Nursing operation tracking and feedback system

Abstract

The invention relates to the technical field of nursing data processing, in particular to a nursing operation tracking and feedback system, which comprises a nursing process sequence verification module, a node sequence setting module, a node sequence setting module, a node sequence setting module, a node sequence setting module, a node sequence setting module and a node sequence setting module, and comparing the state sequence of the current operation state and the standard state, judging the missing intermediate state in the path and whether the path does not reach the termination node, and generating a nursing process sequence verification result. According to the method and the device, the compliance and the efficiency of the operation process are effectively improved through multi-level accurate evaluation of the operation sequence, the execution time and the potential step omission of the nursing process. According to the scheme, a nursing operation process is divided into a state node sequence, and a current operation state is compared with a standard process, so that operation jump and missing nodes which do not conform to specifications can be identified in real time. The detailed verification ensures the operation integrity and the process continuity, and effectively avoids step omission.

Description

Technical Field

[0001] The present invention relates to the technical field of nursing data processing, and particularly to a nursing operation tracking and feedback system. Background Art

[0002] Nursing data processing belongs to a branch of medical nursing information technology, and involves the collection, storage, analysis, and application of various types of data generated during the nursing process. Through information technology means, various data such as nursing operations, patient monitoring, and treatment records are structured and standardized to ensure data accuracy, real-time nature, and traceability. It not only provides accurate patient information support for nursing staff to help them optimize nursing plans, but also provides operational insights for hospital managers to improve medical quality and patient satisfaction.

[0003] Among them, the nursing operation tracking and feedback system is a system that monitors, records, and gives feedback on the nursing operation process in real time. By tracking the operation details of nursing staff, it timely gives feedback on whether it conforms to the standardized process, and helps nursing staff optimize operations when necessary. Its purpose is to improve the standardization of nursing operations, reduce human errors, and at the same time, through data accumulation and analysis, achieve continuous improvement of nursing quality.

[0004] In the prior art, in terms of the sequence specification and detail management of nursing operations, since it can only simply record the process, it is difficult to effectively identify jumps or omissions in complex processes. Especially in scenarios with many nursing steps and long processes, it is easy to miss key nursing links due to improper operation sequences, affecting the overall quality. For example, nursing staff may overlook specific links when performing multi-step handovers, resulting in insufficiently tight nursing connections. The monitoring of execution time is also limited to general feedback and it is difficult to conduct detailed distribution analysis, resulting in unreasonable execution speeds for some operations, affecting the nursing effect. At the same time, it is also difficult to identify the relevance between nursing steps, increasing the risk of key operations in the process being overlooked. This affects the comprehensiveness and accuracy of nursing quality and reduces the safety and consistency of the process. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose a nursing operation tracking and feedback system.

[0006] To achieve the above purpose, the present invention adopts the following technical solution: The nursing operation tracking and feedback system includes:

[0007] The nursing process sequence verification module collects the data of nursing records, initializes the finite state machine model, sets the state sequence through node sequences, inputs the state sequence into the finite state machine model, compares the current operation state with the state sequence of the standard state, judges the missing intermediate states in the path, and whether the path reaches the termination node, and generates the nursing process sequence verification result;

[0008] Based on the nursing process sequence verification result, the execution time evaluation module fits the state sequence into a Gaussian distribution curve, calculates the deviation value of the nursing time for each step according to the data shown in the Gaussian distribution curve, compares the current operation duration with the standard time distribution, determines whether the execution time exceeds the set reasonable range, and generates a nursing operation time deviation evaluation result;

[0009] Based on the nursing operation time deviation evaluation result, the nursing step omission prediction module defines the association probability between each nursing operation step and its subsequent step, constructs the step sequence in the nursing operation record with reference to the historically saved nursing operation standard probability chain, calculates the conditional probability between the current operation and the standard probability chain, predicts the possible subsequent step missing situation, and generates a nursing step omission risk prediction result;

[0010] Based on the nursing step omission risk prediction result, the high-frequency deviation mode optimization module identifies the combination of error steps or step sequences that frequently appear in the nursing operation, marks them as deviation labels, and generates a feedback optimization result of the nursing operation data.

[0011] As a further solution of the present invention, the obtaining step of inputting the state sequence into the finite state machine model is specifically as follows:

[0012] Based on the collection of nursing record data, including recording the time, frequency, and handover situation of each nursing operation step, each nursing operation process is divided into inconsistent state nodes to generate an initialized finite state machine model;

[0013] By setting the order of each node to form a state sequence, the state sequence is input into the initialized finite state machine model, and the operation state monitoring is carried out in sequence according to the node order to obtain the standard state sequence of the nursing operation.

