Diagnosis waiting sorting intelligent management system based on multi-source data analysis

Abstract

The invention belongs to the technical field of diagnosis waiting management, and particularly relates to a diagnosis waiting sorting intelligent management system based on multi-source data analysis, which comprises a diagnosis waiting data fusion preprocessing module, an illness state emergency quantitative analysis module, a multi-dimensional diagnosis waiting priority decision module and an intelligent sorting result visualization module. According to the method, the illness state emergency degree is quantitatively calculated from three dimensions of symptom emergency degree, sign abnormity degree and basic disease risk through an illness state emergency degree quantitative analysis module, and it is ensured that medical resources are reasonably inclined to high-emergency-degree patients; a multi-dimensional waiting priority decision-making module calculates waiting priority coefficients through multi-dimensional analysis and sorts the waiting priority coefficients, an intelligent sorting result visualization module displays waiting queues, medical safety, waiting fairness and diagnosis and treatment efficiency are effectively balanced, and the diagnosis and treatment efficiency is improved. And accurate and reasonable sorting of the waiting patients is realized.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of queuing management, and in particular to a queuing intelligent management system based on multi-source data analysis. BACKGROUND

[0002] In the field of medical service management, a queuing system is a core support for optimizing diagnosis and treatment processes and improving service efficiency, and the scientificity and rationality directly affect the utilization rate of medical resources and the medical experience of patients. With the advancement of medical informatization, various intelligent queuing and calling technologies have emerged.

[0003] At present, queuing technologies mainly focus on the prediction of queuing time or the simple adjustment of queues. For example, a method and device for calculating outpatient queuing time are disclosed in Chinese Patent No. CN106372391A. The core of the technical solution is to obtain the registration time of a patient and the historical diagnosis efficiency data of a department and a doctor, calculate and predict the queuing waiting time of the patient, and reduce the invalid waiting of the patient through time prediction.

[0004] However, the queuing logic of the above-mentioned invention only focuses on the registration time sequence and historical diagnosis efficiency, and cannot integrate the multi-dimensional factors of disease urgency, queuing time fairness, patient special needs and historical queuing performance, so it is difficult to realize accurate and reasonable queuing of queuing patients, resulting in low efficiency of medical resource allocation, delayed treatment of emergency patients and poor patient medical experience, and the balance between medical safety, queuing fairness and diagnosis efficiency cannot be effectively achieved. Therefore, a solution is proposed. SUMMARY

[0005] The purpose of the present application is to provide a queuing intelligent management system based on multi-source data analysis to solve the technical defects proposed in the background art.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical solution: a queuing intelligent management system based on multi-source data analysis, comprising a queuing data fusion preprocessing module, a disease emergency quantification analysis module, a multi-dimensional queuing priority decision module and an intelligent queuing result visualization module.

[0007] The queuing data fusion preprocessing module collects multi-source heterogeneous data of queuing patients and pre-processes and fuses the data, and outputs the fused standardized data set in real time.

[0008] The disease emergency quantification analysis module quantifies the disease emergency degree of the patient from three dimensions of symptom emergency degree, sign abnormality degree and basic disease risk, and sends the disease emergency degree of the patient to the multi-dimensional queuing priority decision module.

[0009] The multi-dimensional queuing priority decision module obtains a queuing priority coefficient of the corresponding queuing patient through multi-dimensional queuing priority analysis calculation, sorts all queuing patients of the corresponding department according to the values of the queuing priority coefficients from large to small, generates a department queuing queue, and sends the department queuing queue to the intelligent queuing result visualization module. The intelligent queuing result visualization module displays the department queuing queue.

[0010] Further, the queuing data fusion preprocessing module connects the hospital HIS system, LIS system and electronic medical record system through an API interface, collects patient basic information, registration information and inspection data, obtains real-time patient current sign data through Internet of Things connection of sign monitoring equipment, and collects special demand declaration and illness supplement description through a patient terminal applet;

[0011] The collected data is cleaned. For missing values, a multiple imputation method based on similar patient medical records is used. For abnormal values, the quartile range method is used to identify and eliminate invalid data. For format differences, all data is converted to JSON format, and indexed according to "patient ID-data type-collection time". The data is fused to generate a standardized data set containing patient unique identification, basic information, real-time signs, chief complaint symptoms, registration duration and special needs, and output.

