Interface-free medical consultation number reporting control method, storage medium, system and number reporting machine

Abstract

The present invention discloses an interface-free medical consultation number reporting control method, storage medium, system and reporting machine, and relates to the field of number reporting control and adjustment of reporting machines. The present invention can dynamically adjust the department resource allocation priority through dual mapping based on real-time queue status and historical weight, reduce the waiting time of emergency patients, and balance the needs of follow-up patients; through the threshold trigger mechanism of the medical consultation accident number model, risks such as equipment failure and process blockage are identified in advance, so that the response speed of accidents is improved; the secondary adjustment mechanism ensures the operability of the adjustment from easy to difficult, thereby minimizing the time-consuming impact of the adjustment operation; by adjusting the number of spare reporting machines and other physical information that requires actual operation, the real-time total medical consultation time data and the real-time medical consultation accident number are further adjusted with greater intensity, ensuring that the real-time medical consultation accident number is controlled within a preset range.

Description

Technical Field

[0001] The present invention belongs to the field of number reporting control and regulation of a number reporting machine, and in particular relates to an interface-free medical number reporting control method, a storage medium, a system and a number reporting machine. Background Art

[0002] Number-calling machines are widely used in hospitals, train stations, banks, and other industries. When used in traditional hospitals, the static queuing and reporting rules during the consultation process are difficult to adapt to fluctuations in outpatient traffic. Furthermore, the traditional reporting model suffers from both idle resources and overcrowding, preventing the optimal utilization of hospital resources. Furthermore, the fragmented data between modules in traditional HIS systems makes it difficult to achieve cross-hospital collaboration and cross-departmental collaboration when reporting numbers. For example, collaboration with the imaging department and pathology department is often required during consultations. This inconvenience in cross-departmental collaboration leads to inconvenience in hospital operations during patient consultations. Summary of the Invention

[0003] In response to the problems in the related art, the present invention proposes an interface-free medical consultation number reporting control method, storage medium, system and reporting machine to overcome the above-mentioned technical problems existing in the existing related art.

[0004] To solve the above technical problems, the present invention is achieved through the following technical solutions:

[0005] The present invention is an interface-free medical consultation number reporting control method, comprising the following steps:

[0006] S1. Collect patient cohort status data corresponding to various medical scenarios;

[0007] S2. Collect historical patient queue status data, historical reporting weights, and equipment data corresponding to the various medical scenarios described in S1 in the current hospital;

[0008] S3: Collect the total time consumption data and the number of unexpected patient visits corresponding to several consultation cycles, and record them as historical consultation indicator data; and build a total time consumption mapping model and a number of unexpected patient visits mapping model by combining the historical patient queue status data, historical report number weights, and equipment data in S2;

[0009] S4, mapping the collected historical hospital number weights, equipment data, departmental remaining resource data, and patient queue status data collected in S1 into the total consultation time mapping model and the number of consultation accidents mapping model respectively;

[0010] S5. Comparing the mapping result in S4 with the corresponding preset threshold, adjusting the patient information weight and the continuous reporting time according to the comparison result;

[0011] S6. Based on the result of the first adjustment in S5, a second adjustment is made to the patient information weight, the continuous reporting time, and the number of spare reporting machines;

[0012] Based on the dual mapping of real-time queue status and historical weights, department resource allocation priorities can be dynamically adjusted to reduce waiting times for emergency patients while balancing the needs of follow-up patients. Through the threshold trigger mechanism of the unexpected number of visits model, risks such as equipment failure and process congestion can be identified in advance, thereby improving the response speed of unexpected events. The secondary adjustment mechanism ensures that the operability of adjustments is from easy to difficult, thereby minimizing the time-consuming impact of adjustment operations.

[0013] Preferably, the S1 comprises the following steps:

[0014] S11. Setting several types of patient visit scenarios to obtain a patient visit scenario type set; the patient visit scenario type set includes an emergency priority type, a checkup return type, and a multi-department collaboration type;

[0015] Then, respectively set the patient information types corresponding to the emergency priority type, the check-up return type, and the multi-department collaboration type to obtain an emergency priority type information type set, a check-up return type information type set, and a multi-department collaboration type information type set;

[0016] S12. Based on the emergency priority information type set, the examination and return visit information type set, and the multi-department collaboration information type set, the queue status of each patient in the HIS or LIS system is obtained in real time to obtain the current emergency priority data set, the current examination and return visit data set, and the current multi-department collaboration data set;

[0017] By dynamically binding emergency classification with vital signs, an automatic queue-jumping mechanism for critically ill patients is implemented, which greatly shortens the average waiting time of patients and greatly improves response efficiency compared to the traditional model; the latest examination results are automatically pushed to the rescue terminal, reducing the time spent on manual review; the check-back queue weight algorithm comprehensively considers parameters such as report review status and original waiting time to calculate the optimal time for return consultation, reducing patients' secondary waiting time and avoiding invalid waiting due to incomplete reports. The original clinic status synchronization function ensures that doctors are aware of the patient's examination progress in real time; and dynamic priorities are generated in combination with clinical parameters, which greatly reduces consultation preparation time and improves the on-time consultation rate.