[0014] As a further solution of the present invention, the obtaining step of the nursing process sequence verification result is specifically as follows:

[0015] Based on the standard state sequence of the nursing operation, the actual operation state is compared with the preset standard state sequence. When it is detected that the actual operation is not executed in sequence, the jump point is marked and the deviated node is recorded to generate an operation deviation record;

[0016] Based on the operation deviation record, the nodes in the nursing operation path are checked, the missing intermediate states are identified, it is judged whether the operation reaches the predetermined termination node, and the missing situation is recorded to generate a nursing process sequence verification result.

[0017] As a further solution of the present invention, the obtaining step of calculating the deviation value of the nursing time for each step is specifically as follows:

[0018] Based on the status sequence of the nursing processes in the verification result of the nursing process sequence, perform distribution fitting on the sequence to reflect the time distribution law of nursing operations under normal conditions, and obtain a standard time distribution curve;

[0019] According to the data of the standard time distribution curve, use the formula:

[0020]

[0021] Calculate the nursing time deviation value D of the i-th step i to obtain the deviation information of the execution time;

[0022] where μ i is the mean value of the standard nursing time of the i-th step, and σ i is the standard deviation of the standard nursing time of the i-th step.

[0023] As a further solution of the present invention, the obtaining step of the nursing operation time deviation evaluation result is specifically as follows:

[0024] Based on the deviation information of the execution time, compare the nursing time deviation value of each nursing step with the set reasonable range, analyze whether the execution time of each nursing step is within the reasonable range, and obtain the execution rationality judgment result of the nursing step;

[0025] Based on the execution rationality judgment result of the nursing step, by comparing the nursing time deviation values, mark the execution speed of each nursing step as abnormal or normal, and generate the nursing operation time deviation evaluation result.

[0026] As a further solution of the present invention, the obtaining step of the step sequence in the nursing operation record is specifically as follows:

[0027] According to the nursing operation time deviation evaluation result, use the formula:

[0028]

[0029] Calculate the association probability S i between step S i+1 and the subsequent step S i,i+1 to obtain the definition result of the association probability of the nursing step;

[0030] where N i,i+1 represents the number of times that step i immediately follows step i + 1, and N represents the total number of nursing operation records;

[0031] According to the result defined by the correlation probability of the nursing steps, referring to the standard probability chain of nursing operations saved historically, compare and analyze the matching degree between the current operation sequence and the standard probability chain, and compare the correlation of each step in turn to generate a standard step sequence of nursing operations.

[0032] As a further solution of the present invention, the specific steps for obtaining the risk prediction result of nursing step omission are as follows:

[0033] Based on the standard step sequence of nursing operations, use the formula:

[0034]

[0035] Calculate the conditional probability ratio C of the current nursing operation sequence relative to the standard probability chain to obtain the matching degree data of the current sequence;

[0036] Among them, n represents the total number of steps of nursing operations, P(A i |A i-1 ) represents the conditional probability that step A i occurs after step A i-1 in the current nursing record, and P std (A i |A i-1 ) represents the conditional probability that step A i occurs after step A i-1 in the standard nursing process;

[0037] According to the matching degree data of the current sequence, compare the matching degree of the current operation sequence with a preset threshold, analyze the occurrence of key steps in the nursing process, identify the steps with high conditional probability in the standard process and not appearing in the actual operation record, and generate a risk prediction result for nursing step omission.

[0038] As a further solution of the present invention, the specific steps for obtaining the feedback optimization result of nursing operation data are as follows:

[0039] Based on the risk prediction result of nursing step omission, combine all nursing operations in the order of the standard process, compare item by item and arrange the sequence, and record the execution status of each step to generate sequence data of nursing operations;

[0040] Based on the sequence data of nursing operations, judge the combination of error steps that repeatedly appear in the nursing record, mark the combination of error steps and classify them as deviation labels to generate a feedback optimization result of nursing operation data.

[0041] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0042] In the present invention, through multi-level precise assessment of the operation sequence, execution time, and potential step omissions in the nursing process, the compliance and efficiency of the operation process are effectively improved. By dividing the nursing operation process into a sequence of state nodes and comparing the current operation state with the standard process, the scheme can identify non-compliant operation jumps and missing nodes in real time. This refined verification ensures the integrity of the operation and the coherence of the process, effectively avoiding missed inspections of steps. By fitting the time distribution of each nursing step with the Gaussian distribution model, and evaluating the rationality of the execution time, abnormal situations of too fast or overtime operations can be identified in a timely manner, which helps to prevent nursing mistakes caused by abnormal execution speeds. In addition, in the nursing process, by analyzing the correlation probability between steps, potential subsequent steps that may be omitted can be pre-warned, providing forward-looking guarantee for nursing operations. This prediction mechanism based on conditional probability further strengthens the continuous improvement effect of nursing quality on the basis of improving the standardization of nursing operations and reducing human errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is the system flow chart of the present invention;