[0012] Further, the operation analysis process of the illness emergency quantification analysis module includes:

[0013] Data extraction: receiving the patient standardized data set transmitted by the queuing data fusion preprocessing module, extracting the key fields related to the illness emergency degree, including chief complaint symptoms, real-time signs and basic disease history;

[0014] Calculating the symptom emergency degree S: mapping the patient's chief complaint symptoms to the initial emergency degree Sy by comparing the set symptom emergency degree classification table, and the symptom emergency degree S = max(Sy1, Sy2, …, Syg), where g is the number of chief complaint symptoms of the queuing patient;

[0015] Calculating the sign abnormality degree T: obtaining the abnormal deviation degree of each sign index by analysis, and calculating the mean value of the abnormal deviation degrees of all sign indexes to obtain the sign abnormality degree T;

[0016] Calculating the basic disease risk value B: according to the relevance value of the basic disease and the current symptoms, obtaining the basic disease risk value B of the corresponding queuing patient;

[0017] Determine the weights a, b, c of S, T, B by the analytic hierarchy process, and calculate the illness emergency degree E by the formula E = a x S + b x T + c x B. The E value and the S, T, B parameters in the calculation process are transmitted to the multi-dimensional queuing priority decision module in real time.

[0018] Furthermore, the specific methods for analyzing and obtaining abnormal deviations in vital signs are as follows:

[0019] All vital signs indicators were obtained, and the measured values ​​of the corresponding vital signs indicators were collected. The abnormal deviation of the corresponding vital signs indicators was calculated using the formula for calculating the abnormality of vital signs.

[0020] Furthermore, the multi-dimensional waiting priority decision module communicates with the correction factor calculation module. The correction factor calculation module analyzes the waiting records of the corresponding waiting patients in the historical stage to obtain the target correction factor e of the corresponding waiting patients, and sends the target correction factor e of the corresponding waiting patients to the multi-dimensional waiting priority decision module in real time.

[0021] Furthermore, the specific method for obtaining the target correction factor e is as follows:

[0022] The system obtains the total number of waiting times and the number of no-shows for the corresponding patients in the historical period. The ratio of no-shows to total waiting times is used to calculate the waiting anomaly value. When there are no no-shows, the time difference between the waiting time and the consultation time is calculated to obtain the consultation delay value. The average of all consultation delay values ​​for the corresponding patients in the historical period is used to calculate the consultation delay table value. The waiting traceability coefficient is obtained by weighted summation of the waiting anomaly value and the consultation delay table value.

[0023] Several preset waiting period traceability coefficient ranges are set in advance, and each preset waiting period traceability coefficient range corresponds to a set of decision correction factors. The waiting period traceability coefficient of the corresponding waiting patient is compared with all preset waiting period traceability coefficient ranges one by one, and the preset waiting period traceability coefficient range containing the corresponding waiting period traceability coefficient is marked as the target range, and the decision correction factor corresponding to the target range is marked as the target correction factor e.

[0024] Furthermore, the operational analysis process of the multi-dimensional waiting priority decision module is as follows:

[0025] The registration time of the corresponding waiting patients is collected, and the waiting time Ds is calculated by the time difference between the current time and the registration time. The maximum reasonable waiting time of 120 minutes is used as the threshold. The influence value D of the waiting time is obtained by analyzing it using the formula D=min{(Ds / 120)×100,100}. If Ds>120, then D=100.

[0026] And determine the special needs coefficient K; among which, K = 1.15 for elderly patients over 65 years old, K = 1.2 for children under 5 years old, K = 1.25 for disabled patients, and K = 1.0 for ordinary patients;

[0027] The urgency level E and target correction factor e of the corresponding waiting patients are retrieved from the urgency quantification analysis module and the correction factor calculation module, respectively. The waiting priority coefficient P of the corresponding waiting patients is then calculated using the multi-dimensional waiting priority calculation formula.

[0028] Furthermore, the intelligent sorting result visualization module is connected to the backend management terminal. The intelligent sorting result visualization module and the backend management terminal communicate bidirectionally. The intelligent sorting result visualization module transmits the received information to the backend management terminal, and the backend management terminal controls the intelligent sorting result visualization module by issuing control commands to it.

