information supervision effects.
Smart Images
Figure CN119811612A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and specifically to an intelligent hospital information supervision method and system based on clustering processing. Background Art
[0002] Hospital information supervision methods refer to strategies and means that use data analysis, algorithm models and intelligent technologies to comprehensively monitor and manage hospital operations, resources, services, etc. This method collects, processes and analyzes various hospital data, monitors the hospital's operating status in real time, and promptly discovers potential problems, providing decision support and early warning functions for hospital management, thereby improving hospital operating efficiency, optimizing resource allocation, improving service quality and ensuring patient safety. However, general hospital information supervision methods have redundant or insufficient data in some departments, and cannot fully monitor the hospital's operating conditions, and cannot adapt to data changes in the hospital's dynamic operation, resulting in poor accuracy of hospital information supervision; general hospital information supervision methods lack real-time dynamic anomaly detection and processing mechanisms. In the face of rapid changes in patient flow and department load, feature selection and weight allocation cannot respond in time, resulting in potential problems that cannot be quickly discovered, affecting hospital operating efficiency, and thus leading to poor hospital information supervision results. Summary of the invention
[0003] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an intelligent hospital information supervision method and system based on clustering processing. In view of the fact that general hospital information supervision methods have redundant or insufficient data in some departments, cannot fully monitor the operation status of the hospital, and cannot adapt to data changes in the dynamic operation of the hospital, resulting in poor accuracy of hospital information supervision, this solution dynamically adjusts the data collection frequency based on department priority and traffic changes, avoids redundant data collection, and ensures real-time monitoring of high-load and high-priority departments, which helps to improve the overall resource allocation efficiency and patient management capabilities of the hospital; introduces information balance optimization to ensure the comprehensive capture of multi-dimensional data of the hospital; based on the local information distribution adjustment mechanism, the data of different departments can be optimized. The process is balanced and biased, thereby improving the accuracy of hospital information supervision. In view of the lack of real-time dynamic anomaly detection and processing mechanism in general hospital information supervision methods, feature selection and weight allocation cannot respond in time when facing rapid changes in patient flow and department load, resulting in potential problems that cannot be discovered quickly, affecting hospital operation efficiency, and thus leading to poor hospital information supervision effects. This scheme introduces local density estimation, density measurement and error adjustment parameters to calculate the anomaly score of each sample, so that the anomaly score can more comprehensively reflect the degree of anomaly of the data and improve the accuracy of anomaly detection. Adaptive optimization is performed based on feature adjustment factors to ensure that it adapts to the continuously changing data characteristics in hospital operations, thereby improving the effect of hospital information supervision.
[0004] The technical solution adopted by the present invention is as follows: The intelligent hospital information supervision method based on clustering processing provided by the present invention comprises the following steps:
[0005] Step S1: data acquisition;
[0006] Step S2: dynamic feature projection;
[0007] Step S3: hospital information supervision;
[0008] Step S4: Adaptive optimization.
[0009] Furthermore, in step S1, the data acquisition is to acquire hospital information data; combining the characteristics of hospital departments and patient flow, key department operation data is collected first, and department priority It is expressed as: ; Dynamically adjust data collection frequency , expressed as: ;in, is the weighting factor of department load; It is the department load; is the weighting factor of patient flow; is the rate of change of patient flow; is the basic sampling frequency; It is the department flow; the hospital information data includes department load data, patient flow data, department resource usage data, resource allocation data and sampling time.
[0010] Furthermore, in step S2, the dynamic feature projection is to select features; filter irrelevant features; the projection matrix of the data is expressed as: ; Introduce the local information distribution adjustment mechanism, the objective function It is expressed as: ; D is the projection matrix; n is the number of hospital data samples, i is the sample index; is the i-th data point; is the projected data; is the expected value; is the L2 norm; P is the total number of features, l is the feature index; is the feature weight; is the local control weight; M is the number of departments.
[0011] Furthermore, in step S3, the hospital information supervision is dynamically grouped based on the clustering algorithm, and abnormal status is detected by calculating the abnormal score; the hospital data will be projected into a new space, and the location y of each cluster center is obtained by K-means clustering; the cluster center is expressed as: ;in, and are the positions of the k-th cluster center at the t-th and t+1-th iterations respectively; is the jth hospital data sample; h is the adjustment parameter that controls the scale of the distance measurement; is a Gaussian kernel function; after clustering, the abnormality score of the hospital data is calculated , expressed as: ; ; ;in, is the average position of the cluster to which the i-th data belongs; is the standard deviation; is the local density estimate; is the maximum local density estimate; is the abnormal adjustment parameter; is a density measure; K is the total number of clusters; is the position of the kth cluster center at the tth iteration; is the error adjustment parameter; is the average anomaly score of the previous N data points; is the scaling factor; a scoring threshold is set; if the abnormal score of hospital data is greater than the scoring threshold, it is marked as an abnormal state; and an early warning is issued to the management personnel.