[0018] Preferably, said S2 comprises the following steps:

[0019] S21. Based on the emergency priority information type set, the examination and return visit information type set, and the multi-department collaboration information type set, queue status data of multiple patients in the current hospital over several historical consultation cycles are collected to obtain a historical emergency priority data set, a historical examination and return visit data set, and a historical multi-department collaboration data set;

[0020] Then, the reporting weight data corresponding to each type of emergency priority information, examination and return visit information, and multi-department collaboration information when the hospital's reporting machine reports the number in each of the several medical cycles are collected to obtain a historical medical cycle reporting weight data set;

[0021] S22. Collect the remaining resources of each department of the current hospital at the beginning of the plurality of consultation cycles to obtain a historical department resource remaining amount set; set the data type of the consultation result within the consultation cycle, including the total consultation time data and the number of accidents of the patients;

[0022] Then, the number of standby number-calling machines configured in the current hospital corresponding to the plurality of medical treatment cycles, as well as the historical failure rate data and the average continuous number-calling time of the corresponding standby number-calling machines are collected to obtain a historical standby number-calling machine number set, a historical standby number-calling machine failure rate data set, and a historical continuous number-calling time data set;

[0023] Data analysis based on historical emergency data sets can identify the average retention patterns of different types of patients, greatly improving the timeliness of patient treatment. By setting the reporting weight, the number of spare reporting machines, the failure rate data of spare reporting machines, the average continuous reporting time of the reporting machines, the remaining resources of each department and the medical results data within the medical cycle, a modeling basis is provided for the subsequent establishment of the corresponding mapping model, and then a judgment model tool is provided for the subsequent determination of whether the currently set reporting weight data and spare reporting machine data are reasonable.

[0024] Preferably, the step S3 includes the following steps:

[0025] S31. According to the data type of the medical results within the medical cycle, the total medical treatment time data and the number of medical accident data corresponding to the historical medical cycles described in S21 are collected to obtain a historical total medical treatment time data set and a historical medical accident number data set;

[0026] S32. Based on the historical department resource surplus set, historical spare number reporting machine quantity set, historical spare number reporting machine failure rate data set, historical continuous reporting time data set, historical emergency priority data set, historical examination and return visit data set, historical multi-department collaboration data set, historical consultation cycle reporting weight data set, historical total consultation time data set, and historical consultation accident number set, construct mapping models between the department resource surplus, spare number reporting machine quantity, spare number reporting machine failure rate data, continuous reporting time data, emergency priority data, examination and return visit data, multi-department collaboration data, consultation cycle reporting weight data, and total consultation time data and the number of consultation accidents, respectively, to obtain a total consultation time mapping model and a consultation accident number mapping model;

[0027] By analyzing the correlation between emergency priority data and medical accidents, high-risk periods where delayed reporting leads to worsening of the condition can be identified, and nonlinear relationship modeling between failure rate and number of accidents can be achieved; by adopting the emergency start-up of the backup reporting machine, the medical risks caused by equipment failure are greatly reduced; by adding examination and follow-up data to the constructed mapping model, the average process time of high-frequency return departments such as the imaging department can be compressed to a reasonable time through subsequent dynamic adjustment of the reporting weight; and by adding multi-department collaboration data to the constructed mapping model, when the deviation between the consultation request time and the department queue status exceeds a certain degree, the comprehensive priority needs to be dynamically recalculated; the diagnosis and treatment preparation time can be shortened and the accident rate can be reduced.

[0028] Preferably, the total time consumption mapping model and the number of unexpected medical consultation mapping model in S32 respectively adopt an XGBoost-TCN cascade model and a graph neural network model;

[0029] The XGBoost-TCN cascade model combines the advantages of decision trees and time series convolution, significantly reducing the time-consuming prediction error for emergency priority data compared to traditional methods. At the same time, the GNN model mines the relationship between department nodes, resulting in a higher recall rate for accidental risk identification. The model's dual-threshold verification mechanism ensures that even when the equipment failure rate fluctuates greatly, it can still maintain a high level of prediction stability, achieving a shift from "post-failure response" to "pre-failure prevention."

[0030] Preferably, the S4 comprises the following steps:

[0031] S41. Based on the historical visit cycle reporting weight data set, collect the real-time reporting weight data corresponding to each type of emergency priority information, examination and return visit information, and multi-department collaboration information of the current hospital, the real-time remaining resources of each department, the real-time number of configured standby reporting machines, the failure rate data of the corresponding standby reporting machines, and the average continuous reporting time, to obtain the real-time visit cycle reporting weight data set, the real-time standby reporting machine number set, the real-time standby reporting machine failure rate data set, the real-time continuous reporting time data set, and the real-time department resource remaining amount set;

[0032] S42, combining and inputting each data item of the current emergency priority data set, the current checkup and return visit data set, the current multi-department collaboration data set, the real-time consultation cycle number reporting weight data set, the real-time standby number reporting machine quantity set, the real-time standby number reporting machine failure rate data set, the real-time continuous number reporting duration data set, and the real-time department resource remaining quantity data set into the total consultation time mapping model and the consultation accident number mapping model for mapping, thereby obtaining real-time consultation total time data and real-time consultation accident number data;

[0033] By mapping the collected data, we can obtain real-time data on the total time spent on medical treatment and the number of real-time medical accidents, that is, we can obtain data indicators to measure the rationality of the hospital's current reporting weight mechanism and the relevant parameters of the reporting machine, which provides a reference basis for whether to adjust the current reporting weight mechanism and the relevant parameters of the reporting machine in the future.

[0034] Preferably, the S5 comprises the following steps:

[0035] S51, setting a first threshold for the total time consumed during the current medical consultation and a threshold for the number of unexpected medical consultations;

[0036] S52, setting a maximum number of repetitions; when the real-time total time consumption data of the medical consultation is greater than or equal to the first current total time consumption threshold or the real-time number of medical consultation accidents is greater than or equal to the current number of medical consultation accidents threshold, the real-time medical consultation cycle reporting weight data set and the real-time continuous reporting duration data set are adjusted using a non-interface technology, and after the adjustment is completed, S42, S51 and S52 are repeated;

[0037] By first adjusting the real-time consultation cycle reporting weight data and the real-time continuous reporting time data, both types of data are digital setting data of the reporting machine, so the adjustment process is relatively quick and simple, thereby ensuring that the real-time total consultation time data will not fluctuate significantly due to the adjustment process.