[0029] Furthermore, the background management terminal communicates with the visualization performance evaluation module. The visualization performance evaluation module evaluates and analyzes the performance of the intelligent sorting result visualization module. Through analysis, it determines whether a visualization anomaly signal is generated. When a visualization anomaly signal is generated, it is sent to the background management terminal. When the background management terminal receives the visualization anomaly signal, it issues a corresponding warning.

[0030] Furthermore, the specific analysis process of the visualization performance evaluation module is as follows:

[0031] The system obtains the number of times the intelligent sorting result visualization module fails to display within a unit of time and marks them as abnormal frequency values ​​for waiting display. It also marks the total duration of the intelligent sorting result visualization module's failure to display within a unit of time as the waiting black screen status value. The abnormal frequency values ​​for waiting display and the waiting black screen status value are compared with the preset abnormal frequency thresholds for waiting display and the preset waiting black screen status thresholds, respectively. If the abnormal frequency value for waiting display or the waiting black screen status value exceeds the corresponding preset threshold, a visualization abnormal signal is generated.

[0032] If the abnormal frequency value and the black screen status value of the waiting room do not exceed the corresponding preset threshold, the time when the intelligent sorting result visualization module receives the department's waiting queue is obtained and marked as the first time. The time when the intelligent sorting result visualization module updates the waiting queue is marked as the second time. The time difference between the first time and the second time is calculated to obtain the waiting update delay value. The average of all waiting update delay values ​​within a unit of time is calculated to obtain the waiting delay coefficient.

[0033] The system obtains the display brightness of the intelligent sorting result visualization module, calculates the difference between the display brightness and the median of the preset suitable brightness range, and takes the absolute value to obtain the waiting display brightness value. The system also calculates the average of the waiting display brightness values ​​within a unit of time to obtain the waiting display brightness characteristic value. When the waiting display brightness value exceeds the preset waiting display brightness threshold, the system determines that the intelligent sorting result visualization module is in a high brightness bias state, and marks the total duration of the intelligent sorting result visualization module in a high brightness bias state within a unit of time as the waiting brightness high bias value.

[0034] The visualization performance evaluation value is obtained by weighted summation of the waiting delay coefficient, the waiting brightness characteristic value, and the waiting brightness high bias value. The visualization performance evaluation value is then compared with the preset visualization performance evaluation threshold. If the visualization performance evaluation value exceeds the preset visualization performance evaluation threshold, a visualization abnormality signal is generated.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. In this invention, the urgency of a patient's condition is quantified and calculated through a urgency quantification and analysis module to ensure that medical resources are reasonably allocated to patients with high urgency. The correction factor calculation module incorporates the patient's historical waiting performance into the priority consideration to avoid patients who frequently miss appointments or delay their visits from excessively occupying medical resources. Furthermore, the waiting priority coefficient is calculated through multi-dimensional analysis, and the waiting queue is sorted and displayed based on the waiting priority coefficient. This effectively balances medical safety, waiting fairness, and treatment efficiency, achieving accurate and reasonable sorting of waiting patients.

[0037] 2. In this invention, the performance of the intelligent sorting result visualization module is evaluated and analyzed through the visualization performance evaluation module. The operation status of the intelligent sorting result visualization module can be monitored in real time, and problems such as display abnormalities, update delays, and unsuitable brightness can be detected and warned in a timely manner. This avoids medical staff misjudging the waiting order or patients becoming confused due to information display problems caused by display failures, ensuring the stability, timeliness and accuracy of the waiting queue display, and maintaining the smooth operation of the entire waiting management process. Attached Figure Description

[0038] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0039] Figure 1 This is a system block diagram of Embodiment 1 of the present invention;

[0040] Figure 2 This is a system block diagram of Embodiment 2 of the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Example 1: As Figure 1As shown, the intelligent waiting list management system based on multi-source data analysis proposed in this invention includes a waiting list data fusion and preprocessing module, a disease urgency quantification and analysis module, a correction factor calculation module, a multi-dimensional waiting list priority decision module, and an intelligent ranking result visualization module.