[0012] Furthermore, in step S4, the adaptive optimization is to automatically adjust the feature weights; when new sampling data is continuously introduced, the projection matrix D is updated in an iterative manner to adapt to the new data distribution, and the feature adjustment factor is introduced at the same time. , which is used to adjust the intensity of department clustering adjustment; the feature weight update is expressed as: ; ;in, is the rate of change of the feature at the tth iteration; is the feature weight at the tth iteration; is the regularization parameter; is the feature adjustment parameter; It is The moderating factor of each feature; and is the characteristic value of the i-th sample data; and It is the eigenvalue of the cluster center to which the i-th sample data belongs.
[0013] The intelligent hospital information supervision system based on clustering processing provided by the present invention includes a data acquisition module, a dynamic feature projection module, a hospital information supervision module and an adaptive optimization module;
[0014] The data acquisition module acquires hospital information data;
[0015] The dynamic feature projection module performs dimensionality reduction processing through feature projection;
[0016] The hospital information supervision module dynamically groups hospital data using a clustering algorithm and calculates anomaly scores to detect abnormal conditions;
[0017] The adaptive optimization module dynamically adjusts the feature weights and utilizes feature adjustment factors to adjust the adjustment strength of department clustering.
[0018] The beneficial effects achieved by the present invention using the above scheme are as follows:
[0019] (1) In view of the fact that general hospital information supervision methods have redundant or insufficient data in some departments, cannot fully monitor the hospital's operating conditions, and cannot adapt to data changes in the hospital's dynamic operations, resulting in poor accuracy of hospital information supervision, this solution dynamically adjusts the data collection frequency based on department priority and traffic changes to avoid redundant data collection while ensuring real-time monitoring of high-load and high-priority departments, which helps to improve the hospital's overall resource allocation efficiency and patient management capabilities; introduces information balance optimization to ensure the comprehensive capture of the hospital's multi-dimensional data; based on the local information distribution adjustment mechanism, the data of different departments can be balanced in the optimization process to avoid bias; thereby improving the accuracy of hospital information supervision.
[0020] (2) In view of the lack of real-time dynamic anomaly detection and processing mechanism in general hospital information supervision methods, feature selection and weight allocation cannot respond in time when facing rapid changes in patient flow and department load, resulting in the inability to quickly discover potential problems, affecting hospital operation efficiency, and thus leading to poor hospital information supervision effects. This scheme introduces local density estimation, density measurement and error adjustment parameters to calculate the anomaly score of each sample, so that the anomaly score can more comprehensively reflect the degree of abnormality of the data and improve the accuracy of anomaly detection; adaptive optimization is performed based on feature adjustment factors to ensure that it adapts to the continuously changing data characteristics in hospital operations; thereby improving the effectiveness of hospital information supervision. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 A schematic diagram of the process of the intelligent hospital information supervision method based on clustering processing provided by the present invention;
[0022] Figure 2 A schematic diagram of the intelligent hospital information supervision system based on clustering processing provided by the present invention.
[0023] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION
[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.
[0025] In the description of the present invention, it is necessary to understand that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the referred system or element must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.
[0026] Example 1, see Figure 1 The present invention provides a smart hospital information supervision method based on clustering processing, which includes the following steps:
[0027] Step S1: data acquisition: acquiring hospital information data;
[0028] Step S2: dynamic feature projection; dimensionality reduction processing is performed through feature projection;
[0029] Step S3: Hospital information supervision: dynamically group hospital data through clustering algorithm and calculate anomaly scores to detect abnormal status;
[0030] Step S4: Adaptive optimization; dynamic feature weights, using feature adjustment factors to adjust the adjustment strength of department clustering.
[0031] Example 2, see Figure 1 This embodiment is based on the above embodiment. In step S1, the key department operation data is collected first in combination with the hospital department characteristics and patient flow. It is expressed as: ; Dynamically adjust data collection frequency , expressed as: ;in, is the weighting factor of department load; It is the department load; is the weighting factor of patient flow; is the rate of change of patient flow; is the basic sampling frequency; It is the flow of people in the department; the hospital information data includes department load data, patient flow data, department resource usage data, resource allocation data and sampling time; the department load includes the number of patients, the workload of medical staff and the bed occupancy rate; the patient flow includes the number of patients received, the flow trend and the severity of the disease; the department resource usage data includes equipment usage and drug consumption.