[0038] Preferably, the S6 comprises the following steps:

[0039] S61, setting a second threshold for the total time consumed in the current medical consultation;

[0040] S62. When the number of repetitions in S52 is greater than or equal to the maximum number of repetitions and the real-time total time consumed for medical consultation is greater than or equal to the first current total time consumed for medical consultation threshold or the number of real-time medical consultation accidents is greater than or equal to the current threshold for the number of medical consultation accidents, the real-time medical consultation cycle reporting weight data set, the real-time continuous reporting duration data set and the real-time standby reporting machine quantity set are adjusted simultaneously using a non-interface technology. After the adjustment is completed, S42, S51, S52, S61 and S62 are repeated until the real-time total time consumed for medical consultation is less than the second current total time consumed for medical consultation threshold and the number of real-time medical consultation accidents is less than the current threshold for the number of medical consultation accidents;

[0041] When simply adjusting the digital setting information of the number reporting machine does not work, the physical information that requires actual operation, such as the number of spare number reporting machines, is adjusted to further strengthen the real-time total time data of medical visits and the number of real-time medical visit accidents, so as to ensure that the number of real-time medical visit accidents is controlled within the preset range; since the actual operation takes more time, a second current total time threshold for medical visits is set to limit the degree of time spent on actual operations.

[0042] Preferably, adjusting the real-time medical consultation cycle reporting weight dataset and the real-time continuous reporting duration dataset in S52 includes the following steps:

[0043] S521. Setting the weight parameter adjustment step corresponding to each type of patient information and the parameter adjustment step of the continuous reporting time data of each reporting machine to obtain a first parameter adjustment step set and a second parameter adjustment step set;

[0044] S522: Adjust the real-time medical consultation cycle reporting weight dataset and the real-time continuous reporting duration dataset according to the first parameter adjustment step set and the second parameter adjustment step set;

[0045] The simultaneous adjustment of the real-time medical treatment cycle reporting weight data set, the real-time continuous reporting duration data set, and the real-time standby reporting machine quantity set in S62 includes the following steps:

[0046] S621. Setting the weight parameter adjustment step corresponding to each type of patient information, the parameter adjustment step for the continuous reporting duration data of each number-reporting machine, and the parameter adjustment step for the number of real-time standby number-reporting machines, to obtain a third parameter adjustment step set, a fourth parameter adjustment step set, and a fifth parameter adjustment step set;

[0047] S622. Simultaneously adjust the real-time medical consultation cycle reporting weight data set, the real-time continuous reporting duration data set, and the real-time standby reporting machine quantity set according to the third parameter adjustment step set, the fourth parameter adjustment step set, and the fifth parameter adjustment step set.

[0048] The interface-free registration machine-based multi-scenario medical consultation adaptation registration control system includes a real-time patient queue information collection module, a current hospital historical medical consultation registration data collection module, a historical medical consultation index data collection module, a medical consultation association mapping model construction module, a real-time medical consultation data mapping module, a current medical consultation index primary adjustment module, and a current medical consultation index secondary adjustment module.

[0049] The present invention also discloses a storage medium on which a program is stored. When the program is executed by a processor, the above method is implemented.

[0050] The present invention also discloses a multi-scenario medical consultation adaptation reporting control system based on an interface-free reporting machine, and the system is used to implement the above method.

[0051] The present invention also discloses a number reporting machine, in which the above-mentioned interface-free number reporting machine multi-scenario medical consultation adaptation number reporting control system is integrated; the number reporting machine can be fixedly installed on the seat body.

[0052] Preferably, the number reporting machine can be fixed on the armrest or backrest of a chair body placed in the hospital.

[0053] The present invention has the following beneficial effects:

[0054] 1. Through dual mapping based on real-time queue status and historical weights, the present invention can dynamically adjust department resource allocation priorities, reduce waiting times for emergency patients, and balance the needs of follow-up patients. Through the threshold trigger mechanism of the unexpected number of visits model, risks such as equipment failure and process congestion can be identified in advance, thereby improving the response speed of unexpected events. The secondary adjustment mechanism ensures that the operability of adjustments is from easy to difficult, thereby minimizing the time-consuming impact of adjustment operations.

[0055] 2. In this invention, the advantages of decision trees and time series convolution are integrated through the XGBoost-TCN cascade model, which greatly reduces the time-consuming prediction error of emergency priority data compared with traditional methods. At the same time, the GNN model mines the relationship between department nodes to achieve a higher recall rate for accident risk identification. The model's dual-threshold verification mechanism, training error threshold + test accuracy threshold, ensures that when the equipment failure rate fluctuates greatly, it can still maintain a high level of prediction stability, realizing the transition from "post-failure response" to "pre-failure prevention."

[0056] 3. In the present invention, a secondary adjustment mechanism is set up. When only adjusting the digital setting information of the number reporting machine does not work, the physical information that requires actual operation, such as the number of spare number reporting machines, is adjusted to further strengthen the real-time total time consumption data of medical consultations and the number of real-time medical consultation accidents, so as to ensure that the number of real-time medical consultation accidents is controlled within a preset range; because the actual operation takes more time, a second current total time consumption threshold is set to limit the degree of time consumed by the actual operation.

