Intelligent interaction and management system for perioperative nursing information of interventional operation patient

Abstract

The invention relates to the technical field of medical information, in particular to an interventional operation patient perioperative period nursing information intelligent interaction and management system which comprises the steps that through an integrated data collection module, a timeline generation module, a risk identification module, a nursing intervention module and an interaction management module, time axis alignment and structured coding are conducted on scattered physiological data and diagnosis and treatment events; and a dynamic comprehensive nursing timeline is formed. And performing real-time analysis on the timeline based on a preset risk knowledge graph, identifying an abnormal data mode and generating an early warning. The system automatically matches a nursing rule base, pushes individualized intervention suggestions including specific operations, priorities and bases, synchronizes information and collects feedback through a multi-role interaction channel, and realizes closed-loop management. According to the system, fragmented nursing information can be integrated into a continuous dynamic view, and risk early warning and structured nursing actions are directly linked, so that the collaboration, accuracy and timeliness of nursing management are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of medical information technology, in particular to an intelligent interaction and management system for perioperative nursing information of interventional surgery patients. BACKGROUND

[0002] The current perioperative nursing management of interventional surgery patients highly depends on multiple heterogeneous information systems. Nursing records, electronic medical records, anesthesia monitoring, and equipment output data are scattered and stored in different systems, and the formats and time bases are not unified. Nursing staff need to manually search, compare, and splice information between different systems to form an intermittent understanding of the patient's state. This data island phenomenon leads to the inability to automatically and continuously construct a dynamic time view reflecting the overall picture of the patient. The formulation of nursing decisions relies on the experiential integration of fragmented information by medical staff, and lacks systematic tool support.

[0003] Existing patient monitoring and warning systems are mostly based on static threshold alarms for single vital parameters. Such methods break the internal relationship between physiological parameters and cannot correlate physiological changes with concurrent diagnostic and therapeutic behaviors. The warning results are only isolated data point anomalies and cannot point to complex clinical situations or risk patterns composed of multiple events. After the alarm is triggered, specific nursing measures still need to be determined and coordinated by medical staff based on experience, and the system itself does not provide step-by-step intervention guidance based on evidence-based rules directly related to specific risk patterns.

[0004] There is a need for a technical solution that can automatically integrate scattered patient information into a dynamic timeline with clear time sequence and unified structure, and based on this, identify abnormal patterns formed by multi-dimensional data combinations in real time, rather than single parameter abnormalities. Further, the system needs to be able to automatically associate and output structured nursing intervention suggestions while identifying risk patterns, thereby directly and accurately linking risk warning and nursing action. SUMMARY

[0005] The purpose of the present application is to solve the problems existing in the prior art, and to provide an intelligent interaction and management system for perioperative nursing information of interventional surgery patients.

[0006] To achieve the above purpose, the present application adopts the following technical solution: an intelligent interaction and management system for perioperative nursing information of interventional surgery patients, comprising: a data acquisition module for acquiring full-dimensional perioperative physiological data and diagnostic and treatment event data of interventional surgery patients; a timeline generation module for time axis alignment and structured coding processing of the full-dimensional perioperative physiological data and the diagnostic and treatment event data, and generating a dynamic comprehensive nursing timeline of the interventional surgery patients; a risk identification module, based on a preset perioperative risk knowledge graph, performs real-time scanning and analysis on the dynamic comprehensive nursing timeline, identifies a set of abnormal data patterns related to high-risk nursing events, and generates a risk warning signal containing a risk type, associated data points, and a confidence score; a nursing intervention module, for the identified set of abnormal data patterns, calls matching entries in a nursing path rule library, automatically generates and pushes an individualized nursing intervention suggestion list containing nursing operation items to be performed, execution priorities, and reference bases; an interaction management module, establishes a multi-role interaction channel between the interventional surgery patient, nursing staff, and doctors, synchronizes the risk warning signal and the individualized nursing intervention suggestion list through the multi-role interaction channel, and collects feedback execution records from nursing staff and doctors, and updates the feedback execution records to the dynamic comprehensive nursing timeline.