[0043] The waiting data fusion and preprocessing module collects multi-source heterogeneous data from waiting patients, performs preprocessing and fusion, and outputs the fused standardized dataset in real time. This effectively solves the problems of scattered data sources, messy formats, and inconsistent quality in the traditional data collection process. It provides high-quality and highly consistent data support for subsequent modules such as emergency quantification analysis and multi-dimensional waiting priority decision-making, avoiding deviations in subsequent analysis results due to data problems and ensuring the reliability of the system's ranking decision from the source.

[0044] Specifically, the waiting data fusion and preprocessing module connects to the hospital's HIS system, LIS system, and electronic medical record system through API interfaces to collect patients' basic information, registration information, and test data. It also connects to vital sign monitoring devices through the Internet of Things to obtain patients' current vital sign data in real time, and collects special needs declarations and supplementary descriptions of their condition through a patient-side mini-program.

[0045] The collected data was cleaned. For missing values, a multiple imputation method based on similar patient medical records was used. For outliers, the interquartile range method was used to identify and remove invalid data. For format differences, all data was uniformly converted to JSON format and indexed according to "patient ID-data type-collection time". The data was then fused to generate a standardized dataset containing the patient's unique identifier, basic information, real-time vital signs, chief symptoms, registration time, and special needs, and then output it.

[0046] The urgency quantification analysis module quantifies the urgency of a patient's condition from three dimensions: symptom urgency, sign abnormality, and underlying disease risk. It then sends this urgency information to a multi-dimensional waiting priority decision module. This overcomes the limitations of traditional methods that rely solely on a single symptom or subjective judgment to assess urgency. By quantifying urgency from three dimensions—symptoms, signs, and underlying diseases—it more accurately identifies patients truly requiring emergency treatment, preventing delays in treatment for critical cases or the occupation of emergency medical resources by ordinary patients. This significantly improves the scientific rigor of medical judgment and the safety of medical treatment. The operational analysis process of the urgency quantification analysis module includes:

[0047] Data extraction: Receive standardized patient datasets transmitted from the waiting data fusion and preprocessing module, and extract key fields related to the urgency of the condition, mainly including chief symptoms (such as chest pain, dyspnea, high fever, etc.), real-time vital signs (such as HR, SpO2, etc.) and underlying medical history (such as coronary heart disease, hypertension, diabetes, etc.).

[0048] Calculate the symptom urgency S (S value range 0-100): Refer to the established symptom urgency grading table, preferably the "ICD-11 Symptom Urgency Grading Table"; map the patient's chief complaint symptoms to an initial urgency score Sy (e.g., acute chest pain Sy = 90, mild headache Sy = 15), symptom urgency S = max{Sy1, Sy2, ..., Syg}, that is, when there are multiple symptoms, the maximum value is taken as S; where g is the number of chief complaints of the waiting patient;

[0049] Calculate the abnormality degree of vital signs T (T value range 0-100): The abnormal deviation degree of each vital sign indicator is obtained through analysis, and the abnormal deviation degree of all vital sign indicators is averaged to calculate the abnormality degree T. The specific method for analyzing and obtaining the abnormal deviation degree of vital sign indicators is as follows:

[0050] All vital signs indicators were obtained, and the measured values ​​of the corresponding vital signs indicators were collected. The abnormal deviation degree of the corresponding vital signs indicators was calculated using the formula for calculating the abnormality degree of vital signs, which is as follows:

[0051]

[0052] Where Ti represents the degree of abnormal deviation of the i-th physical trait (0 to 100, the higher the value, the more severe the abnormality);

[0053] Xi represents the measured value of the i-th physical indicator of the corresponding waiting patient;

[0054] Xn represents the normal clinical reference value for the i-th physical indicator;

[0055] Xmax represents the upper limit of clinical risk for the i-th physical indicator (e.g., heart rate Xmax = 150 beats / min);

[0056] Xmin represents the lower limit of clinical risk for the i-th physical indicator (e.g., heart rate Xmin = 40 beats / min);

[0057] Calculate the underlying disease risk value B (B value range 0-100): Assign a value based on the correlation between the underlying disease and the current symptoms. For example, if a patient has coronary heart disease and complains of chest pain, B=95; if a patient has diabetes but complains of a cold, B=30; and if a patient has no underlying disease, B=10. Based on this, obtain the corresponding underlying disease risk value B for the waiting patient.