[0057] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0059] Figure 1 This is a schematic diagram of the overall flow of the interface-free medical consultation number control method of the present invention;

[0060] Figure 2 A schematic diagram of the process of constructing a total time consumption mapping model and a number of unexpected visits mapping model for the present invention;

[0061] Figure 3 Schematic diagram of the process of primary regulation and secondary regulation of the present invention;

[0062] Figure 4 This is a module schematic diagram of the interface-free number reporting machine multi-scenario medical consultation adaptation reporting control system of the present invention. DETAILED DESCRIPTION

[0063] The following will clearly and completely describe the technical solutions in the embodiments of the invention in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0064] Example 1

[0065] See also Figure 1-3 This embodiment is a method for controlling medical consultation number registration without an interface, and includes the following steps:

[0066] S1. Collect patient cohort status data corresponding to various medical scenarios;

[0067] Said S1 comprises the following steps:

[0068] S11. Set several types of patient visit scenarios to obtain a set of patient visit scenario types; the set of patient visit scenario types includes emergency priority (including cases requiring immediate treatment, such as traumatic bleeding, acute chest pain, and febrile convulsions), follow-up visit (after completing CT / B-ultrasound examinations, the patient needs to return to the doctor for report interpretation), and multi-department collaboration (requiring cross-department consultation, such as MDT diagnosis and treatment of tumors);

[0069] Then, the patient information types corresponding to the emergency priority type, examination and return type, and multi-department collaboration type are set respectively to obtain the emergency priority type information type set, the examination and return type information type set, and the multi-department collaboration type information type set; the emergency priority type information type set includes patient ID, current queue position, emergency level (I-IV), waiting time (minutes), vital sign status, and latest examination results; the examination and return type information type set includes patient ID, original clinic, examination items, examination completion time, report status, and return queue weight; the multi-department collaboration type information type set includes patient ID, main clinic department, collaborative department, consultation request time, queue status of each department, and comprehensive priority;

[0070] S12. Based on the emergency priority information type set, the examination and return visit information type set, and the multi-department collaboration information type set, the queue status of each patient in the HIS or LIS system is obtained in real time to obtain the current emergency priority data set, the current examination and return visit data set, and the current multi-department collaboration data set; wherein, the queue status of each patient in the HIS or LIS system can be obtained in real time by identifying screen information with optical character recognition (OCR) or monitoring serial port data streams;

[0071] For example, the emergency priority dataset, the checkup follow-up dataset, and the multi-department collaboration dataset are as follows:

[0072]

[0073] By dynamically linking emergency department classification (I-IV) with vital signs, an automatic queue-jumping mechanism is implemented for critically ill patients, reducing the average waiting time for Level I patients to less than 3 minutes, improving response efficiency by 80% compared to traditional models. The system automatically pushes the latest examination results to the emergency terminal, reducing the time spent on manual review. The return consultation queue weighting algorithm comprehensively considers parameters such as report review status and original waiting time to calculate the optimal time for return consultation, reducing patients' secondary waiting time and avoiding ineffective waiting due to incomplete reports. The original clinic status synchronization function ensures that doctors are aware of the patient's examination progress in real time. Combined with clinical parameters (tumor TNM staging), dynamic priority is generated, significantly reducing consultation preparation time and improving the on-time consultation rate.

[0074] S2. Collect historical patient queue status data, historical reporting weights, and equipment data corresponding to the various medical scenarios described in S1 in the current hospital;

[0075] The S2 comprises the following steps:

[0076] S21. Based on the emergency priority information type set, the examination and return visit information type set, and the multi-department collaboration information type set, queue status data of multiple patients in the current hospital over several historical consultation cycles are collected to obtain a historical emergency priority data set, a historical examination and return visit data set, and a historical multi-department collaboration data set;

[0077] Then, the reporting weight data corresponding to each type of emergency priority information, examination and return visit information, and multi-department collaboration information when the hospital's reporting machine reports the number in each of the several medical cycles are collected to obtain a historical medical cycle reporting weight data set;

[0078] S22. Collect the remaining resources of each department of the current hospital at the start time of the plurality of consultation cycles to obtain a historical department resource remaining amount set; set the data type of the consultation result within the consultation cycle, including the total consultation time data and the number of patient accidents; for example, the patient accidents include situations where the patient's condition worsens within the hospital due to unreasonable weight distribution of the hospital's number reporting machine or untimely reporting;

[0079] Then, the number of standby number-calling machines configured in the current hospital corresponding to the plurality of medical treatment cycles, as well as the historical failure rate data and the average continuous number-calling time of the corresponding standby number-calling machines are collected to obtain a historical standby number-calling machine number set, a historical standby number-calling machine failure rate data set, and a historical continuous number-calling time data set;

[0080] Data analysis based on historical emergency data sets can identify the average retention patterns of different types of patients, improving the treatment efficiency of Class I patients by more than 35%. By setting the reporting weight, the number of standby reporting machines, the failure rate of standby reporting machines, the average length of time for consecutive reporting by reporting machines, the remaining resources of each department, and the medical results data within the medical cycle, a modeling basis is provided for the subsequent establishment of the corresponding mapping model, thereby providing a judgment model tool for whether the currently set reporting weight data and standby reporting machine data are reasonable.

[0081] S3: Collect the total time consumption data and the number of unexpected patient visits corresponding to several consultation cycles, and record them as historical consultation indicator data; and build a total time consumption mapping model and a number of unexpected patient visits mapping model by combining the historical patient queue status data, historical report number weights, and equipment data in S2;

[0082] The S3 includes the following steps:

[0083] S31. According to the data type of the medical results within the medical cycle, the total medical treatment time data and the number of medical accident data corresponding to the historical medical cycles described in S21 are collected to obtain a historical total medical treatment time data set and a historical medical accident number data set;

[0084] S32. Based on the historical department resource surplus set, historical spare number reporting machine quantity set, historical spare number reporting machine failure rate data set, historical continuous reporting time data set, historical emergency priority data set, historical examination and return visit data set, historical multi-department collaboration data set, historical consultation cycle reporting weight data set, historical total consultation time data set, and historical consultation accident number set, construct mapping models between the department resource surplus, spare number reporting machine quantity, spare number reporting machine failure rate data, continuous reporting time data, emergency priority data, examination and return visit data, multi-department collaboration data, consultation cycle reporting weight data, and total consultation time data and the number of consultation accidents, respectively, to obtain a total consultation time mapping model and a consultation accident number mapping model;

[0085] The construction of the total time consumption mapping model and the number of unexpected medical visits mapping model in S32 includes the following steps:

[0086] S321. Construct an initial XGBoost-TCN cascade model and an initial graph neural network model; set a first training data ratio and a second training data ratio, and use the first training data ratio and the second training data ratio to divide the historical department resource remaining amount set, the historical standby reporting machine quantity set, the historical standby reporting machine failure rate data set, the historical continuous reporting time data set, the historical emergency priority data set, the historical examination return type data set, the historical multi-department collaboration data set, the historical medical treatment cycle reporting weight data set, the historical medical treatment total time data set, and the historical medical treatment accident number set to obtain a first training data set, a first test data set, a second training data set, and a second test data set;

[0087] S322. Set a first training error threshold and a second training error threshold; use the first training data set to train the initial XGBoost-TCN cascade model. During the training process, when the training error is less than the first training error threshold, stop training to obtain a trained XGBoost-TCN cascade model; otherwise, continue training until the training error is less than the first training error threshold; then use the second training data set to train the initial graph neural network model. During the training process, when the training error is less than the second training error threshold, stop training to obtain a trained graph neural network model; otherwise, continue training until the training error is less than the second training error threshold;

[0088] S323. Set a first test accuracy threshold and a second test accuracy threshold; use the first test data set to test the trained XGBoost-TCN cascade model. After the test is completed, obtain the first test accuracy data; when the first test accuracy data is greater than or equal to the first test accuracy threshold, use the trained XGBoost-TCN cascade model as the total time-consuming medical consultation mapping model; otherwise, continue to train the trained XGBoost-TCN cascade model;

[0089] The trained graph neural network model is then tested using the second test dataset. After the test is completed, second test accuracy data is obtained. When the second test accuracy data is greater than or equal to a second test accuracy threshold, the trained graph neural network model is used as a medical accident number mapping model. Otherwise, the trained graph neural network model continues to be trained.

[0090] The initial XGBoost-TCN cascade model includes:

[0091] Front-end XGBoost module: 1. Input layer: structured features (emergency priority, department resources, etc.); 2. Number of decision trees: 100-200 (controlled by early stopping); 3. Output: process type classification (emergency / inspection / collaboration) and basic time-consuming prediction;

[0092] Backend TCN module: Contains dilated convolutional layers to capture long-term dependencies, batch normalization layers to accelerate convergence, and finally a fully connected output.

[0093] The initial graph neural network model includes:

[0094] TCN module: 1. Dilated convolution: dilation = 1 / 2 / 4 incrementally captures multi-scale temporal patterns; 2. Activation function: ReLU to avoid gradient vanishing; 3. Pooling layer: max pooling to compress feature dimensions;

[0095] GNN module: 1. Graph convolution: GCNConv basic feature propagation; Graph attention: GATConv modeling department resource competition; Output layer: sigmoid activation adaptation to binary classification tasks;

[0096] The XGBoost-TCN cascade model combines the advantages of decision trees and time series convolution, significantly reducing the time-consuming prediction error for emergency-priority data compared to traditional methods. Furthermore, the GNN model mines departmental node relationships to achieve a high recall rate for unexpected risk identification. The model's dual-threshold verification mechanism (training error threshold + test accuracy threshold) ensures high prediction stability even when equipment failure rates fluctuate significantly, enabling a shift from "post-failure response" to "pre-failure prevention."

[0097] By analyzing the correlation between emergency priority data and medical accidents, we can identify high-risk periods where delays in reporting numbers lead to worsening of the condition (for example, the accident rate increases sharply by 47% when the waiting time for a Level I emergency room exceeds 5 minutes), and achieve nonlinear relationship modeling between the failure rate and the number of accidents; by using the emergency start-up of the backup reporting machine, the medical risks caused by equipment failure are greatly reduced; by adding examination and return data to the constructed mapping model, the average process time of high-frequency return departments such as the imaging department can be compressed to a reasonable time through subsequent dynamic adjustment of the reporting weight; and by adding multi-department collaboration data to the constructed mapping model, when the deviation between the consultation request time and the department queue status exceeds 30%, the comprehensive priority needs to be dynamically recalculated; this can shorten the MDT diagnosis and treatment preparation time by 58% and reduce the accident rate to below 3‰;

[0098] S4, mapping the collected historical hospital number weights, equipment data, departmental remaining resource data, and patient queue status data collected in S1 into the total consultation time mapping model and the number of consultation accidents mapping model respectively;

[0099] The S4 comprises the following steps:

[0100] S41. Based on the historical visit cycle reporting weight data set, collect the real-time reporting weight data corresponding to each type of emergency priority information, examination and return visit information, and multi-department collaboration information of the current hospital, the real-time remaining resources of each department, the real-time number of configured standby reporting machines, the failure rate data of the corresponding standby reporting machines, and the average continuous reporting time, to obtain the real-time visit cycle reporting weight data set, the real-time standby reporting machine number set, the real-time standby reporting machine failure rate data set, the real-time continuous reporting time data set, and the real-time department resource remaining amount set;

[0101] S42, combining and inputting each data item of the current emergency priority data set, the current checkup and return visit data set, the current multi-department collaboration data set, the real-time consultation cycle number reporting weight data set, the real-time standby number reporting machine quantity set, the real-time standby number reporting machine failure rate data set, the real-time continuous number reporting duration data set, and the real-time department resource remaining quantity data set into the total consultation time mapping model and the consultation accident number mapping model for mapping, thereby obtaining real-time consultation total time data and real-time consultation accident number data;

[0102] By mapping the collected data, we can obtain real-time data on the total time spent on medical consultations and the number of real-time medical accidents. In other words, we can obtain data indicators to measure the rationality of the hospital's current reporting weight mechanism and related parameters of the reporting machine, which provides a reference for whether to adjust the current reporting weight mechanism and related parameters of the reporting machine in the future.