[0007] As a further scheme of the present application, the full-dimension perioperative physiological data and the diagnosis and treatment event data are time axis aligned and structured coded to generate a dynamic comprehensive nursing timeline of the interventional surgery patient, comprising: The full-dimension perioperative physiological data includes real-time vital sign waveforms, continuous test index trends, and drug infusion records, and the diagnosis and treatment event data includes surgery progress markers, complication diagnosis records, and consultation opinion abstracts; The dynamic comprehensive nursing timeline sorts and associates data items from different sources according to a unified timestamp; For each data point in the real-time vital sign waveforms, continuous test index trends, and drug infusion records, a second acquisition timestamp and a data source identifier are attached; For each event in the surgery progress markers, complication diagnosis records, and consultation opinion abstracts, the recorded time of the document or the system entry time is parsed as an event timestamp; A unified time axis with the patient's surgery start time as the zero point is established, and the full-dimension perioperative physiological data and the diagnosis and treatment event data with attached timestamps are mapped onto the unified time axis in chronological order; For data items from different data sources at the same time or adjacent times, the relevance is judged according to the type of the data items, logical links are established for data items with clinical relevance, and the dynamic comprehensive nursing timeline with context association is formed; The dynamic comprehensive nursing timeline is periodically compressed and abstracted to extract key time nodes and data inflection points, forming a timeline abstract view for quick overview.

[0008] As a further scheme of the present application, the preset perioperative risk knowledge graph is loaded, and the perioperative risk knowledge graph defines a plurality of high-risk nursing events, each event being associated with a series of physiological parameter threshold conditions and event logical relationships. The preset perioperative risk knowledge graph is loaded, and the perioperative risk knowledge graph defines a plurality of high-risk nursing events, each event being associated with a series of physiological parameter threshold conditions and event logical relationships; In a sliding time window manner, a newly added data segment in the dynamic comprehensive nursing timeline is traversed; For the data in the current time window, according to the rules in the perioperative risk knowledge graph, whether the physiological parameters exceed the threshold values is compared item by item, and whether the event logical relationship is satisfied is checked; When all the associated conditions of a certain high-risk nursing event are continuously or intermittently satisfied within the time window, it is determined that an abnormal data pattern is identified; The high-risk nursing event type corresponding to the identified abnormal data pattern, the specific data point set triggering the abnormal data pattern, and the confidence score calculated according to the completeness and duration of condition satisfaction are recorded to form the risk warning signal.

[0009] As a further scheme of the present application, the matching item in the nursing path rule library is called for the identified abnormal data pattern set, and an individualized nursing intervention suggestion list is automatically generated and pushed, including: According to the high-risk nursing event type identified in the risk warning signal, an index query is performed in the nursing path rule library; The standard nursing path item corresponding to the high-risk nursing event type is retrieved from the nursing path rule library, and the standard nursing path item includes a set of recommended nursing operations, operation specifications, evaluation frequency and expected targets; The set of nursing operations in the standard nursing path item is adaptively adjusted in combination with the individual information and current clinical state of the interventional surgery patient in the dynamic comprehensive nursing timeline, and the adaptive adjustment includes addition or deletion of operation items and individualized setting of execution parameters; According to the emergency degree of the abnormal data pattern set and the confidence score of the risk warning signal, the adjusted nursing operation items are assigned with execution priorities; The nursing operation items that have been adaptively adjusted and assigned with execution priorities, the corresponding operation specifications and the reference basis are formatted and packaged to generate the individualized nursing intervention suggestion list.

[0010] As a further scheme of the present application, the multi-role interaction channel between the interventional surgery patient, the nursing staff and the doctor is established, including: create a system interaction session for the current responsible caregiver and the supervising physician respectively, said system interaction session being linked to the unique identification of said interventional surgery patient; push the generated risk warning signal and the individualized list of nursing intervention suggestions to the system interaction session interface of the relevant caregivers and physicians in real time; in said system interaction session interface, provide an interactive option for confirming execution, delaying execution or raising objections for each nursing operation item in said list of nursing intervention suggestions; when a caregiver or a physician operates a certain nursing operation item through said interactive option, the system records the operation type, the operator's identity and the operation timestamp to form said feedback execution record; through said system interaction session, support instant messaging between caregivers and physicians on said risk warning signal or specific nursing operation item.

[0011] As a further scheme of the present application, update said feedback execution record to said dynamic comprehensive nursing timeline, including: analyze said feedback execution record to extract the operated nursing operation item, the execution status, the operator information and the time information; in said dynamic comprehensive nursing timeline, locate to the original data point or risk warning trigger point associated with said operated nursing operation item; add a new node representing a nursing intervention event after said original data point or risk warning trigger point, the content of said new node including the description of the nursing operation item, the execution status, the operator information and the corresponding operation timestamp; according to the execution status of the nursing operation item, update the status marker of the related risk warning signal in said dynamic comprehensive nursing timeline; store the updated said dynamic comprehensive nursing timeline and make the updated content synchronously visible in all related said system interaction session interfaces.