[0058] Using the Analytic Hierarchy Process (AHP) combined with the opinions of more than 30 clinical experts, the weights a, b, and c of S, T, and B are determined. Preferably, a = 0.45, b = 0.35, and c = 0.2. The urgency level E is calculated using the formula E = a × S + b × T + c × B. The E value and the S, T, and B parameters used in the calculation process are transmitted in real time to the multi-dimensional waiting priority decision module. Preferably, E ≥ 80 is considered urgent, 60 ≤ E < 80 is considered moderately urgent, and E < 60 is considered normal.

[0059] The correction factor calculation module analyzes the waiting records of corresponding patients in historical periods to obtain the target correction factor e for each patient. This target correction factor e is then sent in real-time to the multi-dimensional waiting priority decision module. The correction factor is used to appropriately adjust the patient's waiting priority, preventing patients from frequently occupying medical resources without attending on time. Simultaneously, it ensures the fair waiting rights of patients with good historical records, further optimizing the efficiency of medical resource allocation and maintaining the fairness of the waiting order. The specific method for obtaining e is as follows:

[0060] The system obtains the total number of waiting times and the number of no-shows (i.e., the number of times a patient fails to see a doctor within the specified time after being called) for the corresponding waiting period in the historical stage. The ratio of the number of no-shows to the total number of waiting times is used to calculate the waiting abnormal value. When there are no no-shows, the time difference between the time of being called and the time of seeing a doctor is calculated to obtain the doctor's delay value. The system then calculates the average of all doctor's delay values ​​for the corresponding patients in the historical stage to obtain the doctor's delay table value.

[0061] The waiting period traceability coefficient is calculated by weighted summation of abnormal waiting period values ​​and delayed appointment values. Specifically, each abnormal waiting period value and delayed appointment value is assigned a corresponding preset weight coefficient, and each value is multiplied by its respective preset weight coefficient. The sum of the two products is then marked as the waiting period traceability coefficient. It should be noted that a large waiting period traceability coefficient indicates that the corresponding patient has a poor history of waiting and is slow in waiting, and their priority can be appropriately reduced.

[0062] Several preset waiting period traceability coefficient ranges are pre-defined, and each preset waiting period traceability coefficient range corresponds to a set of decision correction factors with values ​​greater than zero. Furthermore, the larger the value of the corresponding preset waiting period traceability coefficient range, the larger the value of the decision correction factor that matches it. The waiting period traceability coefficient of the corresponding patient is compared with all preset waiting period traceability coefficient ranges one by one, and the preset waiting period traceability coefficient range containing the corresponding waiting period traceability coefficient is marked as the target range, and the decision correction factor corresponding to the target range is marked as the target correction factor e.

[0063] The multi-dimensional waiting priority decision module calculates the waiting priority coefficient for each patient through multi-dimensional waiting priority analysis. Based on the value of the waiting priority coefficient, all patients waiting in the corresponding department are sorted for treatment from largest to smallest. The module generates a department waiting queue and sends it to the intelligent sorting result visualization module. The intelligent sorting result visualization module displays the department waiting queue, allowing patients to understand their waiting progress, reducing anxiety and disputes caused by information asymmetry, and effectively improving the user experience and communication efficiency for both parties.

[0064] The multi-dimensional waiting priority decision-making module integrates multiple factors such as the urgency of the illness, fairness of waiting time, special needs of patients, and historical waiting performance. This ensures priority access for patients with urgent conditions while preventing patient dissatisfaction due to excessively long waiting times. Furthermore, the special needs coefficient K reflects humanistic care for vulnerable patient groups, effectively balancing medical safety, fairness of waiting times, and efficiency of treatment, significantly improving the rationality and humanization of overall waiting management. The specific operational analysis process of the multi-dimensional waiting priority decision-making module is as follows:

[0065] The registration time of the corresponding waiting patients is collected, and the waiting time Ds is calculated by the time difference between the current time and the registration time. The maximum reasonable waiting time of 120 minutes is used as the threshold. The waiting time influence value D is obtained by analyzing the formula D=min{(Ds / 120)×100,100}. That is, the waiting time Ds is converted into a value of 0 to 100 by the formula. The larger Ds is, the larger D is. If Ds>120, then D=100.