[0044] Figure 2 is the flow chart of the present invention for inputting the state sequence into the finite state machine model;

[0045] Figure 3 is the flow chart of the present invention for obtaining the verification result of the nursing process sequence;

[0046] Figure 4 is the flow chart of the present invention for calculating the deviation value of the nursing time for each step;

[0047] Figure 5 is the flow chart of the present invention for obtaining the evaluation result of the nursing operation time deviation;

[0048] Figure 6 is the flow chart of the present invention for constructing the step sequence in the nursing operation record;

[0049] Figure 7 is the flow chart of the present invention for obtaining the prediction result of the risk of missing nursing steps;

[0050] Figure 8 is the flow chart of the present invention for obtaining the feedback optimization result of the nursing operation data. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0052] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be construed as a limitation to the present invention. In addition, in the description of the present invention, the meaning of "a plurality of" is two or more, unless otherwise specifically defined.

[0053] Please refer to Figure 1 , the nursing operation tracking and feedback system includes:

[0054] The nursing process sequence verification module collects the data of nursing records, including the nursing time, nursing frequency, and handover situation of nursing staff for each nursing operation step, initializes the finite state machine model, divides each nursing operation process into corresponding state nodes, sets the state sequence through the node sequence, inputs the state sequence into the finite state machine model, compares the current operation state with the state sequence of the standard state, records the jump points that are not in order during the operation, judges the missing intermediate states in the path, and whether the path does not reach the termination node, and generates the nursing process sequence verification result;

[0055] The execution time evaluation module, based on the nursing process sequence verification result, fits the state sequence into a Gaussian distribution curve, calculates the deviation value of the nursing time for each step according to the data shown in the Gaussian distribution curve, compares the current operation duration with the standard time distribution, and judges whether the execution time exceeds the set reasonable range. When the execution time is lower than the mean of the standard time distribution, it is marked as an abnormal situation of too fast execution. When it exceeds the set reasonable range, it is marked as an overtime abnormal situation, and generates the nursing operation time deviation evaluation result;

[0056] The nursing step omission prediction module, based on the nursing operation time deviation evaluation result, defines the association probability between each nursing operation step and its subsequent steps, constructs the step sequence in the nursing operation record with reference to the historically saved nursing operation standard probability chain, calculates the conditional probability between the current operation and the standard probability chain, judges whether there are omissions in the steps that do not appear but have a relatively high association probability, predicts the possible subsequent step missing situations, and generates the nursing step omission risk prediction result;

[0057] The high-frequency deviation pattern optimization module, based on the nursing step omission risk prediction result, organizes all the records of nursing operations into sequence data grouped by steps, identifies the frequently occurring incorrect steps or combinations of step sequences in the nursing operations, marks them as deviation labels, and generates the feedback optimization result of the nursing operation data.

[0058] The verification results of the nursing process sequence include the skipped nodes where operations are not performed in sequence, the intermediate state markers where the process is interrupted, and the feedback of path information that does not reach the termination node. The evaluation results of nursing operation time deviation include the deviation data of nursing step time, the marked overtime and overly fast anomalies, and the statistical information of anomaly frequencies. The prediction results of the risk of nursing step omission include the risk analysis of steps with high correlation probability not being executed, the identification of step coherence deviation, and the early warning of possible subsequent operations that may be omitted. The feedback optimization results of nursing operation data include the pattern analysis of frequent operation deviations, the marking of incorrect step combinations, and the recommended information for optimized execution.

[0059] Please refer to Figure 2 , the specific steps for inputting the state sequence into the finite state machine model are as follows:

[0060] Based on the collection of nursing record data, including recording the time, frequency, and handover situation of each nursing operation step, each nursing operation process is divided into inconsistent state nodes to generate an initial finite state machine model;

[0061] The collection of nursing record data first involves recording the time data of each nursing operation step. Each time point uses the standard timestamp format, accurate to the second. The nursing frequency data is recorded by counting the number of executions of each nursing item within a fixed period (such as hours, days, or weeks), and the statistical results are saved for use in subsequent calculation and verification processes. The handover situation of nursing staff is updated in real time through the handover record log, including information such as handover time, handover personnel, and handover content. The system uniformly encodes and saves this information to ensure the integrity of the handover information. When initializing the finite state machine model, first determine the standard process of each nursing operation. Based on the standardized operation steps, each nursing process is divided into corresponding state nodes. For example, the steps of patient cleaning care can be divided into "preparing tools", "disinfection", "wiping", "checking", etc., and each step is set as a state node. In actual operation, these nodes can be created and the transfer conditions of each node can be set by using, for example, state machine modeling software (such as UML tools or state machine design tools) to form a complete finite state machine model.

