Digital twin driven perioperative nursing plan dynamic optimization method

Abstract

The present application relates to a digital twin driven perioperative nursing scheme dynamic optimization method, sets an intervention sensitive area associated with the current postoperative state of the patient, and limits the physical limit conditions of the nursing operation trigger threshold according to the operation type and anatomical position of the patient; the time period from the completion of the operation to the recovery of the related physiological indicators to the steady state is set as the impact period, and other nursing tasks that produce physical effects on the same area or its physiologically related areas are prohibited from entering the execution sequence; a damping feedback component is embedded in the nursing feedback channel of the digital twin, and when the change rate is higher than the preset physiological tolerance threshold, a rhythm slowing mechanism is started to adjust the execution rhythm of the subsequent tasks; an artificial operation behavior monitoring node is set in the nursing terminal or the bed area; based on the data collected by the monitoring node, an execution behavior deviation judgment unit is constructed for comparison with the system recommended path to identify the implicit correction behavior of the system strategy.

Description

TECHNICAL FIELD

[0001] The present application relates to a perioperative nursing plan optimization method, in particular to a digital twin driven perioperative nursing plan dynamic optimization method. BACKGROUND

[0002] At present, there is a perioperative risk assessment and clinical decision intelligent auxiliary system as disclosed in CN111009322A, which realizes intelligent integration in aspects of structured collection of preoperative-intraoperative-postoperative cycle data, risk analysis, intervention plan and quality management, and to some extent solves the problems of fragmentation and heterogeneity of traditional evaluation tools, but there are still some obvious technical shortcomings and practical application drawbacks in responding to perioperative nursing path dynamic optimization, especially in individualized nursing intervention driven by digital twin. First, this kind of system focuses on risk identification and static decision assistance, and the focus is on preoperative data integration and postoperative follow-up quality control, and lacks fine modeling of nursing behavior response mechanism in the key stage of intraoperative to postoperative recovery. Although the decision assistance model constructed by the system has the function of structured plan recommendation, it lacks the ability of rhythm control, execution time sequence adjustment and task dependency dynamic analysis of nursing tasks, and cannot adjust, insert or rhythm suppress tasks according to the real-time fluctuation of postoperative state of patients. This design lacking the feedback ability of nursing-physiological response closed loop is easy to cause problems such as over-dense nursing rhythm, task overlap or patient load exceeding threshold in the process of clinical high-frequency nursing operation, and further causes the mismatch between nursing behavior and patient tolerance, increases the risk of postoperative complications. In addition, the existing system focuses more on risk prediction rather than task execution management, even if it realizes the push of part of nursing suggestions, it is mainly driven by static rule base, and the task recommendation does not have process perception ability, which cannot automatically buffer or solve the next task according to the physiological response generated by the previous task execution, and this is the most critical dynamic scheduling ability of postoperative nursing optimization.

[0003] Secondly, such systems usually lack a deep understanding mechanism of the actual behavior deviation of nursing staff. In real-world scenarios, nursing staff often fine-tune the recommended path based on experience, such as performing certain soothing care in advance, delaying intense operations, or skipping tasks that are not suitable for the moment. However, existing systems cannot automatically identify these implicit correction behaviors and cannot distinguish whether the deviation is driven by human randomness or patient status. They only treat the deviation as an abnormal or incorrect path, thus failing to convert potential clinical experience into learning factors for path optimization. Essentially, such systems remain at the technical implementation of nursing recommendation output, lacking a mechanism to semantically reflect nursing behavior input and use it for model iteration, which severely limits the adaptability and evolution ability of the system. In addition, perioperative management systems usually lack modeling means for the physical association and physiological impact period between tasks. For example, when turning over care causes changes in local wound tension, the system cannot perceive the impact period and will not restrict the nursing tasks in the same area from entering the execution sequence during the impact period. This lack of impact period management capability can easily lead to physical load stacking.

[0004] Furthermore, this technical solution does not construct a mapping mechanism for behavior path deviation and group consensus. In group nursing operations, multiple nursing staff often show consistent deviation trends for the same task in specific pathological scenarios. This phenomenon has a high value for practical optimization and should be actively learned by the system and used for task priority adjustment. However, existing systems lack mechanisms for deviation behavior classification, reproduction, and multi-user trajectory comparison, and do not have dynamic weight decay functions or strategy update models as support, resulting in rigid system paths that cannot incorporate the collective wisdom of real-world operations. At the same time, while this solution involves connecting intelligent terminals and cloud platforms, its data feedback chain still mainly relies on structured forms and event records, lacking real-time perception nodes at the behavior level, such as monitoring means for nursing action paths, execution interruptions, withdrawals, and sequence dislocations. This results in a lack of behavior alignment between task execution and recommended strategies, making it difficult to form precise interventions or real-time optimizations when dealing with complex nursing field operations, and easily falling into the dilemma of disconnection between recommendations and reality. SUMMARY

[0005] The purpose of the present application is to provide a digital twin driven perioperative nursing scheme dynamic optimization method, thereby solving some of the problems and deficiencies pointed out in the background art.

[0006] The technical scheme adopted by the present application to solve the above technical problems is as follows: a digital twin driven perioperative nursing scheme dynamic optimization method, comprising: setting an intervention sensitive area associated with the current postoperative state of a patient, the intervention sensitive area being delimited according to the type of surgery and the anatomical position of the patient, and having a physical limitation condition for defining a triggerable threshold of nursing operation, the physical limitation condition including a body position angle change value, a pressure contact area, and an action duration;

[0007] A reaction period is predefined for each nursing operation in the target area, and the time period from the completion of the operation to the recovery of the relevant physiological indicators to the steady state is set as the impact period, during which other nursing tasks that produce physiological effects on the same area or its physiologically associated areas are prohibited from entering the execution sequence;

[0008] A damping feedback component is provided, embedded in the nursing feedback channel of the digital twin, for receiving the physiological parameter change signal caused by the execution of the nursing task, and when the signal presents a change rate higher than the preset physiological tolerance threshold between consecutive nursing tasks, the damping feedback component starts the rhythm slowing mechanism to adjust the execution rhythm of the subsequent task; A set of artificial operation behavior monitoring nodes are provided in the nursing terminal or bed area to record the operation path, action interruption, withdrawal behavior and duration generated by the nursing staff during the execution process; Based on the data collected by the monitoring nodes, an execution behavior deviation judgment unit is constructed for comparing with the system recommended path to identify the implicit correction behavior of the nursing staff to the system strategy.

[0009] Further, the damping feedback component comprises:

[0010] A physiological change monitoring interface is used to collect the physiological change data of the patient in real time after the execution of the nursing task, including heart rate, blood pressure, muscle tension and blood oxygen saturation;

[0011] A change intensity analysis unit is connected with the physiological change monitoring interface, for identifying the physiological response intensity after the nursing task according to the physiological change data, and outputting a response fluctuation signal;

[0012] A nursing task buffer control unit is connected with the change intensity analysis unit, for dynamically adjusting the execution rhythm of the subsequent nursing task according to the response fluctuation signal, wherein:

[0013] When the fluctuation signal is less than a first threshold value, the original task rhythm is maintained;

[0014] When the fluctuation signal is greater than a second threshold value, a task buffer delay instruction is inserted;

[0015] When the fluctuation signals of two consecutive tasks both exceed the second threshold value, a rhythm suppression state is started;

[0016] An interactive linkage interface is connected with the buffer control unit, for transmitting the task adjustment information to the nursing operation interface in real time, and providing an interactive option of accepting or rejecting the delay execution;

[0017] A twin state update interface is connected with the digital twin master system, for synchronizing the rhythm adjustment results and manual selection results of the nursing task to the twin system to update the patient state and acceptable nursing threshold.