[0012] As a further scheme of the present application, it also includes: a nursing pathway optimization module, which periodically extracts completed case data from the system, said case data including the complete said dynamic comprehensive nursing timeline and all said feedback execution records; analyze the correlation between the triggering of said risk warning signal and the subsequent nursing intervention effect in said dynamic comprehensive nursing timeline, and evaluate the actual impact of different nursing intervention measures on the outcome of risk events; identify the high-efficiency nursing operation mode positively correlated with the significant improvement of the patient's state after execution, and the low-efficiency or high-risk nursing operation mode not producing the expected effect after execution or being related to adverse events; According to the analysis result, the corresponding standard nursing path entry in the nursing path rule base is revised, and the revision includes adding, deleting or modifying nursing operation items, adjusting operation specifications or priorities; The revised nursing path rule base is applied to the nursing management and suggestion generation process for subsequent new interventional surgery patients.

[0013] As a further scheme of the present application, the analysis in the dynamic comprehensive nursing timeline includes the correlation between the triggering of the risk warning signal and the subsequent effect of nursing intervention, specifically including: Select multiple cases with the same or similar type of risk warning signal as the analysis sample set; For each analysis sample, all data segments from the risk warning triggering time to the end of the predetermined observation time window are extracted from the dynamic comprehensive nursing timeline; Identify the sequence of nursing interventions recorded in the data segment, i.e. the sequence of nursing operation items recorded in the feedback execution record; Quantitative analysis of patient state indicators at the risk warning triggering time and patient state indicators at the end of the predetermined observation time window, calculating the change amount or outcome classification of the state indicators; Using statistical analysis methods, analyze the correlation between different sequences of nursing interventions and the change amount or outcome classification of patient state indicators, and determine the correlation strength of specific nursing operation mode and clinical outcome.

[0014] As a further scheme of the present application, it also includes: The family rehabilitation guidance module extracts key rehabilitation indicators, uncompleted nursing targets and discharge medication information related to the patient from the dynamic comprehensive nursing timeline when the patient is about to be discharged or transferred to the rehabilitation stage; Based on the extracted information, a structured family rehabilitation plan is automatically generated, which includes daily monitoring items, rehabilitation training content, medication reminders and re-examination precautions; Through the patient mobile terminal application, the structured family rehabilitation plan is pushed to the patient or his family members, and the input interface of plan execution record and symptom feedback is provided in the mobile terminal application; Collect the execution record and symptom feedback data from the patient mobile terminal application as a new data source to supplement the patient's continuity of care record, forming an extended nursing timeline after discharge.

[0015] As a further scheme of the present application, the structured family rehabilitation plan is pushed to the patient or his family members through the patient mobile terminal application, and the input interface of plan execution record and symptom feedback is provided in the mobile terminal application, including: A patient-oriented mobile terminal application interface is designed, in which daily monitoring items, rehabilitation training task lists and medication reminders are clearly displayed; For each task, the interactive functions of checking completion, recording specific values or uploading pictures and videos are provided to record the plan execution; A symptom feedback area is set up, and patients can report discomfort symptoms through preset symptom classification or free text description; The plan execution records submitted by the patient and the symptom feedback data are synchronized in real time or at a fixed time to the hospital's nursing information management system through a secure data transmission protocol; In the system interaction session interface on the hospital side, a dedicated view is provided for nursing staff to view the patient's home rehabilitation execution and symptom feedback, and the nursing staff can send guidance information or reminders to the patient through the dedicated view.

[0016] Compared with the prior art, the advantages and positive effects of the present application are: Through time axis alignment and structured coding processing, a dynamic comprehensive nursing timeline is generated, and the feedback records of nursing execution are updated to the timeline in reverse, building a continuously evolving patient state digital model containing complete nursing process traces. This process realizes the fundamental change of nursing information from scattered and isolated to centralized and integrated, from static record to dynamic closed loop. Instead of scattered reports, nursing staff face a complete, coherent and visual patient care process chart, which not only records "what happened", but also clearly shows "what measures were taken" and "how the state changes after the measures", improving the continuity and accuracy of clinical situation awareness.