[0062] By setting the order of each node to form a state sequence, the state sequence is input into the initialized finite state machine model, and the operation state is monitored sequentially according to the node order to obtain the standard state sequence of nursing operations;

[0063] In the process of setting the state sequence according to the node order, based on the standard process of nursing operations, the nursing steps are arranged in chronological order, and each step represents a node and is marked with its order. For setting the state sequence, the system arranges the node order in the standard process in sequence to form a set of state sequences. For example, the node order is "1 - Prepare tools, 2 - Disinfect, 3 - Wipe, 4 - Check", etc., and then inputs this state sequence into the finite state machine model. In the finite state machine model, the standard entry conditions and transfer conditions of each node are set to monitor whether the current operation node meets the requirements of the standard sequence during the nursing operation.

[0064] Please refer to Figure 3 , and the specific steps for obtaining the verification result of the nursing process order are as follows:

[0065] Based on the standard state sequence of nursing operations, compare the actual operation state with the preset standard state sequence. When it is detected that the actual operation is not performed in sequence, mark the jump point and record the deviated node to generate an operation deviation record;

[0066] Input the state sequence into the finite state machine model, and gradually compare the actual state of the current nursing operation with the preset standard state sequence. In the example, assume that the standard state sequence of a patient's cleaning and nursing process is: 1 - Prepare tools, 2 - Clean hands, 3 - Disinfect, 4 - Clean the patient, 5 - Check the nursing result. If the nurse skips the "2 - Clean hands" step and directly enters the "3 - Disinfect" step during the actual operation, the current operation is not performed strictly in accordance with the standard order, and the model will find the missing step of "2 - Clean hands". During the retrieval process, the finite state machine model will compare the sequence numbers of each operation step. When it is detected that the actual operation step does not match the sequence number in the standard state sequence, mark this state as a jump point, record the deviated node in the operation, and store the specific information of this node (such as the deviated step, the identifier of the nurse, time, etc.) in the system log for further analysis and correction of the non - sequential execution in the nursing process in the future.

[0067] Based on the operation deviation record, check the nodes in the nursing operation path, identify the missing intermediate states, judge whether the operation reaches the predetermined termination node, and record the missing situation to generate the verification result of the nursing process order.

[0068] Based on the jump point records, the finite state machine model further checks one by one in the operation path whether there are missing intermediate states and cases where the termination node is not reached. For example, in a nursing process of measuring vital signs, the standard process includes: 1 - Prepare instruments, 2 - Measure body temperature, 3 - Record data, 4 - Replace equipment, 5 - End process. In the actual operation path, if the nursing process skips the step of "3 - Record data" and directly goes from "2 - Measure body temperature" to "4 - Replace equipment", the finite state machine model will find through sequential retrieval that the actual operation does not pass through the "3 - Record data" node and mark this node as a missing intermediate state. The system will record this missing state in detail, including the name of the skipped step and the timestamp, and generate a corresponding exception log. In addition, during the path retrieval, the model will also check whether the operation path reaches the predetermined termination node, that is, whether it enters "5 - End process" after completing all nursing processes. If the process does not reach this termination state, the system will record in the log the situation that the path is not completed, marked as an exception of "not reaching the termination node". Finally, these records generate the complete sequential verification result of the nursing process for subsequent analysis and optimization of the nursing process.

[0069] Please refer to Figure 4 , the specific steps for obtaining the deviation value of the nursing time for each step are as follows:

[0070] Based on the state sequence of the nursing process in the nursing process sequential verification result, perform distribution fitting on the sequence to reflect the time distribution law of nursing operations under normal conditions and obtain a standard time distribution curve;

[0071] In the state sequence of the nursing process, collect and organize the actual operation time data of each nursing step into a complete time data sequence. Then, use professional data analysis software (such as MATLAB or SPSS) to perform Gaussian distribution curve fitting on this data sequence to reveal the time distribution law of each step under the standard nursing process. Specifically, the software will calculate the central position (i.e., the average time) and the fluctuation range (i.e., the standard deviation) of the sequence based on the input time series data, and then generate a Gaussian curve that conforms to the data distribution characteristics. For example, for the "dressing change" step of a certain nursing process, assuming that from a large amount of historical data, it is known that the average operation time of this step is 15 minutes, and most operation durations are concentrated between 12 and 18 minutes. The Gaussian distribution curve generated by the software will take 15 minutes as the central position and gradually decrease symmetrically on both sides to show the distribution of operation time. The shape of this curve indicates that the operation time between 12 and 18 minutes accounts for a relatively large proportion and belongs to a reasonable range, while the cases below 12 minutes or above 18 minutes gradually decrease, indicating a lower probability of deviation from the normal range.