[0103] S5. Comparing the mapping result in S4 with the corresponding preset threshold, adjusting the patient information weight and the continuous reporting time according to the comparison result;

[0104] The S5 comprises the following steps:

[0105] S51, setting a first threshold for the total time consumed during the current medical consultation and a threshold for the number of unexpected medical consultations;

[0106] S52, setting a maximum number of repetitions; when the real-time total time consumption data of the medical consultation is greater than or equal to the first current total time consumption threshold or the real-time number of medical consultation accidents is greater than or equal to the current number of medical consultation accidents threshold, the real-time medical consultation cycle reporting weight data set and the real-time continuous reporting duration data set are adjusted using a non-interface technology, and after the adjustment is completed, S42, S51 and S52 are repeated;

[0107] Adjusting the real-time medical consultation cycle reporting weight dataset and the real-time continuous reporting duration dataset in S52 includes the following steps:

[0108] S521. Setting the weight parameter adjustment step corresponding to each type of patient information and the parameter adjustment step of the continuous reporting time data of each reporting machine to obtain a first parameter adjustment step set and a second parameter adjustment step set;

[0109] S522: Adjust the real-time medical consultation cycle weight dataset and the real-time continuous medical consultation duration dataset according to the first parameter adjustment step set and the second parameter adjustment step set; the adjustment formulas are as follows:

[0110] ;

[0111] ;

[0112] Where, 、 They represent the first j The corresponding type of patient information i +1 round and i The weight data of the wheel; 、 Respectively represent j The corresponding number reporting machine i +1 round and i The continuous reporting time data of the wheel; 、 They represent the first j The first parameter adjustment step and the second parameter adjustment step of the weight data corresponding to the type of patient information jThe first parameter adjustment step length of the continuous reporting time data corresponding to each reporting machine is the first parameter adjustment step length set and the second parameter adjustment step length set. j Parameter adjustment step size; 、 、 、 Respectively represent the real-time total time consumption data of medical consultation, the first current total time consumption threshold of medical consultation, the real-time number of medical consultation accidents and the current threshold of the number of medical consultation accidents;

[0113] Exemplary:

[0114] First, the total time threshold for the current visit is 180 minutes; the threshold for the number of accidents during the current visit is 5 times / hour; the maximum number of repeated adjustments is 30 times;

[0115] The first parameter adjustment step set (patient type weight):

[0116] {

[0117] "Emergency priority": {

[0118] "Emergency Classification": {"Step Size": 0.25, "Direction": "Bidirectional"},

[0119] "Waiting time": {"Step size": 0.01, "Direction": "Forward", "Upper limit": 0.2},

[0120] "vital signs": {"step length": 0.15, "direction": "bidirectional"},

[0121] "Check result": {"Step size": 0.1, "Direction": "Bidirectional"}

[0122] },

[0123] "Check-up type": {

[0124] "Check Item": {"Step Size": 0.12, "Direction": "Bidirectional"},

[0125] "Report Status": {"Step Size": 0.08, "Direction": "Forward"},

[0126] "Return queue weight": {"Step size": 0.05, "Direction": "Bidirectional"}

[0127] },

[0128] "Multi-department collaboration": {

[0129] "Consultation Time": {"Step": 0.03, "Direction": "Forward", "Upper Limit": 0.15},

[0130] "Comprehensive priority": {"step size": 0.2, "direction": "bidirectional"},

[0131] "Department Load": {"Step Size": 0.1, "Direction": "Bidirectional"}

[0132] }

[0133] };

[0134] The second parameter adjustment step set (number reporting machine duration):

[0135] {

[0136] "First number reporting machine": ±30 seconds,

[0137] "Second number reporting machine": ±20 seconds,

[0138] "Third number reporting machine": ±10 seconds

[0139] }

[0140] Execute when real-time data triggers a threshold (e.g., total time consumption reaches 180 minutes or the number of accidents reaches 5 times per hour);

[0141] After adjustment, re-monitor the data and repeat the adjustment if it still exceeds the threshold;

[0142] Effect Verification: After implementation, the average consultation time dropped to 160 minutes, and the number of accidents dropped to 3 per hour;

[0143] By first adjusting the real-time consultation cycle reporting weight data and the real-time continuous reporting duration data, both of which are digital setting data of the reporting machine, the adjustment process is relatively quick and simple, thereby ensuring that the real-time total consultation time data will not fluctuate significantly due to the adjustment process;

[0144] S6. Based on the result of the first adjustment in S5, a second adjustment is made to the patient information weight, the continuous reporting time, and the number of spare reporting machines;

[0145] The S6 comprises the following steps:

[0146] S61, setting a second threshold for the total time consumed in the current medical consultation;

[0147] S62. When the number of repetitions in S52 is greater than or equal to the maximum number of repetitions and the real-time total time consumed for medical consultation is greater than or equal to the first current total time consumed for medical consultation threshold or the number of real-time medical consultation accidents is greater than or equal to the current threshold for the number of medical consultation accidents, the real-time medical consultation cycle reporting weight data set, the real-time continuous reporting duration data set and the real-time standby reporting machine quantity set are adjusted simultaneously using a non-interface technology. After the adjustment is completed, S42, S51, S52, S61 and S62 are repeated until the real-time total time consumed for medical consultation is less than the second current total time consumed for medical consultation threshold and the number of real-time medical consultation accidents is less than the current threshold for the number of medical consultation accidents;

[0148] The simultaneous adjustment of the real-time medical treatment cycle reporting weight data set, the real-time continuous reporting duration data set, and the real-time standby reporting machine quantity set in S62 includes the following steps:

[0149] S621. Setting the weight parameter adjustment step corresponding to each type of patient information, the parameter adjustment step for the continuous reporting duration data of each number-reporting machine, and the parameter adjustment step for the number of real-time standby number-reporting machines, to obtain a third parameter adjustment step set, a fourth parameter adjustment step set, and a fifth parameter adjustment step set;