[0018] Further, the physiological change monitoring interface comprises a nursing behavior recognition subunit for judging the starting time and execution mode of the nursing behavior and synchronously collecting physiological change data to align the nursing behavior and physiological response; the change intensity analysis unit comprises a critical point recognition logic structure for dynamically adjusting the threshold recognition standard when a nonlinear change mode is monitored.

[0019] Further, the nursing task buffer control unit has a task grading reconstruction mechanism for decomposing a scheduled high-load nursing task into multiple subtasks, which are executed in stages under rhythm suppression, and the decomposing scheme is derived from the twin in-vivo patient individual tolerance data; the interactive linkage interface is connected with a nursing staff identity recognition module for dynamically adjusting the execution strategy of the task delay suggestion given by the system according to the executor level and experience level.

[0020] Further, the first threshold value is that within a preset time period after the execution of the nursing task, the change rate of the physiological parameter does not exceed 1.5 times the average change rate under the preoperative stable state of the parameter, which is used to determine that the patient is in a physiological stable state and allows the original nursing task rhythm to be maintained.

[0021] Further, the second threshold value is that within a preset time period after the execution of the nursing task, the change rate of the physiological parameter exceeds 2.5 times the average change rate under the preoperative stable state of the parameter, which is used to determine that the patient is in a high-load physiological response state, triggering task delay execution or rhythm suppression mode; the physiological parameter includes heart rate, systolic pressure, diastolic pressure, blood oxygen saturation or muscle tension, and the change rate is the numerical variation amplitude in unit time divided by the time length.

[0022] Further, the method for identifying the implicit correction behavior of the nursing staff to the system strategy comprises:

[0023] The nursing path suggestion generated by the digital twin system is recorded as a recommended path, which includes operation type, recommended execution time, sequence and frequency information of the nursing task; the action information of the nursing staff when actually executing the nursing task is collected by the collection device set on the nursing terminal, bed or nursing equipment, including actual operation time, operation type, execution sequence, task skipping or repeated execution record;

[0024] The actual execution path is compared with the recommended path to identify whether there is a persistent deviation behavior, including task advance, delay, frequency change, replacement behavior or sequence misplacement; after identifying the above deviation behavior, it is judged whether it constitutes a stable operation mode deviation, and if the preset trend condition is met, it is determined as the implicit correction behavior of the nursing staff to the system suggestion.

[0025] Further, the step of determining whether the implicit modification behavior is constituted includes:

[0026] The identified deviation behaviors are classified by category, including rhythm modification, alternative action, skip operation, and sequence rearrangement; different types of deviations are set with identification logic and processing rules, so that the system can determine the path optimization strategy according to the classification results; the current nursing staff's execution behavior is compared with the operation trajectory of a plurality of users in a similar situation; when a plurality of users have a convergent deviation mode for the same recommended task, the task is marked as a weak recommendation item and the priority is reduced;

[0027] The deviation behavior of the current nursing staff under a specific pathological state is matched with the behavior trajectory of a plurality of nursing staff in a similar clinical situation; when a consistent deviation trend is identified for the same recommended task, the system marks the task as a weak recommendation item and dynamically reduces its priority weight in the path generation model; to achieve the function, the following nursing behavior deviation weight decay function is defined:

[0028]

[0029] Wherein:

[0030] is the dynamic priority weight factor of a recommended nursing task at the current time , the value range ; the smaller the value, the more the priority should be lowered by the system; is the system adjustment coefficient, used to control the sensitivity of the deviation behavior to the path strategy; it is a positive real number, the larger the decay is faster; is the number of deviation behavior types associated with the current recommended task (up to 4 types: rhythm, alternative, skip, and rearrangement); is the importance weight of the th type of deviation behavior set in the system, reflecting the strength of its impact on the overall path quality; is the behavior deviation intensity of the th type of deviation behavior detected within the time , such as the degree of frequency deviation, operation advance amplitude, normalized to represent; is the deviation behavior amplification index, which controls the nonlinearity of behavior intensity on the decay result, usually a real number greater than 1; is the behavior consensus factor, indicating the coincidence degree of the th type of deviation behavior of the historical users in the same situation, the value range , the larger the deviation behavior is more common;

[0031] The intensity, importance and user consensus degree of different types of deviation behaviors are fused, and the path priority weight of the nursing task at the current time point is calculated in real time; when the deviation intensity is high, there is high consensus, and the type influence weight is high, the path priority weight of the nursing task will decrease rapidly;

[0032] On the contrary, if the deviation is only occasional, low intensity or without user consensus, the task still maintains a high recommendation intensity; the function can be embedded into the task priority queue in the path generation module to realize dynamic convergence and adaptive recommendation update of the nursing strategy.

[0033] Further, the identification of the implicit correction behavior comprises:

[0034] Cross-judgment is performed on the patient's current physiological state data and the execution deviation behavior; when the nursing staff continuously avoids or adjusts the nursing task in a certain pathological state, the deviation behavior is attributed to the patient state driven correction behavior, and priority learning is performed; after identifying similar deviation behaviors for three times in succession, an active inquiry prompt is sent to the nursing staff through the nursing terminal, and the reason for the correction is requested, and the nursing staff is allowed to select experience optimization, patient intolerance or equipment limitation options, and the feedback will be used to label the deviation behavior attribute.

[0035] The beneficial effects of the present application are: by incorporating the preoperative, intraoperative and postoperative multi-source physiological state data of the patient into the twin model for real-time mapping, the rhythm, order and mode of the nursing task can be dynamically adjusted according to the individual tolerance of the patient, the recovery rhythm and the sudden reaction, avoiding the one-size-fits-all fixed nursing process, and significantly improving the accuracy and safety of nursing intervention. By embedding the damping feedback component, the system can automatically judge whether the patient is in an overload state based on the physiological fluctuation intensity between consecutive tasks, and real-time trigger the rhythm slowing down, task delay or task decomposition strategy, effectively reducing the risk of secondary injury caused by over-frequent nursing intervention.

[0036] The actual operation behavior of the nursing staff is collected in real time, compared and analyzed with the system recommended path, and the implicit correction behaviors such as rhythm adjustment, task skipping and order rearrangement are identified, especially when the operation trend deviates, the system can actively attribute and learn the group experience, thereby improving the clinical fit and practicality of the recommended path. By accurately aligning the nursing behavior nodes with the patient's physiological response intensity and combining the nonlinear change recognition algorithm, not only the task execution effect can be judged, but also the task rhythm, execution intensity and even the nursing strategy can be adjusted based on the real-time state of the patient, truly realizing the closed-loop optimization mechanism with the patient response as the core. By innovatively introducing the deviation behavior weight decay function, the system can adjust the priority of each nursing task in the path generation model in real time according to the behavior intensity, deviation type influence degree and group consensus degree, so that the nursing system has the ability of self-evolution and self-learning. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating the digital twin-driven perioperative nursing optimization process of this invention.

[0038] Figure 2 This is a structural diagram showing the functional relationship of the damping feedback component of the present invention.

[0039] Figure 3 This is a diagram showing the relationship between the implicit modification and dynamic optimization functions of the nursing pathway in this invention.

[0040] Figure 4 This is a flowchart illustrating the dynamic control of postoperative care rhythm supported by digital twins in Embodiment 1 of the present invention.