[0072] According to the data of the standard time distribution curve, use the formula:

[0073]

[0074] Calculate the nursing time deviation value D at the i-th step i , and obtain the deviation information of the execution time;

[0075] where D i is used to measure the degree of deviation of the actual execution time of this step from the standard time, T i is the actual nursing time of the i-th step, collected by a real-time monitoring device, and μ i is the mean value of the standard nursing time of the i-th step, obtained based on historical data statistics, and the mean value can be calculated through a large number of historical execution time data, and σ i is the standard deviation of the standard nursing time of the i-th step, obtained by calculating the distribution of nursing time data to reflect the fluctuation of nursing time. This standard deviation is calculated using the formula The calculation is obtained, n is the number of samples, and x k is the time data of the k-th execution, and μ is the mean value.

[0076] For example, if the execution duration of nursing step 1 is T 1 = 18 minutes, the execution duration of nursing step 2 is T 2 = 14 minutes, and the execution duration of nursing step 3 is T 3 = 20 minutes. The mean value of step 1 is μ 1 = 15 minutes, the mean value of step 2 is μ 2 = 15 minutes, and the mean value of step 3 is μ 3 = 18 minutes. After statistical calculation, the standard deviation of step 1 is σ 1 = 2 minutes, the standard deviation of step 2 is σ 2 = 1.5 minutes, and the standard deviation of step 3 is σ 3 = 2 minutes.

[0077] Nursing step 1:

[0078]

[0079] Nursing step 2:

[0080]

[0081] Nursing step 3:

[0082]

[0083] The results show that the nursing time deviation value of nursing step 1 is 1.5, the nursing time deviation value of nursing step 2 is -0.67, and the nursing time deviation value of nursing step 3 is 1.

[0084] Please refer to Figure 5 , and the specific steps for obtaining the evaluation result of the deviation of the nursing operation time are as follows:

[0085] Based on the deviation information of the execution time, compare the nursing time deviation value of each nursing step with the set reasonable range, analyze whether the execution time of each nursing step is within the reasonable range, and obtain the judgment result of the execution rationality of the nursing step;

[0086] For the operation duration of the current nursing step, use the calculated deviation value D 1 = 1.5, D 2 = -0.67, D 3 = 1 to perform the judgment and analysis of the reasonable range. First, compare each deviation value with the preset normal operation range [-2, 2] item by item, and judge whether the operation duration falls within the reasonable range according to the position of its value. The specific steps are as follows: Step 1: The deviation value D 1 = 1.5, and this value is within [-2, 2], indicating that the operation duration of Step 1 is close to the normal range of the standard time distribution and no further processing is required. Step 2: The deviation value D 2 = -0.67. This negative value indicates that the execution time of Step 2 is slightly shorter than the standard time, but it is still within the reasonable range and meets the requirements of the normal distribution. Step 3: The deviation value D 3 = 1. This value is positive and within the reasonable range, indicating that the execution time of Step 3 is slightly longer than the standard time, but it does not exceed the normal distribution range. Therefore, through the above item-by-item judgment, it can be seen that the execution times of all steps do not exceed the set reasonable range, and the operation durations are all regarded as normal.

[0087] Based on the judgment result of the execution rationality of the nursing step, by comparing the nursing time deviation values, mark the execution speed of each nursing step as abnormal or normal, and generate the evaluation result of the deviation of the nursing operation time;

[0088] After passing the judgment of the reasonable range, further determine whether to mark as abnormal according to the deviation value result of each step. The judgment criteria include: when the deviation value is significantly lower than the mean of the standard time distribution (that is, the deviation value is negative, indicating that the operation time is significantly lower than the standard time distribution), mark it as "abnormal for too fast execution"; if the deviation value is higher than the reasonable range, mark it as "abnormal for overtime". Combining the actual results of each deviation value: Step 1: The deviation value D 1 = 1.5, which is not lower than the mean and within the reasonable range, so there is no need to mark it as abnormal. Step 2: The deviation value D 2 = -0.67. Although this value is negative, it is within the reasonable range, so it is marked as "abnormal for too fast execution". Step 3: The deviation value D 3= 1, within a reasonable range and not exceeding the mean, so there is no need to mark it as abnormal. The final evaluation result of the nursing operation time deviation shows that: Step 2 is marked as "abnormally fast execution" because it is executed too fast, and both Step 1 and Step 3 are normal.

[0089] Please refer to Figure 6 , and the specific steps for obtaining the step sequence in the nursing operation record are as follows:

[0090] According to the evaluation result of the nursing operation time deviation, use the formula:

[0091]

[0092] Calculate the correlation probability S i of step S i+1 and the subsequent step S i,i+1 , and obtain the definition result of the correlation probability of the nursing steps;

[0093] Among them, N i,i+1 represents the number of times that step i immediately follows step i + 1. By analyzing the nursing data record, N' represents the total number of nursing operation records, that is, the total number of all nursing records, and this value is obtained through the total number of data entries in the nursing record.