[0017] Based on the preset perioperative risk knowledge graph, the dynamic timeline is scanned in real time to identify abnormal data pattern sets, and accordingly the personalized nursing intervention suggestion list is automatically matched from the nursing path rule library. This technology upgrades the risk identification from single-point threshold judgment to multi-dimensional and time-series pattern matching, enabling the system to discover risks hidden in complex data correlations. It establishes an automatic link from "identifying risk patterns" to "recommending evidence-based intervention measures". The output suggestion list clearly specifies the specific operation items, execution priority and basis, enabling nursing response to change from immediate decision-making relying on personal experience to a standardized, interpretable and traceable normative operation guide, improving the timeliness, pertinence and consistency of nursing intervention. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 The timing diagram of the perioperative nursing information intelligent interaction and management system for interventional surgery patients described in the present application; Figure 2 Flowchart for generating individualized nursing intervention suggestion list; Figure 3 A timeline broken line chart for the pre-warning processing state of the postoperative hemorrhage risk of the interventional surgery patient P; Figure 4 A correlation analysis chart for the nursing operation execution quality score and the positive outcome rate; Figure 5 A radar chart for the postoperative rehabilitation multidimensional index of the interventional surgery patient. DETAILED DESCRIPTION

[0019] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not used to limit the present application.

[0020] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only used to facilitate the description of the present application and simplify the description, and therefore cannot be understood as indicating or implying that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0021] Referring to Figure 1 , the data acquisition module is responsible for continuously acquiring multidimensional physiological data and various diagnosis and treatment event data of the interventional surgery patient in the whole perioperative period from various medical devices and information systems. The timeline generation module receives these heterogeneous data, performs time axis alignment and structured coding processing, and thus generates a comprehensive nursing timeline capable of dynamically reflecting the overall situation of the patient. The risk identification module has a built-in perioperative risk knowledge graph, which scans and analyzes the dynamic comprehensive nursing timeline in real time, identifies a set of abnormal data patterns conforming to the characteristics of high-risk nursing events, and generates a risk warning signal with detailed information. The nursing intervention module, upon receiving the risk warning signal, automatically matches and generates an individualized nursing intervention suggestion list containing specific operations, priorities and basis according to the nursing path rule base. The interactive management module is responsible for establishing and maintaining the multi-role interactive channel between the patient, the nursing staff and the doctor, synchronizing the risk and intervention information through the channel, collecting the feedback execution records from the medical staff, and finally updating these key intervention records back to the dynamic comprehensive nursing timeline, forming a nursing closed loop.

[0022] In one embodiment of the present application, the system processes the full-dimension perioperative physiological data and the medical event data acquired by the data acquisition module through the timeline generation module. The full-dimension perioperative physiological data includes real-time vital sign waveforms such as ECG and blood pressure waveforms, continuous test index trends such as blood gas analysis and electrolyte sequences, and drug infusion records such as nitroglycerin infusion records. The medical event data includes surgical process markers such as surgical start and puncture success events, complication diagnosis records such as puncture point hematoma and contrast-induced nephropathy records, and consultation opinion summaries such as cardiovascular medicine consultation conclusions. To construct a dynamic comprehensive nursing timeline, the timeline generation module attaches a time stamp accurate to seconds and a data source identifier to each data point in the real-time vital sign waveforms, the continuous test index trends, and the drug infusion records. For the surgical process markers, the complication diagnosis records, and the consultation opinion summaries in the medical event data, the timeline generation module analyzes the time of the document record or the system entry time as the event timestamp. The timeline generation module establishes a unified time axis with the patient's surgical start time as the zero point, and maps all the full-dimension perioperative physiological data and medical event data carrying timestamps onto the unified time axis in chronological order. The mapping process is performed by the formula:

[0023] wherein: represents the relative time of the data point on the unified time axis, represents the original timestamp of the data point, represents the absolute timestamp of the patient's surgical start time. In specific implementations, the timeline generation module performs relevance judgment on data items from different data sources at the same time or adjacent times, establishes logical links for data items with clinical relevance, such as establishing a logical link between a blood pressure drop data point at a certain time and an increased nitroglycerin infusion rate data point recorded earlier, and forms a dynamic comprehensive nursing timeline with contextual association. In some embodiments, the timeline generation module periodically compresses and generates summaries of the dynamic comprehensive nursing timeline, extracts key time nodes and data inflection points, and forms a timeline summary view for quick overview, such as extracting the time of surgical key steps and the time of abnormal vital sign fluctuations from a timeline lasting for several hours. Optionally, the timeline generation module uses a sliding window average algorithm to retain trend characteristics during compression. It can be understood that the generation of the dynamic comprehensive nursing timeline depends on accurate timestamp alignment and structured coding.