[0041] Figure 5 This is a flowchart of the closed-loop process for digital twin nursing pathway recommendation and implicit correction in Embodiment 2 of the present invention. Detailed Implementation

[0042] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0043] Combined with appendix Figure 1The digital twin driven perioperative nursing scheme dynamic optimization method of the application sets an intervention sensitive zone associated with the current postoperative state of the patient. The intervention sensitive zone is a limited area defined according to the type of operation received by the patient and the anatomical site information corresponding to the postoperative recovery stage. The system dynamically retrieves the preoperative image information, intraoperative operation record and postoperative risk distribution map according to the operation type classification and postoperative stage, generates an individualized three-dimensional sensitive area mapping, and displays the sensitive area in the digital twin model with markable and high response attributes, which is used to constrain the intervention range and operation intensity of specific nursing behaviors. Each intervention sensitive zone is assigned at least one set of physical limitation conditions to define whether the nursing behavior can be allowed or recommended to be executed by the system under the current state. The physical limitation conditions at least include a body position angle change value, a pressure contact area and an action duration. The body position angle change value is used to limit the upper limit of the amplitude of the body adjustment operation involved in the nursing, so as to prevent the risk of pulling, compressing or poor drainage of the suture area after the operation caused by too large angle. The pressure contact area is used to control the contact range of a single nursing operation on the skin, wound or tenderness area of the patient, so as to avoid the induction or aggravation of pressure ulcers and postoperative tissue damage. The action duration is used to determine whether the persistence of a specific nursing behavior (such as turning over, physical drainage, scrubbing, assisting sitting up, etc.) in a unit time exceeds the tissue tolerance threshold. This parameter is dynamically updated in combination with the real-time monitoring of the patient's muscle tension change, autonomous action response and heart rate fluctuation and other indicators, so that the system can determine the allowable degree of the current nursing operation based on the simulation output of the twin body model. Under the premise of meeting the physical limitation conditions, the system allows the subsequent nursing tasks to enter the candidate execution sequence, thereby forming a perioperative nursing optimization control mechanism based on the dynamic perception and response constraints of the intervention sensitive zone.

[0044] The reaction period of each nursing operation in the target area of the patient is predefined, which refers to the whole process time period from the initiation, development to the recovery to the steady state of the local or systemic physiological reaction caused by the nursing behavior after the specific execution. The reaction period is parameterized modeled and the basic time interval is generated by the system through the fusion of historical case data, postoperative recovery model and digital twin simulation results. The target area can be the location of the postoperative wound, the functionally sensitive area (such as the chest and abdominal cavity, the lower limb venous access, the peripheral nerve distribution area) or the specific part with enhanced reactivity after multiple interventions. The system monitors the change trend of the relevant physiological indicators (including but not limited to heart rate, blood pressure, respiratory rate, skin temperature, muscle tension, drainage volume, blood oxygen saturation, etc.) in real time after the completion of the nursing task execution. When the monitoring indicators gradually return to the preoperative steady state level or enter the set physiological fluctuation safety threshold range, it can be determined that the reaction period of the current nursing operation has been completed, which provides the basis for the next path recommendation of the system. The time period between the completion of the nursing operation and the recovery of the physiological indicators to the steady state is defined as the impact influence period. In the impact influence period, the system activates the mutual exclusion scheduling mechanism of the nursing task, and eliminates all nursing tasks that will cause physiological effects in the same target area or its physiologically related area from the candidate task pool. The physical effects include direct contact, indirect pulling, regional compression, liquid or gas intervention and stimulating operation, etc. to avoid continuous operation causing excessive load of local tissue, superimposed stress reaction or monitoring indicators confusion. At the same time, the definition of the physiologically related area is realized by logical linkage judgment according to the anatomical path mapping and the system physiological response path table, so that even if the physical contact parts of the nursing tasks are different, if they cause synchronous influence on the same tissue system, they will also be restricted by the system to enter the current execution sequence, thereby constructing the nursing operation rhythm regulation structure with the impact influence period as the core.

[0045] A damping feedback component is arranged in the care feedback channel of the digital twin, which receives the physiological parameter change signals generated by the patient after each care task is completed. The physiological parameters include, but are not limited to, heart rate, blood pressure, body temperature, blood oxygen saturation, muscle tension, respiratory rate, and subjective score feedback. The damping feedback component has a continuous response recognition function, which can form time series data of parameter changes between multiple care tasks. When the system identifies that the physiological response rate between consecutive tasks exceeds the preset physiological tolerance threshold, it is determined that the patient's current physiological adjustment ability to consecutive care behaviors is insufficient, and there is a high risk of high tissue load, rhythm disorder or potential risk signals. At this time, the damping feedback component starts the rhythm slowing mechanism, which adjusts the execution rhythm of subsequent care tasks dynamically to achieve time rhythm buffering of care tasks. The specific way includes postponing the task start time, introducing a rest period between tasks, or recommending a low-stimulation alternative task to ensure that the patient's physiological response recovers to an acceptable range before the next step of intervention. At the same time, in order to identify the adjustment intention of the nursing staff on the system recommendation path at the execution level, a group of artificial operation behavior monitoring nodes are arranged at the nursing terminal or the bed area. The monitoring nodes include a manual recording button, an execution path tracker, an action trigger sensor, and a time stamp recording device, which are used to record the key behavior parameters of the nursing staff during task execution, including operation path selection, number of interruption actions, task withdrawal records, and duration of each operation. The system constructs an execution behavior deviation judgment unit based on the monitoring data, compares the actual operation path with the system recommended path, and identifies whether there is a non-accidental, repetitive or trend implicit correction behavior. Implicit correction behavior refers to the behavior of the nursing staff adjusting, avoiding, changing the order or performing alternative of the system recommended strategy without explicit feedback channels. If it is determined that such deviation behavior appears continuously in multiple cycles, the system will mark the care task as a potential optimization path node and include the deviation feature in the subsequent path generation for iterative optimization.

[0046] The accompanying drawings Figure 2The damping feedback component is embedded in the care feedback channel of the digital twin and consists of multiple functional units, and the structure comprises: a physiological change monitoring interface for real-time collection of physiological change data of the patient after completion of the care task, the data including but not limited to heart rate, blood pressure, muscle tension and blood oxygen saturation, and the collection can be based on bedside monitoring equipment, wearable sensing devices or embedded care sensing nodes; the physiological change monitoring interface synchronously transmits the collected data to the change intensity analysis unit connected thereto, the change intensity analysis unit is used for interval analysis and trend identification of the physiological data, and the intensity of the physiological response of the patient after the care task is calculated, the intensity can be expressed as the deviation amplitude, change slope or cross-parameter coordinated fluctuation degree of the target index in a unit time, and the result is output in the form of response fluctuation signal; the response fluctuation signal is transmitted to the care task buffer control unit connected thereto, and the buffer control unit sets the dynamic adjustment logic of the task execution rhythm according to the response intensity value, wherein: when the fluctuation signal is lower than the first threshold value set by the system, it indicates that the patient responds stably to the care task, and the system maintains the current task rhythm unchanged; when the fluctuation signal is higher than the second threshold value, the system inserts a task buffer delay instruction, and sets a minimum waiting window for the next care operation in the task sequence; if the fluctuation signal exceeds the second threshold value caused by two continuous tasks, it is judged that the patient enters a high sensitivity state, the system triggers the rhythm suppression state, suspends the recommendation of all high-load care tasks, and activates the priority recommendation strategy of the recovery care suggestion; the rhythm adjustment information is transmitted to the care operation interface in real time through the interactive linkage interface connected with the buffer control unit, the interface provides an interactive option of explicitly accepting or rejecting the delay execution, the nursing staff can manually select based on clinical judgment, and the system synchronously records the selection result and operation behavior; at the same time, the damping feedback component also comprises a twin state updating interface, which is used for synchronously uploading the rhythm adjustment result of each care task and the manual interactive selection result of the nursing staff to the digital twin main system, so as to update the physiological response model, operation tolerance threshold and rhythm sensitivity parameters of the corresponding virtual patient.