[0032] Example 3, see Figure 1 This embodiment is based on the above embodiment. In step S2, dynamic feature projection is to select features that contribute to supervision; filter irrelevant features of the patient's dietary preferences; the projection matrix of the data is expressed as: ; Optimize the selection of feature weights through information balance; ensure the balanced distribution of feature weights, avoid bias in feature selection, ensure multi-dimensional processing of hospital information data, and enhance the comprehensive capture of hospital data; introduce a local information distribution adjustment mechanism to avoid the bias of a few features in global processing, so that the characteristics of different departments can be balanced in the optimization, avoiding over-reliance on high-frequency department data; objective function It is expressed as: ; D is the projection matrix; n is the number of hospital data samples, i is the sample index; is the i-th data point; is the projected data; is the expected value; is the L2 norm; P is the total number of features, l is the feature index; is the feature weight; is the local control weight; M is the number of departments.
[0033] By performing the above operations, the general hospital information supervision method has redundant or insufficient data in some departments, cannot fully monitor the hospital's operating conditions, and cannot adapt to data changes in the hospital's dynamic operations, resulting in poor accuracy of hospital information supervision. This solution dynamically adjusts the data collection frequency based on department priority and traffic changes to avoid redundant data collection while ensuring real-time monitoring of high-load and high-priority departments, which helps to improve the hospital's overall resource allocation efficiency and patient management capabilities; introduces information balance optimization to ensure the comprehensive capture of the hospital's multi-dimensional data; based on the local information distribution adjustment mechanism, the data of different departments can be balanced in the optimization process to avoid bias; thereby improving the accuracy of hospital information supervision.
[0034] Example 4, see Figure 1This embodiment is based on the above embodiment. In step S3, hospital information supervision is dynamically grouped based on a clustering algorithm, and abnormal status is detected by calculating anomaly scores. In hospital operations, anomaly detection is crucial and can detect problems such as resource abuse, equipment failure, or unreasonable staff scheduling. Hospital data will be projected into a new space, and the location y of each cluster center is obtained by K-means clustering. The cluster center is expressed as: ;in, and are the positions of the k-th cluster center at the t-th and t+1-th iterations respectively; is the jth hospital data sample; h is the adjustment parameter that controls the scale of the distance measurement; is a Gaussian kernel function; after clustering, the abnormality score of the hospital data is calculated , expressed as: ; ; ;in, is the average position of the cluster to which the i-th data belongs; is the standard deviation; is the local density estimate; is the maximum local density estimate; is the abnormal adjustment parameter; is a density measure; K is the total number of clusters; is the position of the kth cluster center at the tth iteration; is the error adjustment parameter; is the average anomaly score of the previous N data points; is the scaling factor; sets the scoring threshold; scoring threshold It is expressed as: ;in, and are the mean and standard deviation of the anomaly score, respectively; is the adjustment factor; if the abnormal score of the hospital data is greater than the score threshold, it is marked as an abnormal state; and the management personnel are warned.
[0035] Example 5, see Figure 1 This embodiment is based on the above embodiment. In step S4, the adaptive optimization automatically adjusts the feature weights to adapt to the dynamic changes of hospital data. The data in hospital operation is dynamically changing, so it is necessary to adjust the clustering mode through adaptive optimization to adapt to the changes of hospital data. When new sampling data is continuously introduced, the projection matrix D is updated in an iterative manner to adapt to the new data distribution, and the feature adjustment factor is introduced at the same time. , which is used to adjust the intensity of department clustering adjustment; the feature weight update is expressed as: ; ;in, is the rate of change of the feature at the tth iteration; is the feature weight at the tth iteration; is the regularization parameter; is the feature adjustment parameter; It is The moderating factor of each feature; and is the characteristic value of the i-th sample data; and It is the eigenvalue of the cluster center to which the i-th sample data belongs.
[0036] By performing the above operations, in view of the lack of real-time dynamic anomaly detection and processing mechanism in general hospital information supervision methods, feature selection and weight allocation cannot respond in time when facing rapid changes in patient flow and department load, resulting in the inability to quickly discover potential problems, affecting hospital operation efficiency, and thus leading to poor hospital information supervision effects. This solution introduces local density estimation, density measurement and error adjustment parameters to calculate the anomaly score of each sample, so that the anomaly score can more comprehensively reflect the degree of abnormality of the data and improve the accuracy of anomaly detection; adaptive optimization is performed based on feature adjustment factors to ensure adaptation to the continuously changing data characteristics in hospital operations; thereby improving the effect of hospital information supervision.