[0150] S622, according to the third parameter adjustment step set, the fourth parameter adjustment step set and the fifth parameter adjustment step set, the real-time medical cycle number reporting weight data set, the real-time continuous number reporting time data set and the real-time standby number reporting machine quantity set are adjusted simultaneously; the adjustment formulas are as follows:

[0151] ;

[0152] ;

[0153] ;

[0154] Where, 、 Respectively represent i +1 round and i The number of spare number reporting machines for the ship; 、 、 They represent the first j The second parameter adjustment step length of the weight data corresponding to the type of patient information, the first parameter adjustment step length j The second parameter adjustment step length of the continuous reporting time data corresponding to the number reporting machine and the parameter adjustment step length of the number of spare number reporting machines, namely, the third parameter adjustment step length set, the fourth parameter adjustment step length set and the fifth parameter adjustment step length set. j Parameter adjustment step size; Indicates the total time threshold of the second current visit;

[0155] When merely adjusting the digital setting information of the number-reporting machine does not produce any effect, the number of spare number-reporting machines and other physical information that requires actual operation are adjusted to further and more intensively adjust the real-time total time consumption data and the number of real-time unexpected medical visits, so as to ensure that the number of real-time unexpected medical visits is controlled within a preset range; because the actual operation requires more time, a second current total time consumption threshold is set to limit the time consumption of the actual operation;

[0156] Based on the dual mapping of real-time queue status and historical weights, department resource allocation priorities can be dynamically adjusted to reduce waiting times for emergency patients while balancing the needs of follow-up patients. Through the threshold trigger mechanism of the unexpected number of visits model, risks such as equipment failure and process congestion can be identified in advance, thereby improving the response speed of unexpected events. The secondary adjustment mechanism ensures that the operability of adjustments is from easy to difficult, thereby minimizing the time-consuming impact of adjustment operations.

[0157] Example 2

[0158] See also Figure 4 This embodiment discloses a multi-scenario medical consultation adaptation and reporting control system based on an interface-free reporting machine. The system can implement the method of the above embodiment, including a real-time patient queue information collection module, a current hospital historical medical consultation reporting data collection module, a historical medical consultation index data collection module, a medical consultation association mapping model construction module, a real-time medical consultation data mapping module, a current medical consultation index primary adjustment module, and a current medical consultation index secondary adjustment module;

[0159] The real-time patient queue information collection module collects patient queue status data corresponding to various medical treatment scenarios;

[0160] The current hospital's historical medical report number data collection module collects historical patient queue status data, historical report number weights, and device data corresponding to the current hospital's various medical scenarios in S1;

[0161] The historical medical treatment index data collection module collects the total medical treatment time data and the number of accidental cases of the patients corresponding to a number of medical treatment cycles, and records them as historical medical treatment index data;

[0162] The medical consultation association mapping model construction module uses the historical patient queue status data, historical report number weights and equipment data in S2 and the historical medical consultation index data to construct a total medical consultation time mapping model and a medical consultation accident number mapping model;

[0163] The real-time medical data mapping module maps the collected historical hospital number weights, equipment data, departmental remaining resource data, and patient queue status data collected by S1 into the total medical time mapping model and the number of medical accident mapping model, respectively, to obtain real-time total medical time data and real-time number of medical accidents;

[0164] The current medical index adjustment module compares the real-time total medical treatment time data and the real-time number of medical treatment accidents with the corresponding thresholds, and adjusts the patient information weight and the continuous reporting time according to the comparison results;

[0165] The current medical treatment index secondary adjustment module performs secondary adjustment on the patient information weight, the continuous reporting time and the number of spare reporting machines according to the result of the primary adjustment in S5.

[0166] Example 3

[0167] This embodiment discloses a storage medium having a program stored thereon. When the program is executed by a processor, the method of the above embodiment is implemented.

[0168] Example 4

[0169] This embodiment discloses a number-reporting machine, which integrates the interface-free, multi-scenario medical consultation adaptation number-reporting control system of the above embodiment; the number-reporting machine can be fixedly mounted on the seat body;

[0170] Furthermore, the number reporting machine can be fixed on the armrest or backrest of a chair body placed in the hospital.

[0171] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0172] The preferred embodiments of the invention disclosed above are intended only to help illustrate the invention. These preferred embodiments do not exhaust all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A method for controlling medical consultation number registration without an interface, characterized in that: The following steps are involved: S1. Collect patient cohort status data corresponding to various medical scenarios; S2. Collect historical patient queue status data, historical reporting weights, and equipment data corresponding to the various medical scenarios described in S1 in the current hospital; S3: Collect the total time consumption data and the number of unexpected patient visits corresponding to several consultation cycles, and record them as historical consultation indicator data; and build a total time consumption mapping model and a number of unexpected patient visits mapping model by combining the historical patient queue status data, historical report number weights, and equipment data in S2; S4, mapping the collected historical hospital number weights, equipment data, departmental remaining resource data, and patient queue status data collected in S1 into the total consultation time mapping model and the number of consultation accidents mapping model respectively; S5. Comparing the mapping result in S4 with the corresponding preset threshold, adjusting the patient information weight and the continuous reporting time according to the comparison result; S6. Based on the result of the first adjustment in S5, a second adjustment is made to the patient information weight, the continuous reporting time, and the number of spare reporting machines.

2. The interface-free medical consultation number control method according to claim 1, characterized in that: Said S1 comprises the following steps: S11. Setting several types of patient visit scenarios to obtain a patient visit scenario type set; the patient visit scenario type set includes an emergency priority type, a checkup return type, and a multi-department collaboration type; Then, respectively set the patient information types corresponding to the emergency priority type, the check-up return type, and the multi-department collaboration type to obtain an emergency priority type information type set, a check-up return type information type set, and a multi-department collaboration type information type set; S12. According to the emergency priority information type set, the examination and return visit information type set, and the multi-department collaboration information type set, the queue status of each patient in the HIS or LIS system is obtained in real time to obtain the current emergency priority data set, the current examination and return visit data set, and the current multi-department collaboration data set.