[0047] The physiological change monitoring interface in the damping feedback assembly further includes a nursing behavior identification subunit, which is used to determine the starting time point and execution mode of the nursing behavior during the actual execution of the nursing task based on the operation end input signal, the action trigger mechanism or the behavior path tracking result, including but not limited to whether the operation is a planned task, whether it is a substitute behavior, whether there is a step skipping or reverse sequence phenomenon, and recording the actual position of the behavior in the nursing sequence in combination with the time stamp, thereby maintaining strict time synchronization with the patient physiological change data acquisition, establishing a one-to-one logical mapping relationship between the nursing behavior trigger event and the physiological response data, achieving precise alignment of the nursing behavior and the physiological reaction, avoiding physiological signal interpretation errors caused by acquisition delay, operation drift or unintended intervention, and improving the system's ability to determine the cause and effect relationship; at the same time, the change intensity analysis unit also includes a critical point identification logic structure for monitoring and determining whether there is a nonlinear change mode in the physiological change data. The logic structure can identify nonlinear change characteristics including sudden response, hysteresis response, biphasic fluctuation, platform delay and amplification type perturbation, etc. When the above abnormal fluctuation mode is identified, the system triggers a dynamic threshold adjustment mechanism, adjusts the originally set physiological parameter judgment threshold according to the response curve trend, or recalculates the window interval, thereby avoiding false positives, missed judgments or false delays in the nonlinear physiological response stage.

[0048] The nursing task buffer control unit possesses a task hierarchical reconstruction mechanism. When the system identifies a patient in a state of rhythm suppression, it structurally decomposes originally planned high-load, highly continuous, or physiologically impactful nursing tasks into multiple sub-tasks with lower operational intensity, shorter execution time, or less physiological burden. These sub-tasks are then re-embedded into the nursing task sequence in a progressive manner to reduce the overall intervention pressure on the patient during periods of high sensitivity. The decomposition scheme is derived from the patient's historical physiological tolerance data, postoperative response stability score, and nursing behavior response model within the digital twin. Before each task decomposition, the system calls upon the corresponding virtual patient model within the twin system to evaluate the simulated response of different sub-tasks, thereby determining whether the decomposed task combination is feasible and physiologically safe in the current state. This ensures that the hierarchically reconstructed task structure both meets nursing objectives and provides a buffer function for intervention rhythm. At the same time, the interactive linkage interface is connected to the nursing staff identification module. After the task postponement, dismantling, or buffering rhythm suggestions are generated, the system dynamically adjusts the suggested strategies based on dimensions such as the identity, professional title, clinical experience level, and execution authority level of the personnel currently performing the nursing task. When the system identifies the operation as being performed by a high-level or senior nursing staff, it allows for relaxation of the mandatory execution conditions of some delay instructions or provides more autonomous path adjustment options to ensure the flexibility of the human-machine collaborative decision-making mechanism. Without disrupting the system's rhythm control logic, the system respects the initiative of clinical experience in optimizing the nursing pathway. At the same time, the system incorporates the nursing staff's acceptance or modification of the task rhythm suggestions into the twin model feedback loop, which is further used to improve the behavior deviation identification mechanism and experience knowledge learning module. This enables multi-dimensional dynamic collaborative control of the perioperative nursing pathway among the twin system, physiological state, and manual execution.

[0049] The first threshold in the nursing task buffer control unit is used to determine whether the patient is in an acceptable physiological stable state after the nursing task is performed, so as to decide whether the pace of the subsequent nursing task needs to be adjusted. Specifically, the first threshold is defined as follows: within a preset time period after the completion of the nursing task, the change rate of the monitored one or more physiological parameters does not exceed 1.5 times the average change rate of the corresponding parameters in the preoperative stable state of the patient, that is, the physiological parameters in the preoperative resting state are sampled to form a basic steady state model, and the average value of the change rate is extracted as a baseline index. Then, a response window is set after the nursing task is performed, and the core parameters such as heart rate, blood pressure, muscle tension and blood oxygen saturation are sampled and the change rate is calculated in real time. If the rate value obtained always maintains within 1.5 times the steady state baseline value, the system determines that the patient is currently in a physiological stable state, indicating that the patient's response to the current nursing task is within a controllable range, and there is no risk of obvious overload, stress activation or regulation imbalance. Therefore, the nursing task buffer control unit does not adjust the execution pace of the subsequent task, and allows the nursing path to continue according to the original set time pace, so as to ensure the implementation efficiency of the continuous nursing process. At the same time, the threshold strategy combines parameterized modeling and dynamic window judgment, has good individual adaptability and cross-cycle scalability, and can dynamically adapt to various nursing scenes and patient states and realize safe pace control.

[0050] The second threshold value is set in the nursing task buffer control unit for rhythm adjustment judgment, which is used to identify whether the short-time physiological response of the patient after the execution of the nursing task is in a high load state, and is specifically defined as: within a preset time period after the completion of the execution of the nursing task, the monitored core physiological parameters are continuously sampled and the rate is calculated, and if the obtained change rate exceeds 2.5 times of the average change rate of the corresponding parameters in the preoperative stable state of the patient, the system determines that the nursing operation has caused a physiological impact or stress response beyond the normal tolerance range, and triggers the task rhythm intervention mechanism, including generating a task delay execution instruction or activating a rhythm suppression mode, wherein the physiological parameters include but are not limited to heart rate, systolic pressure, diastolic pressure, blood oxygen saturation and muscle tension, the sampling mode can be obtained based on a continuous monitoring device or a periodic data interface, the change rate is the numerical variation amplitude of the current parameter in a unit of time divided by the length of the time, which reflects the instantaneous intensity of parameter change, the system compares the instantaneous change rate with the steady-state mean value of the patient in the preoperative resting state, judges whether the relative deviation degree exceeds the second threshold value, and decides whether to delay, buffer or completely suspend the next nursing task, so as to prevent the superposition of multiple tasks from causing irreversible physiological damage to the patient, and at the same time, the threshold judgment mechanism combines the physiological baseline data of the patient individual in the digital twin system and the real-time feedback model, has dynamic adaptability and individual difference recognition ability, ensures that the rhythm control has reaction sensitivity and clinical safety redundancy, and improves the stability and self-adjusting ability of the perioperative nursing process.

[0051] Combining the attached Figure 3The judgment method for identifying the implicit correction behavior of nursing staff to the system recommended strategy, the specific steps are: the individualized nursing path suggested by the digital twin system is recorded as a recommended path, the recommended path includes the nursing task item generated for the current postoperative state of a specific patient, each task item contains operation type, recommended execution time point or time period, execution order and repetition frequency and other key scheduling parameters, forming a standard nursing task sequence in the system; then, through the data acquisition device deployed on the nursing terminal, bed area or intelligent nursing equipment, the operation behavior data of the nursing staff in the process of executing the nursing task is collected in real time, the operation data includes but is not limited to actual operation time, operation type, task order, task skipping condition, repetition execution times, operation duration and whether there is manual interruption and other information, the information is packaged and mapped into the same data structure as the recommended path; the system compares the actual execution path with the recommended path item by item, identifies whether there is a persistent deviation behavior, the deviation behavior includes the advance or delay execution of the task, the change of execution frequency, the use of non-recommended task to replace the original task, the disturbance of recommended order or the skipping of key steps and other operation differences, the judgment standard is based on the time sequence deviation between behavior and path, the inconsistency of execution strategy and the repeated behavior correction characteristics; when the above deviation behavior appears more than the minimum trend recognition times set by the system in the continuous nursing period, or meets the behavior deviation amplitude threshold and time persistence standard, the system judges that the behavior constitutes a stable operation mode deviation, and accordingly judges that it is the implicit correction behavior of the nursing staff to the system strategy, that is, the recommended path is not modified through explicit feedback or operation interface, but the path adjustment tendency is expressed in the actual operation, the identification result will be pushed to the digital twin system feedback module, for behavior factor modeling and strategy self-adaptive correction in the subsequent path re-optimization process.