[0037] Example 6, see Figure 2 , this embodiment is based on the above embodiment, the intelligent hospital information supervision system based on clustering processing provided by the present invention includes a data acquisition module, a dynamic feature projection module, a hospital information supervision module and an adaptive optimization module;
[0038] The data acquisition module acquires hospital information data;
[0039] The dynamic feature projection module performs dimensionality reduction processing through feature projection;
[0040] The hospital information supervision module dynamically groups hospital data using a clustering algorithm and calculates anomaly scores to detect abnormal conditions;
[0041] The adaptive optimization module dynamically adjusts the feature weights and utilizes feature adjustment factors to adjust the adjustment strength of department clustering.
[0042] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.
[0043] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.
[0044] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.
Claims
1. A smart hospital information supervision method based on clustering processing, characterized by: The method comprises the following steps: Step S1: data acquisition: acquiring hospital information data; Step S2: dynamic feature projection; Step S3: Hospital information supervision: dynamically group hospital data through clustering algorithm and calculate anomaly scores to detect abnormal status; Step S4: adaptive optimization; In step S1, the data acquisition is to acquire hospital information data; combining the characteristics of hospital departments and patient flow, key department operation data is collected first, and department priority It is expressed as: ; Dynamically adjust data collection frequency , expressed as: ;in, is the weight factor of department load; It is the department load; is the weighting factor of patient flow; is the rate of change of patient flow; is the basic sampling frequency; It is the department flow; the hospital information data includes department load data, patient flow data, department resource usage data, resource allocation data and sampling time.
2. The intelligent hospital information supervision method based on clustering processing according to claim 1 is characterized by: In step S2, the dynamic feature projection is a selection feature; Filter irrelevant features; The projection matrix of the data is represented as: ; Introduce the local information distribution adjustment mechanism, the objective function It is expressed as: ; D is the projection matrix; n is the number of hospital data samples, and i is the sample index; is the i-th data point; is the projected data; is the expected value; is the L2 norm; P is the total number of features, l is the feature index; is the feature weight; is the local control weight; M is the number of departments.
3. The intelligent hospital information supervision method based on clustering processing according to claim 2 is characterized by: In step S3, the hospital information supervision is based on dynamic grouping by clustering algorithm, and abnormal status is detected by calculating abnormal score; hospital data will be projected into a new space, and the location y of each cluster center is obtained by K-means clustering; the cluster center is expressed as: ;in, and are the positions of the k-th cluster center at the t-th and t+1-th iterations respectively; is the jth hospital data sample; h is the adjustment parameter that controls the scale of the distance measurement; is a Gaussian kernel function; after clustering, the abnormality score of the hospital data is calculated , expressed as: ; ; ;in, is the average position of the cluster to which the i-th data belongs; is the standard deviation; is the local density estimate; is the maximum local density estimate; is the abnormal adjustment parameter; is a density measure; K is the total number of clusters; is the position of the kth cluster center at the tth iteration; is the error adjustment parameter; is the average anomaly score of the previous N data points; is the scaling factor; a scoring threshold is set; if the abnormal score of hospital data is greater than the scoring threshold, it is marked as an abnormal state; and an early warning is issued to the management personnel.
4. The intelligent hospital information supervision method based on clustering processing according to claim 3 is characterized by: In step S4, the adaptive optimization is to automatically adjust the feature weights; When new sample data is continuously introduced, the projection matrix D is updated iteratively to adapt to the new data distribution, and the feature adjustment factor is introduced , which is used to adjust the intensity of department clustering adjustment; the feature weight update is expressed as: ; ;in, is the rate of change of the feature at the tth iteration; is the feature weight at the tth iteration; is the regularization parameter; is the feature adjustment parameter; It is The moderating factor of each feature; and is the characteristic value of the i-th sample data; and It is the eigenvalue of the cluster center to which the i-th sample data belongs.
5. A smart hospital information supervision system based on clustering processing, used to implement a smart hospital information supervision method based on clustering processing as described in any one of claims 1 to 4, characterized in that: It includes data acquisition module, dynamic feature projection module, hospital information supervision module and adaptive optimization module.
6. The intelligent hospital information supervision system based on clustering processing according to claim 5 is characterized by: The data acquisition module acquires hospital information data; The dynamic feature projection module performs dimensionality reduction processing through feature projection; The hospital information supervision module dynamically groups hospital data using a clustering algorithm and calculates anomaly scores to detect abnormal conditions; The adaptive optimization module dynamically adjusts the feature weights and utilizes feature adjustment factors to adjust the adjustment strength of department clustering.