[0094] For example, there are a total of 50 nursing records, and the statistical data is as follows: In 40 nursing records, step S i (such as "patient information verification") immediately follows step S i+1 (such as "drug preparation") appears, that is, N i,i+1 = 40, and the total number of nursing records is N' = 50.

[0095] Substitute these values into the formula for calculation:

[0096]

[0097] This result indicates that in this nursing process, the probability that the step "patient information verification" is immediately followed by "drug preparation" is 0.8, that is, 80%.

[0098] According to the definition result of the correlation probability of the nursing steps, refer to the historically saved standard probability chain of nursing operations, compare and analyze the matching degree between the current operation sequence and the standard probability chain, and compare the correlation of each step in turn to generate the standard step sequence of nursing operations;

[0099] Based on the calculated correlation probability S i,i+1 , compare and analyze it with the historically saved standard probability chain of nursing operations to construct a standard step sequence of nursing operations. During the construction process, first read the standard probability chain of nursing steps recorded in the historical data. Assume that in the standard chain, step S i and Si+1 The associated probability is 0.85. Then, compare the actually calculated associated probability S i,i+1 = 0.8 with the standard probability of 0.85. According to the set threshold, if the difference between the currently calculated probability and the standard probability is within 0.1, it is considered to conform to the standard care process sequence; otherwise, it is marked as having a difference. The specific construction process is as follows: Assume that multiple sets of associated probabilities of care steps are recorded in the standard probability chain. For example, S 1,2 = 0.85, S 2,3 = 0.9, S 3,4 = 0.78, etc. Compare the calculated associated probability of each step with the corresponding probability in the standard chain one by one. Take the step S 1 immediately following S 2 as an example. If the actually calculated associated probability is S 1,2 = 0.8, with a difference of 0.05 from the standard probability of 0.85, then this step conforms to the standard. Through multiple comparisons, construct a step sequence that conforms to the standard chain. Assume that the associated probabilities of steps S 1 to S 4 conform to the standard, but there is no standard step S 4 following S 5 , then it is marked as having a deviation, and based on the comparison results, construct a sequence of care operation steps that is closest to the standard. Finally, substituting specific values for illustration can obtain a relatively reasonable step sequence in the current care operation, which is used for further evaluation and optimization of the care process.

[0100] Please refer to Figure 7 , the steps for obtaining the prediction result of the risk of omission of care steps are specifically as follows:

[0101] Based on the standard step sequence of care operations, use the formula:

[0102]

[0103] Calculate the conditional probability ratio C of the current care operation sequence relative to the standard probability chain to obtain the matching degree data of the current sequence;

[0104] Among them, C is used to evaluate the matching degree of the current operation sequence with the standard care process. If this value is lower than the set threshold, it may indicate the risk of step omission. i represents the step index in the sequence, traversing each step sequence of the care operation step by step, n represents the total number of steps in the care operation, P(A i |A i-1 ) represents the conditional probability that step A i occurs after step A i-1 in the current care record. This conditional probability is obtained through actual data statistics. For example, in the care operation history record, count step A i-1Followed by step A i The frequency of occurrence, divided by the number of occurrences of step A i-1 to obtain the conditional probability, P std (A i |A i-1 ) represents the conditional probability that occurs after step A in the standard care process. This conditional probability is obtained based on historical standard care records and is part of the standard probability chain. For example, in the standard care data, count the frequency of step A i followed by step A i-1 in all care records, divide by the number of occurrences of step A i-1 to obtain the standard conditional probability. i The frequency of occurrence, divided by the number of occurrences of step A i-1 to obtain the standard conditional probability.

[0105] For example, if the statistical data of the care records is as follows, the conditional probability of the current care operation sequence: P(A 2 |A 1 ) = 0.9, P(A 3 |A 2 ) = 0.8, the conditional probability of the standard care step sequence: P std (A 2 |A 1 ) = 0.85, P std (A 3 |A 2 ) = 0.9.

[0106] Calculate the joint conditional probability of the current care sequence:

[0107]

[0108] Calculate the joint conditional probability of the standard probability chain:

[0109]

[0110] Substitute these values into the formula for calculation:

[0111]

[0112] The result shows that the matching degree of the current care operation sequence is 0.941.