[0024] In a specific implementation, in connection with an example scenario, a patient undergoing a percutaneous coronary intervention procedure generates multi-source data during the procedure, real-time vital sign waveforms including electrocardiogram waveforms show a heart rate rising from 70 beats per minute to 95 beats per minute, blood pressure waveforms show a systolic pressure dropping from 120 millimeters of mercury to 100 millimeters of mercury, continuous laboratory test trend data including arterial blood gas analysis shows a partial pressure of oxygen dropping from 90 millimeters of mercury to 80 millimeters of mercury, medication infusion records show nitroglycerin being pumped in at a rate of 10 micrograms per minute, procedure event data includes a procedural progress marker record of“balloon dilation” being completed at a certain time, and a complication diagnosis record of“vagal reflex” being marked at a later time. The timeline generation module attaches a millisecond-level acquisition timestamp and an electrocardiogram monitor data source identifier to each heartbeat data point of the electrocardiogram waveforms, attaches an acquisition timestamp and a non-invasive blood pressure monitoring device identifier to each measurement value of the blood pressure waveforms, attaches a laboratory information system identifier and a test system entry timestamp to the arterial blood gas analysis result, attaches a medication infusion pump record timestamp and a medication administration system identifier to the nitroglycerin infusion record, attaches a procedural documentation record timestamp to the“balloon dilation” event, and attaches an electronic medical record system entry timestamp to the“vagal reflex” event. The timeline generation module sorts and maps the converted data according to relative time, and in the mapping process, the timeline generation module detects that the blood pressure drop data point and the nitroglycerin infusion increase data point are close in time, establishes a logical link indicating that the blood pressure drop can be associated with the drug effect. In some embodiments, the timeline generation module displays the original data scattered in different systems and different time bases in the data comparison, while the processed dynamic comprehensive care timeline provides a sequence view arranged according to a unified time axis. The original electrocardiogram data and the procedural event data timestamp can be off by several minutes, and the dynamic comprehensive care timeline eliminates the deviation by aligning through the unified time axis. Optionally, the timeline generation module identifies the blood pressure drop inflection point and the“vagal reflex” event as key nodes for compressed display when generating the timeline summary view. It can be understood that the dynamic comprehensive care timeline provides a coherent time sequence by integrating multi-source data.

[0025] Referring to Figure 2In an embodiment of the present application, the risk identification module loads a pre-set perioperative risk knowledge graph, which defines a plurality of high-risk nursing events such as postoperative bleeding and acute heart failure, each of which is associated with a series of physiological parameter threshold conditions and event logical relationships. The risk identification module iterates over the newly added data segments in the dynamic comprehensive nursing timeline in a sliding time window manner. For the data in the sliding time window, the risk identification module compares whether the physiological parameters exceed the threshold according to the rules in the pre-set perioperative risk knowledge graph, and checks whether the event logical relationship is satisfied, for example, the rule may require “heart rate continuously greater than 100 beats per minute” and “systolic blood pressure continuously less than 90 mmHg” and “hemoglobin decreased by more than 2 g / dl within 3 hours”. When all the associated conditions of a high-risk nursing event are continuously or intermittently satisfied within the sliding time window, the risk identification module determines that an abnormal data pattern is identified, records the high-risk nursing event type corresponding to the identified abnormal data pattern, the specific data point set triggering the abnormal data pattern, and the confidence score calculated according to the completeness and duration of condition satisfaction, and finally forms a risk warning signal. In some embodiments, the confidence score by the formula:

[0026] wherein: represents the confidence score of the risk warning signal, the value range is 0-1, the higher the score represents the higher the accuracy of risk identification, represents the total number of conditions associated with the high-risk nursing event, represents the th condition, represents the pre-set weight of the th condition, the function returns the satisfaction degree of the th condition in the sliding time window (the value range is 0 to 1), the function is an abnormal pattern duration-based gain function. Upon receiving the risk warning signal, the nursing intervention module indexes and queries the nursing path rule base according to the high-risk nursing event type identified in the risk warning signal. The nursing intervention module retrieves the standard nursing path entry corresponding to the high-risk nursing event type from the nursing path rule base, which contains a set of recommended nursing operations, operation specifications, evaluation frequency, and expected targets. The nursing intervention module combines the individual information and current clinical state of the patient undergoing surgery in the dynamic comprehensive nursing timeline to adaptively adjust the set of nursing operations in the standard nursing path entry, including the addition or deletion of operation items and the individualization of execution parameters. For example, for a patient with renal dysfunction, the "monitor urine output" item is added to the standard nursing operation set, or the infusion rate parameter of the "intravenous fluid infusion" item is adjusted. The nursing intervention module assigns execution priorities to the adjusted nursing operation items according to the urgency of the abnormal data pattern set and the confidence score of the risk warning signal. The nursing intervention module formats and packages the adjusted nursing operation items, corresponding operation specifications, and reference basis to generate an individualized nursing intervention suggestion list.

[0027] In a specific implementation, in combination with an example scenario, the dynamic comprehensive nursing timeline shows that, in the time period of 2-3 hours after the surgery, the heart rate of patient A continuously increases from 72 beats per minute to 115 beats per minute, the systolic blood pressure gradually decreases from 118 mmHg to 82 mmHg, and the hemoglobin test value decreases from 12.5 g / dl to 10.2 g / dl within 3 hours. The risk identification module sets the sliding time window to 30 minutes, and when traversing the time window containing the above data period, the module calls the preset perioperative risk knowledge graph rule about the "postoperative bleeding" event, which contains three conditions: "heart rate greater than 100 beats per minute", "systolic blood pressure less than 90 mmHg", and "hemoglobin decrease greater than 2 g / dl within 3 hours". The risk identification module compares the data, and the heart rate and blood pressure data satisfy the first two conditions at multiple consecutive time points, and the hemoglobin decrease value satisfies the third condition. The risk identification module determines that the abnormal data pattern related to "postoperative bleeding" is identified, records the event type as "postoperative bleeding", and triggers the data point set as the above specific heart rate, blood pressure, and hemoglobin data points. The risk identification module calculates the condition satisfaction degree, the first two conditions are completely satisfied in the window, and the third condition is satisfied, and calculates the confidence score. In comparison, the heart rate of patient B in the same time period is 88 beats per minute, the blood pressure is 102 / 65 mmHg, and the hemoglobin is stable. After the risk identification module traverses the sliding time window, it is determined that there is no matching abnormal data pattern. After receiving the "postoperative bleeding" risk warning signal for patient A, the nursing intervention module indexes the nursing path rule base, retrieves the standard nursing path entry corresponding to "postoperative bleeding", which contains a set of nursing operations and operation specifications such as "check the puncture point", "monitor vital signs", "review blood routine", "establish two intravenous access", and "rapid fluid infusion". The nursing intervention module adjusts the standard nursing operation set adaptively in combination with the individual information recorded in the dynamic comprehensive nursing timeline of patient A, which is 75 years old and has a history of chronic heart failure, adds the "monitor central venous pressure" operation item, and individualizes and lowers the initial infusion rate parameter of "rapid fluid infusion". The nursing intervention module allocates the highest priority to "check the puncture point" and "establish two intravenous access" according to the emergency degree of increased heart rate and decreased blood pressure, and allocates the next highest priority to "review blood routine".

[0028] In one embodiment of the present application, the interaction management module creates a system interaction session for each of the caregivers currently responsible for the interventional procedure patient care and the supervising physician, respectively, the system interaction session is associated to the unique identification of the interventional procedure patient. The interaction management module pushes the risk alert signal generated by the risk identification module and the individualized nursing intervention suggestion list generated by the nursing intervention module to the system interaction session interface of the relevant caregivers and physicians in real time. In the system interaction session interface, each of the nursing operation items in the individualized nursing intervention suggestion list is provided with an interactive option for confirming execution, delaying execution or raising objection. When the caregivers or physicians operate on a certain nursing operation item through the interactive option of the system interaction session interface, the interaction management module records the operation type, the operator identity and the operation timestamp, forming a feedback execution record. In the system interaction session interface, the interaction management module supports the caregivers and physicians to initiate instant messaging of text or voice on the risk alert signal or a specific nursing operation item. For the collected feedback execution records, the interaction management module analyzes and extracts the operated nursing operation item, the execution status, the operator information and the time information. In the dynamic comprehensive nursing timeline, the interaction management module locates to the original data point or the risk alert trigger point associated with the operated nursing operation item. After the original data point or the risk alert trigger point, the interaction management module adds a new node representing a nursing intervention event, the content of the new node includes the description of the nursing operation item, the execution status, the operator information and the corresponding operation timestamp. According to the execution status of the nursing operation item, the interaction management module updates the state marker of the relevant risk alert signal in the dynamic comprehensive nursing timeline. The interaction management module stores the updated dynamic comprehensive nursing timeline and makes the updated content synchronously visible in all associated system interaction session interfaces.