[0052] In the process of judging whether the nursing staff makes implicit correction behavior to the system recommended path, the system classifies the identified deviation behaviors by categories, including rhythm correction category, i.e. advance, delay or interval adjustment of the nursing task execution time; substitution action category, i.e. replacing the system recommended task with a non-recommended task; skipping operation category, i.e. directly skipping part of the system specified steps; sequence rearrangement category, i.e. adjusting the execution sequence or structural arrangement of the task in the path, and setting differentiated identification logic and processing rules for deviation behaviors of different categories, so that the system can respond to adjustment according to the behavior classification results by calling the corresponding path optimization strategy template; then, the system compares the operation behavior of the current nursing staff in the actual execution process with the behavior trajectory of multiple historical nursing staff in similar clinical situations, including patient type, operation stage, physiological state and operation frequency, etc. When the system identifies that most users have the same deviation mode for the same recommended task, especially the same type of correction trend repeatedly appears under similar pathological state, the system will mark the nursing task as a weak recommended item and dynamically down-regulate its priority in the path generation model, and further propose a nursing behavior deviation weight decay function for the fusion calculation of the intensity, importance and user consensus degree of different types of deviation behaviors, so as to obtain the priority adjustment result of the recommended nursing task at the current time point, which is defined as follows:

[0053]

[0054] wherein, is the dynamic priority weight factor of a certain recommended nursing task at the current time , the value range is [0, 1], and the smaller the value is, the more the task should be down-regulated in priority by the system; is the system adjustment coefficient, used to control the sensitivity of the influence of deviation behavior on path strategy, which is a positive real number, and the larger the value is, the higher the influence weight of deviation is; is the number of deviation behavior types associated with the task, at most four, i.e. rhythm correction, substitution action, skipping operation and sequence rearrangement; is the importance weight of the first category of deviation behavior set in the system, used to reflect its influence on the integrity of the path and the safety of nursing; is the deviation intensity of the first category of deviation behavior within the time , which is represented as the deviation degree of the actual execution frequency, amplitude or way compared with the system recommended value, and is normalized; is the deviation amplification index, which controls the nonlinear influence degree of high intensity deviation on the final priority weight, which is usually set as a real number greater than 1; is the behavior consensus factor, which represents the consensus degree of multiple historical users for the first The coincidence degree of the class deviation behavior, with a value range of [0, 1], the greater the value, the more the behavior is convergent, when a deviation behavior has high deviation intensity, high importance weight and significant user consensus, the value will quickly decrease, the system will identify it as a path node with poor strategy adaptability and adjust it preferentially, on the contrary, if the deviation behavior is only occasional, low intensity and lacks user consistency, the value will be maintained at a high level, retaining its recommendation priority, the function can be embedded into the task priority queue of the path generation module as a dynamic sorting weight basis, so that the nursing path has real-time adaptive recommendation updating capability during execution, realizing the closed-loop cooperative adjustment mechanism between nursing behavior, user practice and twin strategy. The coincidence degree of the class deviation behavior, with a value range of [0, 1], the greater the value, the more the behavior is convergent, when a deviation behavior has high deviation intensity, high importance weight and significant user consensus, the value will quickly decrease, the system will identify it as a path node with poor strategy adaptability and adjust it preferentially, on the contrary, if the deviation behavior is only occasional, low intensity and lacks user consistency, the value will be maintained at a high level, retaining its recommendation priority, the function can be embedded into the task priority queue of the path generation module as a dynamic sorting weight basis, so that the nursing path has real-time adaptive recommendation updating capability during execution, realizing the closed-loop cooperative adjustment mechanism between nursing behavior, user practice and twin strategy.

[0055] The identification of the implicit correction behavior of the nursing staff not only based on the comparison between the execution path and the system recommended path, but also further combined with the current physiological state data of the patient and the deviation behavior characteristics for cross judgment, including synchronously collecting the core physiological parameters of the patient (such as heart rate, blood pressure, respiratory rate, muscle tension, oxygen saturation, etc.) after the execution of the nursing task, and performing time corresponding analysis with the identified deviation behavior event, when the system continuously detects that the nursing staff repeatedly avoids, replaces, delays or sequentially adjusts the same type of nursing task under a certain class of specific pathological state, and the physiological state parameter fluctuation of the patient at that time has similarity, the system attributes this deviation behavior to the patient state driven correction behavior, which is different from the path adjustment caused only by the experience preference of the operator, the driven behavior will be preferentially included in the learning sample library of the path generation model, for iterative optimization of the path recommendation mechanism of the system under similar physiological state; in addition, in order to enhance the understanding of the system on the semantic of behavior deviation, when the system identifies that the nursing staff has a stable deviation trend for the same type of task in three consecutive nursing cycles, the nursing terminal pops up an active inquiry prompt, the prompt content includes a brief description of the deviation behavior and its system identification logic, and requests the nursing staff to select the reason option behind the deviation behavior, including experience optimization (indicating that the operation is based on clinical experience adjustment to be more reasonable), patient intolerance (indicating that the patient cannot accept the original recommended task physiologically or psychologically), and equipment limitation (indicating that the task cannot be completed due to the current conditions), the nursing staff can complete the feedback operation through touch selection or voice confirmation, the system will bind this artificial feedback information with the behavior deviation event, mark it as an attribute factor of the deviation behavior, and synchronously transmit it to the path correction module of the digital twin master system, for subsequent accurate attribution, behavior classification and multi-source fusion modeling of the recommended strategy, thereby enhancing the path adaptation ability of the system under the influence of multiple cause variables in complex nursing scenarios, realizing the deep closed-loop learning and dynamic knowledge sedimentation of the digital twin model in actual operation.