3. The interface-free medical consultation number control method according to claim 2, characterized in that: The S2 comprises the following steps: S21. Based on the emergency priority information type set, the examination and return visit information type set, and the multi-department collaboration information type set, queue status data of multiple patients in the current hospital over several historical consultation cycles are collected to obtain a historical emergency priority data set, a historical examination and return visit data set, and a historical multi-department collaboration data set; Then, the reporting weight data corresponding to each type of emergency priority information, examination and return visit information, and multi-department collaboration information when the hospital's reporting machine reports the number in each of the several medical cycles are collected to obtain a historical medical cycle reporting weight data set; S22. Collect the remaining resources of each department of the current hospital at the beginning of the plurality of consultation cycles to obtain a historical department resource remaining amount set; set the data type of the consultation result within the consultation cycle, including the total consultation time data and the number of accidents of the patients; Then collect the number of spare number reporting machines configured in the current hospital corresponding to the several medical cycles, the historical failure rate data of the corresponding spare number reporting machines, and the average continuous reporting time, to obtain the historical spare number reporting machine number set, the historical spare number reporting machine failure rate data set, and the historical continuous reporting time data set.

4. The interface-free medical consultation number control method according to claim 3, characterized in that: The S3 includes the following steps: S31. According to the data type of the medical results within the medical cycle, the total medical treatment time data and the number of medical accident data corresponding to the historical medical cycles described in S21 are collected to obtain a historical total medical treatment time data set and a historical medical accident number data set; S32. Based on the historical department resource surplus set, historical spare number reporting machine quantity set, historical spare number reporting machine failure rate data set, historical continuous number reporting time data set, historical emergency priority data set, historical examination and follow-up data set, historical multi-department collaboration data set, historical consultation cycle number reporting weight data set, historical total consultation time data set and historical consultation accident number set, construct mapping models between the department resource surplus, spare number reporting machine quantity, spare number reporting machine failure rate data, continuous number reporting time data, emergency priority data, examination and follow-up data, multi-department collaboration data, consultation cycle number reporting weight data and the total consultation time data and the number of consultation accidents, respectively, to obtain the total consultation time mapping model and the consultation accident number mapping model.

5. The interface-free medical consultation number control method according to claim 4, characterized in that: The S4 comprises the following steps: S41. Based on the historical visit cycle reporting weight data set, collect the real-time reporting weight data corresponding to each type of emergency priority information, examination and return visit information, and multi-department collaboration information of the current hospital, the real-time remaining resources of each department, the real-time number of configured standby reporting machines, the failure rate data of the corresponding standby reporting machines, and the average continuous reporting time, to obtain the real-time visit cycle reporting weight data set, the real-time standby reporting machine number set, the real-time standby reporting machine failure rate data set, the real-time continuous reporting time data set, and the real-time department resource remaining amount set; S42. Each piece of data in the current emergency priority data set, the current check-up follow-up data set, the current multi-department collaboration data set, the real-time consultation cycle number reporting weight data set, the real-time standby number reporting machine quantity set, the real-time standby number reporting machine failure rate data set, the real-time continuous number reporting time data set, and the real-time department resource remaining quantity set are combined and input into the total consultation time mapping model and the consultation accident number mapping model for mapping, so as to obtain the real-time total consultation time data and the real-time consultation accident number.

6. The interface-free medical consultation number control method according to claim 5, characterized in that: The S5 comprises the following steps: S51, setting a first threshold for the total time consumed during the current medical consultation and a threshold for the number of unexpected medical consultations; S52, setting a maximum number of repetitions; when the real-time total time consumption data of the medical consultation is greater than or equal to the first current total time consumption threshold or the real-time number of medical consultation accidents is greater than or equal to the current number of medical consultation accidents threshold, the real-time medical consultation cycle reporting weight data set and the real-time continuous reporting duration data set are adjusted using a non-interface technology, and after the adjustment is completed, S42, S51 and S52 are repeated; Adjusting the real-time medical consultation cycle reporting weight dataset and the real-time continuous reporting duration dataset in S52 includes the following steps: S521. Setting the weight parameter adjustment step corresponding to each type of patient information and the parameter adjustment step of the continuous reporting time data of each reporting machine to obtain a first parameter adjustment step set and a second parameter adjustment step set; S522: Adjust the real-time medical consultation cycle reporting weight data set and the real-time continuous reporting duration data set according to the first parameter adjustment step set and the second parameter adjustment step set.

7. The interface-free medical consultation number control method according to claim 6, characterized in that: The S6 comprises the following steps: S61, setting a second threshold for the total time consumed in the current medical consultation; S62. When the number of repetitions in S52 is greater than or equal to the maximum number of repetitions and the real-time total time consumed data for medical consultation is greater than or equal to the first current total time consumed threshold for medical consultation or the number of real-time medical consultation accidents is greater than or equal to the current number threshold for medical consultation accidents, the real-time medical consultation cycle reporting weight data set, the real-time continuous reporting time data set and the real-time standby reporting machine quantity set are adjusted simultaneously using interface-free technology. After the adjustment is completed, S42, S51, S52, S61 and S62 are repeated until the real-time total time consumed data for medical consultation is less than the second current total time consumed threshold for medical consultation and the number of real-time medical consultation accidents is less than the current number threshold for medical consultation accidents.

8. A storage medium, characterized in that: A program is stored thereon, and when the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

9. An interface-free number reporting machine with multi-scenario medical consultation adaptation and reporting control system, characterized by: The system is used to implement the method according to any one of claims 1 to 7.

10. A number reporting machine, characterized in that: A system as claimed in claim 9 is integrated.

Images