[0066] And determine the special needs coefficient K, which can reflect humanistic care; for example, K = 1.15 for elderly patients over 65 years old, K = 1.2 for children under 5 years old, K = 1.25 for disabled patients, and K = 1.0 for ordinary patients;

[0067] The urgency level E and target correction factor e of the corresponding waiting patients are retrieved from the urgency quantification analysis module and the correction factor calculation module, respectively. The waiting priority coefficient P of the corresponding waiting patients is calculated using the multi-dimensional waiting priority calculation formula, which is as follows:

[0068] P = K × (w1 × E + w2 × D) - e; where w1 and w2 are preset weight coefficients, and w1 + w2 = 1; it should be noted that the larger the value of the waiting priority coefficient P, the higher the waiting priority of the corresponding patient and the earlier the waiting order.

[0069] Example 2: Figure 2As shown, the difference between this embodiment and Embodiment 1 is that the intelligent sorting result visualization module is communicatively connected to the backend management terminal, and the intelligent sorting result visualization module and the backend management terminal communicate bidirectionally. The intelligent sorting result visualization module transmits the received information to the backend management terminal, and the backend management terminal controls the intelligent sorting result visualization module by issuing control commands to it.

[0070] Furthermore, the background management terminal communicates with the visualization performance evaluation module. The visualization performance evaluation module evaluates and analyzes the performance of the intelligent sorting result visualization module. Through analysis, it determines whether a visualization anomaly signal is generated. When a visualization anomaly signal is generated, it is sent to the background management terminal. When the background management terminal receives the visualization anomaly signal, it issues a corresponding warning.

[0071] The visualization performance evaluation module monitors the operational status of the intelligent sorting result visualization module in real time, promptly identifying issues such as display anomalies, update delays, and inappropriate brightness. These issues are then promptly addressed and resolved via the backend management terminal. This prevents medical staff from misjudging the waiting order or patients from becoming confused due to information display problems caused by display malfunctions, ensuring the stability, timeliness, and accuracy of the waiting queue display and maintaining the smooth operation of the entire waiting management process. The specific analysis process of the visualization performance evaluation module is as follows:

[0072] The number of times the intelligent sorting result visualization module fails to display within a unit of time is obtained and marked as the abnormal frequency value of the waiting display. The total duration of the intelligent sorting result visualization module failing to display within a unit of time is marked as the waiting black screen status value. The abnormal frequency value of the waiting display and the waiting black screen status value are compared with the preset abnormal frequency threshold and the preset waiting black screen status threshold, respectively. If the abnormal frequency value of the waiting display or the waiting black screen status value exceeds the corresponding preset threshold, it indicates that the intelligent sorting result visualization module is difficult to display stably and reliably within a unit of time, and a visualization abnormal signal is generated.

[0073] If the abnormal frequency value and the black screen status value of the waiting room do not exceed the corresponding preset threshold, the time when the intelligent sorting result visualization module receives the department's waiting queue is obtained and marked as the first time. The time when the intelligent sorting result visualization module updates the waiting queue is marked as the second time. The time difference between the first time and the second time is calculated to obtain the waiting update delay value. The average of all waiting update delay values ​​within a unit of time is calculated to obtain the waiting delay coefficient.

[0074] The system obtains the display brightness of the intelligent sorting result visualization module, calculates the difference between the display brightness and the median of the preset suitable brightness range, and takes the absolute value to obtain the waiting display brightness value. The system also calculates the average of the waiting display brightness values ​​within a unit of time to obtain the waiting display brightness characteristic value. When the waiting display brightness value exceeds the preset waiting display brightness threshold, the system determines that the intelligent sorting result visualization module is in a high brightness bias state, and marks the total duration of the intelligent sorting result visualization module in a high brightness bias state within a unit of time as the waiting brightness high bias value.

[0075] The visualization performance evaluation value is obtained by weighting and summing the waiting delay coefficient, the waiting brightness feature value, and the waiting brightness high bias value. Specifically, the waiting delay coefficient, the waiting brightness feature value, and the waiting brightness high bias value are each assigned a corresponding preset weight coefficient, and the waiting delay coefficient, the waiting brightness feature value, and the waiting brightness high bias value are each multiplied by the corresponding preset weight coefficient. The sum of the three sets of product results is marked as the visualization performance evaluation value.