[0056] Embodiment 1: ​

[0057] In conjunction with Figure 4In this embodiment, a male patient completed laparoscopic cholecystectomy at 8:30 am, and was transferred to the intelligent nursing ward supported by the digital twin system. The system established a patient individual twin model and started the perioperative nursing rhythm regulation mechanism. At 11:30, the system arranged the first bedside turning-over nursing task. At the same time of task execution, the physiological change monitoring interface started, and the patient's heart rate, systolic pressure, diastolic pressure, muscle tension and oxygen saturation were collected in real time. Five minutes before turning over, the physiological baseline was stable, with a heart rate of 82 beats per minute, a systolic pressure of 125 mmHg, a diastolic pressure of 78 mmHg, an oxygen saturation of 98%, and a muscle tension score of 2 (based on the revised Ashworth scale). After the turning-over nursing action started, within the 7-minute task window, the system recorded that the heart rate rose to 98 beats per minute, the systolic pressure rose to 138 mmHg, the muscle tension increased to level 3, and the oxygen saturation decreased to 95%. The change intensity analysis unit compared each index with the average change rate before the operation, and obtained that the comprehensive fluctuation signal value caused by the nursing task was 2.1 times the baseline rate, between the first threshold (1.5 times) and the second threshold (2.5 times). The buffer control unit judged that the original nursing rhythm could be maintained, and the system entered the observation state. At 12:10, the system arranged the second nursing task - assisting the patient to sit up for respiratory function exercise. This task challenged the patient's abdominal pressure control, balance ability and respiratory system load. After the start of the nursing task, the monitoring data showed that the heart rate rapidly rose to 112 beats per minute, the systolic pressure rose to 151 mmHg, the diastolic pressure was 91 mmHg, the muscle tension reached the upper level of 3, and the oxygen saturation decreased to 93%. The change intensity analysis unit output the fluctuation signal value of 2.7 times, which had exceeded the second threshold. The buffer control unit immediately triggered the task buffer delay instruction, postponed the next task involving abdominal load (such as scrubbing the abdominal belt and changing the drainage bag), and pushed a pop-up window to the nursing terminal through the interactive linkage interface, prompting the performer to delay the execution of the next task. The current patient was in a high response period. Did you agree to postpone for 10 minutes? The nursing staff clicked to accept, and the operation record was uploaded to the twin state update interface. When the third nursing task - bed-side sputum suction and back-patting operation was performed at 13:20, the patient had recovered to a relatively stable state. During the task execution process, the heart rate again rose to 116 beats per minute, the blood pressure rose to 157 / 93 mmHg, the muscle tension jumped to level 4, and the oxygen saturation decreased to 91%. The fluctuation signal value rose to 2.8 times, the system determines that the patient enters a physiological high-load state, immediately starts the rhythm suppression state, suspends all tasks that need physical contact with the upper abdomen or back, and pushes low-load replacement tasks such as environmental adjustment, static psychological pacification, and music intervention in the system nursing path, while recording the continuous high response and the interaction selection results of the nursing staff to the digital twin master system, updating the physiological tolerance down-regulation state label of the patient at 4-6 hours after the operation, adjusting the nursing threshold and rhythm recommendation standard of the patient in future path generation, and forming a closed-loop response mechanism from physiological monitoring, behavioral response, task rhythm control to model state update, effectively reducing the risk of continuous nursing complications.

[0058] When the system identifies that the physiological parameter fluctuation values of its two consecutive nursing tasks exceed the second threshold, it determines that it enters a physiological high-load state and officially starts the rhythm suppression mechanism. At this time, the nursing task buffer control unit activates the built-in task hierarchical reconstruction mechanism to evaluate the bedside full-range body position assistance + abdominal compression fixation belt replacement + cleaning drainage site comprehensive nursing operation planned to be performed at 5 hours after the operation. This task belongs to the high-load task level in the system annotation, involves multiple operation sites, has a long action duration, and involves direct contact with the abdomen. The system calls the preoperative tolerance evaluation and actual monitoring data within 3 hours after the operation in the digital twin model of the patient to form an individual tolerance curve, and accordingly disassembles the task into three subtasks: bed tail-head side position adjustment (expected load level L1), drainage port wiping + dressing replacement (expected load level L2), and fixation belt inspection and tightness adjustment (expected load level L1-L2). The time planning is based on the individual physiological recovery window, and the first subtask is executed at 6 hours, the second subtask is inserted with a buffer recovery period, and the third task is determined whether to remain or delayed by artificial decision based on the physiological response after the execution of the first two. The task disassembly information is transmitted to the nursing terminal through the interactive linkage interface. The A-class nursing staff (experience level E4) responsible for the current nursing task identifies through the identity recognition module, and the system defaults that they have intermediate and advanced intervention authority, so all three options of accepting disassembly suggestions, merging execution, and requesting senior nursing staff to take over are opened in the pop-up window. The A-class nursing staff chooses to accept disassembly and execute in stages, and the system records their behavior path and writes feedback to the twin body state update interface, forming a rhythm optimization node at 6 hours after the operation.

[0059] Meanwhile, the nursing behavior recognition subunit is further enabled during the task execution process, and the starting point of each subtask is confirmed in real time. The system confirms the behavior starting point according to signals such as the starting of the task by the nursing staff, the activation of the hand trigger sensor, the change of the motion tracking image node, and the like, forms a one-to-one correspondence relationship between operation and physiological response, and collects the heart rate of the patient within 5 minutes of the execution of the first subtask from 88 to 91 times per minute, the systolic pressure from 124 to 129 mmHg, the muscle tension maintains level 2, and the blood oxygen maintains 97%. The fluctuation rate is calculated as (91-88) / 5=0.6 times per minute per minute, which is far lower than the 1.5 times critical value of the average rate (0.9 times per minute per minute) in the stable state before the operation, that is, the first threshold value. The system determines that the patient is in a physiological stable state, maintains the current rhythm, and continues to execute the subsequent low-load task. When the second subtask is executed, the system monitors that the heart rate rises to 99 times per minute in a short time, the systolic pressure rises to 137 mmHg, the muscle tension slightly rises, the blood oxygen drops to 95%, and the fluctuation rate is (99-88) / 5=2.2 times per minute per minute. The critical point recognition logic structure identifies that the heart rate change and the blood pressure rise exist in asymmetric rise and present a light non-linear jump curve. The system adjusts the judgment mechanism and increases the originally set second threshold value of 2.5 times to 2.7 times to avoid misjudgment, thereby judging that the light stress state is acceptable, suspending the rhythm suppression mechanism, but retaining the task disassembly path. Finally, the nursing staff completes all subtasks at the 7th hour after the operation according to the nursing terminal prompt, and marks in the completion confirmation interface that the patient accepts the phased nursing well without rejection or subjective discomfort feedback. This information is immediately written into the twin state record as experience feedback of the nursing path, and is used to update the intervention rhythm tolerance parameter in the postoperative recovery model of the patient.

[0060] From the data level, the embodiment shows the real-time linkage capability of multiple modules, for example: the task disassembly is based on specific physiological tolerance value to form a differentiated path, the change rate is calculated based on (Δ physiological value) / Δt, for example, the heart rate change rate is 2.2 times per minute, and the blood pressure rise rate is (137-124) / 5=2.6 mmHg per minute. In the case where neither of them exceeds the second threshold value (about 2.7 times the average rate) adjusted dynamically, the misstart of the rhythm suppression is avoided, the sensitivity and safety redundancy of the system rhythm control are improved, and the nursing behavior recognition subunit establishes a precise mapping chain between the nursing operation and the physiological fluctuation to avoid misassociation and data drift, and ensures that the judgment logic is traceable.

[0061] Embodiment 2:

[0062] In combination with the accompanying Figure 5, based on embodiment 1, at 08:00 on the first day after the operation of the patient, the system generates a set of recommended paths according to the operation type of the patient, the amount of intraoperative bleeding, the postoperative anesthesia recovery performance, and the standard nursing recovery template in the twin model, including the first time of body position assistance turning over at 08:30, the first time of drainage tube outlet disinfection and dressing change at 09:30, encouraging sitting up for abdominal breathing exercise at 10:30, guiding the patient to perform water drinking test at 11:00. The system records the path as a recommended path and uploads it to the ward nursing terminal. Each nursing task in the path contains information such as operation type (such as assisting turning over), recommended execution time point, operation sequence, and recommended frequency (such as breathing training recommended 3 times a day), and embeds task ID for subsequent comparison.

[0063] Nurse B is on duty that day, and her behavior data is collected by the operation record subsystem accessed by the nursing terminal and the action tracking module on the bedside equipment. The system records that she actually completed the body position assistance task at 08:20 in advance and skipped the dressing change operation at 09:30 because there was no obvious leakage of the drainage tube. The second task (drainage dressing change) was performed at 10:00, and the breathing training task was delayed to 11:10, and the water drinking test was also delayed to 11:30. The system compares the recommended path with the actual execution path and identifies multiple deviation behaviors such as task execution time in advance (08:20 instead of 08:30), task sequence dislocation (drainage tube treatment earlier than breathing training), task skipping and re-doing (09:30 task skipped and re-done at 10:00), and overall delay (the last two items are executed in sequence). The system further determines whether a stable operation mode deviation is formed after detecting that three tasks have deviated in sequence.