[0113] According to the matching degree data of the current sequence, compare the matching degree of the current operation sequence with the preset threshold, analyze the occurrence of key steps in the care process, identify the steps with high conditional probability in the standard process but not appearing in the actual operation records, and generate the prediction result of the risk of missing care steps;

[0114] Based on the calculated current nursing operation sequence matching degree of 0.941, a predictive analysis is carried out on the possible missing subsequent steps. First, compare this matching degree value with the standard threshold (e.g., 0.95). Since the matching degree of 0.941 is lower than this threshold, it indicates that there is a deviation between the current nursing operation and the standard process, and there may be a risk of missing steps. Further analyze the specific steps, focusing on the key steps that should appear in the standard process and have a relatively high conditional probability, especially the subsequent steps following the current operation sequence. For example, assume that in the standard nursing process, "patient information verification" should be followed by "drug preparation", and the standard conditional probability of this subsequent step is 0.88, but the "drug preparation" step is not recorded in the current record. Based on the relatively low matching degree value of 0.941 and the relatively high value of the standard conditional probability (0.88), "drug preparation" is marked as a possible missing subsequent step. Finally, based on the matching degree analysis and the missing situation of the subsequent steps, a prediction result of the risk of missing nursing steps is generated to provide a reference for process inspection.

[0115] Please refer to Figure 8 , the steps for obtaining the feedback optimization result of nursing operation data are specifically as follows:

[0116] Based on the prediction result of the risk of missing nursing steps, all nursing operations are combined in the order of the standard process. Through item-by-item comparison and sequence arrangement, and recording the execution status of each step, sequence data of nursing operations is generated;

[0117] In actual nursing operations, assume that the operation record of a certain nurse shows that he / she has performed the following steps: "patient information verification -> vital sign monitoring -> drug preparation -> drug distribution". According to the prediction result of the risk of missing nursing steps, this series of operation records are sorted into standardized step combination data for easy comparison. During the comparison process, the nursing operation sequence is corresponded and arranged one by one with the standard nursing process. Assume that in the standard nursing process, the correct step order should be "patient information verification -> drug preparation -> vital sign monitoring -> drug distribution". Through comparison, it is found that the order of the two steps "vital sign monitoring" and "drug preparation" in this nursing operation sequence does not meet the requirements of the standard process. In order to reduce the possible risks, when this sequence is output as standard sequence data, it will be rearranged in the order of the standard process, and the deviated steps will be marked so that in subsequent nursing operation management or system feedback, such non-standard operation sequence deviations can be identified in a timely manner. In addition, in the generated sequence data, the system will also record information such as the number of deviations and frequencies to provide data support for analyzing the consistency of nursing operations.

[0118] Based on the sequence data of nursing operations, judge the error step combinations that repeatedly appear in the nursing record, mark the error step combinations and classify them as deviation labels, and generate the feedback optimization result of nursing operation data.

[0119] Based on the standardized nursing operation sequence data, the system further identifies and analyzes the error steps or incorrect step combinations that frequently appear in different nursing records, and marks them as deviation labels. For example, in the nursing records of the recent month, by counting the order of operation steps in 100 records, it is found that the next step "vital sign monitoring" after the combination of "patient information verification -> drug preparation" is omitted 20 times; in addition, 15 records show that directly after "patient information verification", it enters "drug distribution", skipping the two steps of "drug preparation" and "vital sign monitoring". Such deviation patterns have high repeatability and frequency, so they are marked as deviation labels and recorded in the system data. By classifying and identifying these deviation labels, feedback optimization results can be generated. For example, the system will feedback such high-frequency incorrect operation sequences according to the deviation information of the nursing data, so that the nursing management department can improve the process management targeted or provide specialized training to help nursing staff improve the standardization of operations, reduce the recurrence of similar problems, and improve the overall nursing quality and safety.

[0120] The above are only the preferred embodiments of the present invention, and do not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still belong to the protection scope of the technical solution of the present invention.

Claims

1. Nursing operation tracking and feedback system, characterized by: The system comprises: The nursing process sequence verification module collects the data of nursing records, initializes the finite state machine model, sets the state sequence through the node sequence, inputs the state sequence into the finite state machine model, compares the state sequence of the current operation state with the state sequence of the standard state, determines the missing intermediate state in the path, and whether the path does not reach the terminal node, and generates the nursing process sequence verification result; The execution time evaluation module fits the state sequence into a Gaussian distribution curve based on the nursing process sequence verification result, calculates the deviation value of the nursing time of each step according to the data displayed by the Gaussian distribution curve, compares the current operation time with the standard time distribution, determines whether the execution time exceeds the set reasonable range, and generates a nursing operation time deviation evaluation result; The nursing step omission prediction module defines the association probability between each nursing operation step and its subsequent steps based on the nursing operation time deviation evaluation result, builds the step sequence in the nursing operation record with reference to the historically saved nursing operation standard probability chain, calculates the conditional probability between the current operation and the standard probability chain, predicts the possible missing of subsequent steps, and generates the nursing step omission risk prediction result; The high-frequency deviation pattern optimization module identifies the incorrect steps or combinations of step sequences that frequently occur in nursing operations based on the nursing step omission risk prediction results, marks them as deviation labels, and generates feedback optimization results for nursing operation data.