[0076] It should be noted that the higher the value of the visualization performance evaluation, the more abnormal the operation of the intelligent sorting result visualization module is within a unit of time, and the worse the visualization performance is. The visualization performance evaluation value is compared with the preset visualization performance evaluation threshold. If the visualization performance evaluation value exceeds the preset visualization performance evaluation threshold, it indicates that the operation of the intelligent sorting result visualization module is abnormal within a unit of time, and the visualization performance is poor. In this case, a visualization abnormality signal is generated.

[0077] The working principle of this invention is as follows: During use, the waiting data fusion preprocessing module cleans and standardizes multi-source heterogeneous data, generating a unified dataset containing key patient information. The urgency quantification analysis module quantifies the urgency of the condition from three dimensions: symptom urgency, abnormal signs, and underlying disease risk. This accurately identifies patients requiring priority treatment, such as those with acute chest pain, high fever accompanied by underlying diseases, effectively avoiding medical safety risks associated with delayed treatment of urgent conditions and ensuring that medical resources are rationally allocated to patients with high urgency. The correction factor calculation module incorporates the patient's historical waiting performance into the priority consideration, preventing patients who frequently fail to show up or delay their visits from excessively occupying medical resources and further maintaining the fairness of the waiting order. The multi-dimensional waiting priority decision module integrates the urgency of the condition, the impact value of waiting time, the special needs coefficient, and the target correction factor. Through multi-dimensional analysis, it calculates and ranks the waiting priority coefficients, effectively balancing medical safety, waiting fairness, and treatment efficiency, reducing patient dissatisfaction. The intelligent ranking result visualization module displays the waiting queue, reducing anxiety and disputes caused by information asymmetry.

[0078] In this invention, the threshold, preset value, or preset range settings are for result comparison and analysis to determine whether the result is good or bad. The magnitude of these values ​​is determined by a combination of large-scale model analysis of sample data and human experience, and can also be appropriately adjusted based on seasonal or common-sense influence conditions. Similarly, the preset weight coefficients and influence factors are assigned specific values ​​based on the magnitude of each parameter's influence on the result, ultimately reflecting the impact on the result. These settings are also determined by a combination of large-scale model analysis of sample data and human experience, and can also be appropriately adjusted based on seasonal or common-sense influence conditions.