[0064] To determine whether it is a stable implicit correction behavior, the system calls the nursing path data of patients who received the same operation within nearly 7 days in the same ward, and compares the operation behavior tracks of 5 different nurses at the same postoperative 6-hour stage. It is found that 4 of them appropriately delayed the drainage tube handling task and performed the respiratory training task in advance, and there was a higher frequency of pre-advance behavior of body position assistance task. The system infers that this deviation is not a casual event but an experience correction logic formed by the operation end in long-term practice, determines that the preset trend condition of multiple user convergence deviation on similar task nodes is met, and further identifies the behavior of the nursing staff as an implicit correction behavior of the system strategy; This judgment will be marked into the behavior label system of the nursing path strategy model, and trigger the path adaptation module in the twin model to correct the strategy of low-priority tasks. Further, in the system statistics, the recommended time of the respiratory training task is 10:30, but the average execution time is concentrated between 09:50 and 10:10, with a deviation frequency of 78%, and the execution sequence deviates from the normal behavior in most users, so the system adjusts the recommended execution time window of the task from the original 10:00-10:30 to 09:30-10:00, and closes the forced reminder mechanism of the drainage handling task under the condition of no exudation after operation, making the path recommendation more in line with the actual execution logic of nursing staff.

[0065] Combined with the data in this case, the system collected the task execution sequence of nurse B as T1-turning over (08:20), T2-dressing change (10:00), T3-breathing training (11:10), and T4-water tasting (11:30), which deviated from the recommended path T1 (08:30)→T2 (09:30)→T3 (10:30)→T4 (11:00) by ±40 minutes, ±30 minutes, and ±40 minutes, with a deviation rate of 33%~44%. The system sets the time tolerance threshold for stable deviation recognition as ±15 minutes, and the task sequence dislocation threshold as 1-level dislocation (e.g., the task sequence becomes T1→T3→T2→T4), so this combination meets the constitutive requirements of deviation trend and is marked by the system as a stable implicit correction behavior event; This judgment is used to drive the twin model to directly move the body position assistance time forward when generating the path for this type of patient next time, set the drainage-related tasks as conditional trigger type nursing tasks, and adjust the recommended rhythm parameters of the breathing training task.

[0066] ,The system further enters the implicit correction behavior classification and priority decay calculation phase. In actual execution, nurse B delays the drainage dressing change task from the recommended 09:30 to 10:00 and advances the respiratory training task to 09:50. The system preliminarily identifies that the behavior has pace correction (advance, delay), sequence rearrangement (operation sequence change), and task skipping (the dressing task was originally skipped and then done) and other deviation modes. Then, the system classifies all nursing staff's deviation behaviors into the following four categories: pace correction category (such as advance / delay ≥ 15 minutes), alternative action category (such as replacing the recommended task with other operations), skipping operation category (task not executed and no system confirmation), and sequence rearrangement category (execution sequence different from the recommended logic chain). The system sets corresponding identification logic for each category of deviation: for example, the time error exceeds ± 15 minutes as the judgment threshold for pace deviation, and the longest common subsequence difference > 1 as the criterion for sequence misplacement.

[0067] Compare this deviation data with the historical operation trajectory database in the digital twin system. The system retrieves the nursing trajectories of 162 patients in the same hospital area who underwent the same procedure and were in the 6-hour postoperative stage in the past 90 days. It finds that 128 people (about 79%) in the nursing path have the same task pace adjustment behavior, of which 71% have a pace advance or delay of ± 20 minutes, 63% have a sequence misplacement of more than one level, and 58% have a drainage task skipped and then done. Combined with this high consistency data, the system marks the drainage dressing change task in the morning of the first postoperative day as a weak recommendation item and starts the path priority decay calculation phase.

[0068] At this time, the nursing behavior deviation weight decay function is called:

[0069]

[0070] Enter the following example parameters for calculation:

[0071] : This time, 3 types of deviation behaviors are detected: pace correction ( ), skipping operation ( ), and sequence rearrangement ( );

[0072] , , : The system assesses that pace correction has the greatest impact on path integrity, followed by skipping, and rearrangement is relatively less important;

[0073] , , : represent the deviation intensity, which has been normalized (e.g., a 30-minute advance is 0.75 deviation intensity).

[0074] , , : consensus factor derived from historical data, representing the group consensus of deviation behavior

[0075] : system adjustment coefficient, controlling the sensitivity of deviation response, the recommended range is 2.0~6.0, and 4.0 is the medium sensitivity

[0076] : behavior amplification index, enhancing the nonlinear impact of high-intensity deviation in overall priority calculation, the recommended setting is 1.5~2.5, and 1.8 is a relatively robust value

[0077] Substitute calculation:

[0078]

[0079] In turn, calculate:

[0080] ;

[0081] ;

[0082] ;

[0083] Corresponding to the consensus factor:

[0084] ;

[0085] ;

[0086] ;

[0087] Sum: ;

[0088] Substitute again:

[0089]

[0090] Final result , which indicates that the current recommended priority weight of the task is reduced to about 67% of the original, although it is not completely eliminated by the system, but it has been significantly downgraded in the task scheduling priority ranking, and the task will be in a non-mandatory recommendation state in the subsequent path generation process, only if the nursing staff has spare time or the patient's condition permits. If the same deviation trend continues to be observed and more high-consensus operations occur, the priority value will further decrease.

[0091] On the first day to the second day morning after the operation, the system recommends a total of 4 times of breathing training + abdominal belt inspection + water stimulation combination tasks, with an interval of 6 hours each time. Among the 3 times of execution path performed by nurse B, the breathing training part is delayed or disassembled to the end of the combination task or even performed separately. The system monitoring data shows that the three behaviors all occur in the time window when the patient has mild abdominal distension, systolic pressure > 140 mmHg, and muscle tension score > 3. For example, the recommended task at 21:00 on the first day after the operation is to perform breathing training first (recommended period 21:00-21:15), and then perform abdominal belt inspection (21:15-21:30). The nurse actually delays the breathing training to 21:45 and completes the rest of the operations within 21:00-21:15. The physiological state data collected by the system shows that the patient's heart rate was 98 beats per minute, systolic pressure was 146 mmHg, and muscle tension score was 3.5 (obvious subjective resistance) at 21:00. By 21:45, the systolic pressure decreased to 135 mmHg, the heart rate decreased to 91 beats per minute, and the muscle tension score decreased to 2.5, which is an acceptable state. No decrease in blood oxygen or significant discomfort occurred after task execution.

[0092] Thereafter, in the 03:00 and 09:00 time periods on the second day after the operation, similar paths are replaced by behavior structures that delay the execution of breathing training, and each delay node corresponds to a patient with high blood pressure or muscle tension. The system compares the physiological parameter collection values with the execution path and determines that the deviation behavior has a high consistency of pathological state condition trigger characteristics. The cross-judgment mechanism is started to bind and mark the deviation behavior with the corresponding physiological state. The system sets the trigger rule as follows: under the same pathological state (such as blood pressure > 140 mmHg or muscle tension score ≥ 3), the same deviation behavior occurs for more than three times in a row, which is initially classified as a state-driven modification behavior. The system triggers a pop-up window on the nursing terminal.

[0093] At this time, the system pops up a query prompt on the nursing terminal: the system has found that you have delayed the execution of the breathing training task for the patient for the third time in a row under the condition of elevated blood pressure. Please select a reason to optimize the system recommendation logic and provide three options: ① experience optimization (based on your judgment that it is safer to stabilize blood pressure first and then train), ② patient intolerance (such as the patient actively resisting cooperation), and ③ device limitation (such as the auxiliary air bag device cannot be immediately activated). Nurse B selected ② patient intolerance and added a note that the patient actively requested a delay each time the blood pressure was high and exhibited obvious breath-holding resistance. The system records this feedback and marks it as a subjective attribute of the deviation behavior, and uploads it to the behavior attribution module in the twin main system along with the three behavior-state combinations monitored.