2. The nursing operation tracking and feedback system according to claim 1, characterized in that: The acquisition step of inputting the state sequence into the finite state machine model is specifically as follows: Based on the collection of nursing record data, including recording the time, frequency, and handover status of each nursing operation step, each nursing operation process is divided into inconsistent state nodes to generate an initialized finite state machine model; A state sequence is formed by setting the order of each node, the state sequence is input into the initialized finite state machine model, and the operation state is monitored in sequence according to the node order to obtain a standard state sequence of the nursing operation.

3. The nursing operation tracking and feedback system according to claim 2, characterized in that: The steps for obtaining the nursing process sequence verification result are specifically as follows: Based on the standard state sequence of the nursing operation, the actual operation state is compared with the preset standard state sequence, and when it is detected that the actual operation is not performed in sequence, the jump point is marked and the deviation node is recorded to generate an operation deviation record; Based on the operation deviation record, check the nodes in the nursing operation path, identify the missing intermediate states, determine whether the operation reaches the predetermined termination node, record the missing conditions, and generate the nursing process sequence verification result.

4. The nursing operation tracking and feedback system according to claim 3, characterized in that: The steps for obtaining the deviation value of the nursing time for each step are specifically as follows: Based on the state sequence of the nursing process in the nursing process sequence verification result, distribution fitting is performed on the sequence to reflect the time distribution law of the nursing operation under normal conditions, and a standard time distribution curve is obtained; Based on the data of the standard time distribution curve, the formula is used: Calculate the nursing time deviation value D of step i i , get the deviation information of execution time; Among them, μ i is the mean standard nursing time of step i, σ i is the standard deviation of the standard care time for step i.

5. The nursing operation tracking and feedback system according to claim 4, characterized in that: The steps for obtaining the nursing operation time deviation evaluation result are specifically as follows: Based on the deviation information of the execution time, the nursing time deviation value of each nursing step is compared with the set reasonable range, and whether the execution time of each nursing step is within the reasonable range is analyzed to obtain the execution rationality judgment result of the nursing step; Based on the rationality judgment result of the execution of the nursing steps, by comparing the nursing time deviation value, the execution speed of each nursing step is marked as abnormal or normal, and the nursing operation time deviation evaluation result is generated.

6. The nursing operation tracking and feedback system according to claim 5, characterized in that: The steps for obtaining the step sequence in constructing the nursing operation record are specifically as follows: According to the nursing operation time deviation evaluation results, the formula is used: Calculation step S i With the subsequent steps i+1 The association probability S i,i+1 , get the associated probability definition result of the nursing steps; Among them, N i,i+1 represents the number of times step i appears immediately after step i+1, and N represents the total number of nursing operation records; According to the association probability definition result of the nursing steps, referring to the historically saved nursing operation standard probability chain, the matching degree between the current operation sequence and the standard probability chain is compared and analyzed, and the association of each step is compared in turn to generate a nursing operation standard step sequence.

7. The nursing operation tracking and feedback system according to claim 6, characterized in that: The steps for obtaining the nursing step omission risk prediction result are specifically as follows: Based on the standard sequence of steps in the nursing procedure, the formula is used: Calculate the conditional probability ratio C of the current nursing operation sequence relative to the standard probability chain to obtain the matching data of the current sequence; Where n represents the total number of nursing steps, P(A i |A i-1 ) means that in the current nursing record, step A i In step A i-1 The conditional probability of occurrence after the occurrence, P std (A i |A i-1 ) indicates that in the standard nursing process, step A i In step A i-1 The conditional probability of occurrence after appearance; According to the matching data of the current sequence, the matching degree of the current operation sequence is compared with the preset threshold, the occurrence of key steps in the nursing process is analyzed, the steps with high conditional probability in the standard process and not appearing in the actual operation records are identified, and the risk prediction results of missing nursing steps are generated.

8. The nursing operation tracking and feedback system according to claim 7, characterized in that: The steps for obtaining the feedback optimization result of the nursing operation data are specifically as follows: Based on the nursing step omission risk prediction result, all nursing operations are combined in the order of the standard process, and the sequence data of the nursing operations are generated by comparing and arranging the sequence item by item, and the execution status of each step is recorded; Based on the sequence data of the nursing operation, the wrong step combination that appears repeatedly in the nursing record is determined, the wrong step combination is marked and classified as a deviation label, and a feedback optimization result of the nursing operation data is generated.

Images