[0079] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, enabling those skilled in the art to better understand and utilize it. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A queuing intelligent management system based on multi-source data analysis, characterized in that, The application comprises a waiting data fusion preprocessing module, a disease emergency quantification analysis module, a multi-dimensional waiting priority decision module and an intelligent sorting result visualization module. The waiting data fusion preprocessing module collects multi-source heterogeneous data of the waiting patients and pre-processes and fuses the data, and outputs the fused standardized data set in real time. The disease emergency quantification analysis module quantifies the emergency degree of the patient from three dimensions of symptom emergency degree, sign abnormality degree and basic disease risk, and sends the emergency degree of the patient to the multi-dimensional waiting priority decision module. The multi-dimensional waiting priority decision module obtains the waiting priority coefficient of the corresponding waiting patient through multi-dimensional waiting priority analysis calculation, sorts all the waiting patients in the corresponding department according to the values of the waiting priority coefficients from large to small, generates a department waiting queue and sends it to the intelligent sorting result visualization module, and the intelligent sorting result visualization module displays the department waiting queue. 2.The intelligent management system for queuing based on multi-source data analysis according to claim 1, wherein, The waiting data fusion preprocessing module connects the hospital HIS system, LIS system and electronic medical record system through an API interface, collects the basic information, registration information and inspection data of the patients, obtains the current sign data of the patients in real time through the Internet of Things, and collects the special demand declaration and disease supplement description through the patient terminal applet. The collected data is cleaned, the multiple imputation method based on similar patient medical records is used for missing values, the quartile range method is used to identify and eliminate invalid data for abnormal values, and all data is converted to JSON format for format differences; the data is fused to generate a standardized data set and output. 3.The intelligent management system for queuing based on multi-source data analysis according to claim 1, wherein, The operation analysis process of the disease emergency quantification analysis module includes: Receiving the patient standardized data set transmitted by the waiting data fusion preprocessing module, extracting the key fields related to the emergency degree of the disease, including the chief complaint, real-time signs and basic disease history; mapping the patient's chief complaint to the initial emergency degree Sy according to the set symptom emergency degree classification table, and the symptom emergency degree S = max (Sy1, Sy2, …, Syg), wherein g is the number of chief complaints of the waiting patient; The abnormal deviation degree of each sign index is obtained by analysis, and the mean value of the abnormal deviation degrees of all sign indexes is calculated to obtain the sign abnormality degree T; the basic disease risk value B of the corresponding waiting patient is obtained according to the correlation value of the basic disease and the current symptoms; the weights a, b and c of S, T and B are determined by the analytic hierarchy process, and the disease emergency degree E is calculated by the formula E = a × S + b × T + c × B. 4.The intelligent management system for queuing based on multi-source data analysis according to claim 3, characterized in that, The abnormal deviation degree of the sign index is obtained by analyzing the measured value of the corresponding sign index, and the abnormal deviation degree of the corresponding sign index is calculated by the sign abnormality degree calculation formula. 5.The intelligent management system for queuing based on multi-source data analysis according to claim 1, wherein, The multi-dimensional waiting priority decision module is connected with the correction factor calculation module, which analyzes the waiting records of the corresponding waiting patient in the historical stage to obtain the target correction factor e of the corresponding waiting patient, and sends the target correction factor e of the corresponding waiting patient to the multi-dimensional waiting priority decision module in real time. 6.The intelligent management system for queuing based on multi-source data analysis according to claim 5, wherein, The analysis acquisition method of the target correction factor e is specifically as follows: The waiting patient tracing coefficient is calculated by weighted summation of the waiting abnormal value and the treatment delay table value, the waiting patient tracing coefficient of the corresponding waiting patient is compared with all preset waiting patient tracing coefficient ranges one by one, the preset waiting patient tracing coefficient range containing the corresponding waiting patient tracing coefficient is marked as a target range, and the decision correction factor corresponding to the target range is marked as a target correction factor e. 7.The intelligent management system for waiting list ordering based on multi-source data analysis of claim 1, wherein, The running analysis process of the multi-dimensional waiting priority decision module is specifically as follows: The waiting time Ds is calculated by time difference between the current time and the registration time, the maximum reasonable waiting time is 120 minutes, the waiting time influence value D is obtained by analysis through the formula D = min {(Ds / 120) x 100, 100}; And determine the special demand coefficient K, retrieve the emergency degree E and the target correction factor e of the corresponding waiting patient, and calculate the waiting priority coefficient P of the corresponding waiting patient through the multi-dimensional waiting priority calculation formula. 8.The intelligent management system for waiting list ordering based on multi-source data analysis of claim 1, wherein, The intelligent sorting result visualization module is connected with the background control terminal, the intelligent sorting result visualization module transmits the received information to the background control terminal, and the background control terminal controls the intelligent sorting result visualization module by sending control instructions to it. 9.The intelligent management system for queuing based on multi-source data analysis according to claim 8, wherein, The background control terminal is connected with the visualization performance evaluation module, the visualization performance evaluation module evaluates and analyzes the running performance of the intelligent sorting result visualization module, judges whether to generate a visualization abnormal signal through analysis, and sends the visualization abnormal signal to the background control terminal when the visualization abnormal signal is generated. 10.The intelligent management system for queuing based on multi-source data analysis according to claim 9, wherein, The specific analysis process of the visualization performance evaluation module is as follows: if the waiting display abnormal frequency value or the waiting black screen situation value exceeds the corresponding preset threshold value, a visualization abnormal signal is generated; if the waiting display abnormal frequency value and the waiting black screen situation value do not exceed the corresponding preset threshold value, the visualization performance evaluation value is calculated by weighted summation of the waiting more delay coefficient, the waiting display bright characteristic value and the waiting brightness high bias value, and if the visualization performance evaluation value exceeds the preset visualization performance evaluation threshold value, a visualization abnormal signal is generated.

Images