[0094] The system thereby forms three data closed loops: first, a behavior recognition logic chain from task path deviation to behavior type classification to trend confirmation; second, a state correspondence mechanism to establish a causal mapping between physiological parameter dynamic capture and deviation behavior; third, nursing staff interaction feedback and label attribution to provide semantic input for model learning. Finally, the deviation behavior is identified by the system as a patient state-driven corrective behavior and is preferentially included in the strategy relearning of the digital twin model. In future path recommendations, if the system identifies that the patient has a similar blood pressure / muscle tension state, it will delay the recommendation of respiratory training task execution or actively provide an optional flexible execution window, avoiding execution conflicts or patient discomfort caused by rigid sequencing, and improving the adaptability, safety, and naturalness of human-machine collaboration of the recommended path.

[0095] The above shows and describes the basic principles, main features and advantages of the present application. Those skilled in the art should understand that the present application is not limited to the above-mentioned embodiments, and the above-mentioned embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the claimed present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A method for dynamic optimization of perioperative nursing plans driven by digital twins, characterized in that... include: An interventional sensitive area is set that is associated with the patient’s current postoperative status. The interventional sensitive area is defined according to the patient’s surgical type and anatomical location, and has physical limitations that limit the threshold that can be triggered by nursing operations. The physical limitations include changes in body position angle, pressure contact area and duration of action. For each nursing operation, a reaction cycle in the target area is predefined. The time period from the completion of the operation to the recovery of relevant physiological indicators to a steady state is set as the impact period. During the impact period, other nursing tasks that have a physical effect on the same area or its physiologically related areas are prohibited from entering the execution sequence. A damping feedback component is set up and embedded in the nursing feedback channel of the digital twin to receive physiological parameter change signals caused by the execution of nursing tasks. When the rate of change of the signal between consecutive nursing tasks exceeds a preset physiological tolerance threshold, the damping feedback component activates a rhythm slowing mechanism to adjust the execution rhythm of subsequent tasks. A set of manual operation behavior monitoring nodes are set up in the nursing terminal or bed area to record the operation path, action interruption, withdrawal behavior and duration generated by the nursing staff during the execution process. Based on the data collected by the monitoring nodes, an execution behavior deviation judgment unit is constructed to compare with the system's recommended path, which is used to identify the implicit modification behavior of nursing staff to the system strategy. The methods for identifying implicit modification behaviors of nursing staff to system strategies include: The nursing path suggestions generated by the digital twin system are recorded as recommended paths. The paths include the operation type of the nursing task, the recommended execution time, the sequence and frequency information. The action information of the nursing staff when actually performing the nursing task is collected by the acquisition device set on the nursing terminal, the hospital bed or the nursing equipment, including the actual operation time, operation type, execution sequence, and records of task skipping or repeated execution. The actual execution path is compared with the recommended path to identify whether there is a continuous deviation behavior, including task advancement, delay, frequency change, substitution behavior or sequence misalignment; after identifying the above deviation behavior, it is determined whether it constitutes a stable operation mode deviation. If the preset trend conditions are met, it is identified as the nursing staff's implicit correction behavior to the system suggestions. The steps for determining whether the implicit modification behavior constitutes the aforementioned behavior include: The identified deviation behaviors are categorized into rhythm correction, alternative actions, skipped operations, and sequence rearrangement. Identification logic and processing rules are set for different types of deviations, enabling the system to optimize paths based on the classification results. The current nursing staff's performance is compared with the operation trajectories of multiple users in similar situations in the past. When multiple users have similar deviation patterns for the same recommended task, the task is marked as a weak recommendation and its priority is reduced. The identification of the implicit correction behavior includes: The system combines the patient's current physiological state data with the execution of deviation behaviors for cross-judgment. When nursing staff continuously avoid or adjust nursing tasks under a certain pathological state, the deviation behavior is attributed to patient state-driven corrective behavior and given priority for learning. After identifying three consecutive similar deviation behaviors, the system sends an active inquiry prompt to the nursing staff through the nursing terminal, requesting an explanation of the correction reason and allowing the nursing staff to choose from experience optimization, patient intolerance, or device limitation options. The feedback will be used to label the deviation behavior attributes.

2. The method for dynamic optimization of perioperative nursing plans driven by digital twins according to claim 1, characterized in that... The damping feedback component comprises: A physiological change monitoring interface is used to collect real-time physiological change data of patients after the execution of nursing tasks. The data includes heart rate, blood pressure, muscle tone and blood oxygen saturation. The intensity of change analysis unit is connected to the physiological change monitoring interface and is used to identify the intensity of physiological response after the nursing task based on the physiological change data, and output the response fluctuation signal. A nursing task buffer control unit, connected to the change intensity analysis unit, is used to dynamically adjust the execution rhythm of subsequent nursing tasks based on the response fluctuation signal, wherein: When the fluctuation signal is less than the first threshold, maintain the original task rhythm; When the fluctuation signal exceeds the second threshold, a task buffer delay instruction is inserted. When the fluctuation signals of two consecutive tasks both exceed the second threshold, the rhythm suppression state is activated. An interactive linkage interface, connected to the buffer control unit, is used to transmit the task adjustment information to the nursing operation interface in real time and provide interactive options for accepting or rejecting delayed execution. The twin status update interface, connected to the digital twin master system, is used to synchronize the rhythm adjustment results of nursing tasks and the results of manual selection to the twin system in order to update the patient status and acceptable care threshold.

3. The method for dynamic optimization of perioperative nursing plans driven by digital twins according to claim 2, characterized in that... The physiological change monitoring interface includes a nursing behavior recognition subunit, which is used to determine the start time and execution method of nursing behavior and synchronously collect physiological change data to align nursing behavior with physiological response. The change intensity analysis unit includes a critical point identification logic structure, which is used to dynamically adjust the threshold identification standard when a nonlinear change pattern is detected.

4. The method for dynamic optimization of perioperative nursing plans driven by digital twins according to claim 3, characterized in that... The nursing task buffer control unit has a task classification and reconstruction mechanism, which is used to break down the original high-load nursing task into multiple sub-tasks and execute them in stages under rhythm suppression. The breakdown scheme is derived from the individual patient tolerance data in twins. The interactive linkage interface is connected to the nursing staff identification module and is used to dynamically adjust the execution strategy of the task postponement suggestions given by the system according to the executor's level and experience level.

5. The method for dynamic optimization of perioperative nursing plans driven by digital twins according to claim 2, characterized in that... The first threshold in the nursing task buffer control unit is that, within a preset time period after the execution of the nursing task, the rate of change of the physiological parameters does not exceed 1.5 times the average rate of change of the parameters under the preoperative stable state, which is used to determine that the patient is in a physiologically stable state and allows the original nursing task rhythm to be maintained.

6. The method for dynamic optimization of perioperative nursing plans driven by digital twins according to claim 2, characterized in that... The second threshold is defined as the rate of change of the physiological parameter exceeding 2.5 times the average rate of change of the parameter in the preoperative stable state within a preset time period after the execution of the nursing task. This threshold is used to determine that the patient is in a state of high physiological response load, triggering a delayed task execution or rhythm suppression mode. The physiological parameters include heart rate, systolic blood pressure, diastolic blood pressure, blood oxygen saturation, or muscle tone. The rate of change is the magnitude of the numerical change per unit time divided by the length of time.

Images