[0029] In a specific implementation, in combination with an example scenario, the patient P generates a “postoperative hemorrhage” risk warning signal and an individualized nursing intervention suggestion list containing items such as “check puncture point”, “establish two venous access”, etc. The interaction management module creates system interaction sessions for the nurse A responsible for the patient P and the supervising doctor B, and both system interaction sessions are associated with the medical record number of the patient P. The interaction management module pushes the warning signal and the suggestion list to the mobile nursing terminal interface of the nurse A and the computer workstation interface of the doctor B at the same time. In the interface of the nurse A, the “establish two venous access” item in the list is displayed with three buttons of “confirm execution”, “delay execution” and “raise objection”. The nurse A clicks the “confirm execution” button. The interaction management module records the operation type as “confirm execution”, the operator as “nurse A”, and the operation timestamp as “2026-02-12 14:25:30”, forming a feedback execution record about the “establish two venous access” item. At the same time, the doctor B sees the same list on the interface and clicks “raise objection” on the “electrocardiogram monitoring level” item and inputs “please confirm whether to upgrade to continuous monitoring” in the text box. The interaction management module records this operation and generates another feedback execution record. In data comparison, at the same time, the “establish venous access” item in the interface of the nurse A changes to “executed”, while the item in the interface of the doctor B is displayed as “executed by the nurse A”; for the “electrocardiogram monitoring level” item, the interface of the nurse A displays “doctor B raises objection: please confirm...”, and the interface of the doctor B displays “objection sent”. In the timeline integration part, the interaction management module analyzes the feedback execution record about “establish two venous access”, extracts the nursing operation item as “establish two venous access”, the execution status as “completed”, and the operator as “nurse A”. In the dynamic comprehensive nursing timeline, the interaction management module locates the original data points of “blood pressure drop” and “heart rate increase” that triggered the “postoperative hemorrhage” warning, and adds a new node after these points. The content of the new node is described as “nursing intervention: establish venous access (2)”, the execution status is “completed”, the operator is “A nurse”, and the operation timestamp is “14:25:30”. The interaction management module updates the state marker of the “postoperative hemorrhage” risk warning signal in the dynamic comprehensive nursing timeline from “new warning” to “processing” according to the completion of the “establish venous access” operation. In some embodiments, the logic of the interaction management module to update the state marker is based on the formula:

[0030] wherein: represents the updated risk warning signal state value (1 represents released), represents the state value before updating, represents the number weight of “completed” operations in this feedback execution record, is a preset attenuation coefficient according to the risk type. The updated entire dynamic comprehensive nursing timeline is saved, and the timeline view embedded in the system interaction session interface of the nurse A and the doctor B is synchronously refreshed to display the newly added nursing intervention node and the updated warning state. In terms of communication and integration, the nurse A sees the dissent text proposed by the doctor B for the "electrocardiogram monitoring level" in the interface, initiates a voice communication request to the doctor B through the instant messaging function built in the system interaction session interface, and after communication, the doctor B changes the "propose dissent" operation to "confirm execution" in the interface. The interaction management module records the change to form a new feedback execution record. Optionally, the system interaction session interface can set different color highlights for different types of operations (such as confirmation, delay). It can be understood that the multi-role interaction channel realizes the synchronization and closed loop of information. In specific implementation, compared with the scene without the system, the nurse needs to inform the doctor of the warning by telephone, the doctor orally issues the medical order, and the nurse writes down the execution and then transcribes, and the information flow is broken; and in the embodiment, the warning, list, operation, feedback and timeline update are all real-time synchronized in the unified interface to form a structured record.

[0031] Referring to Figure 3 This is a postoperative bleeding risk warning processing state timeline broken line graph of the patient P after the interventional surgery, which directly shows the dynamic processing process of the "postoperative bleeding" risk warning of the patient P from 14:00 to 14:55 on February 12, 2026. The whole process from risk warning triggering, intervention execution to risk removal is clearly presented, verifying the effectiveness of the system closed loop management. Through the change of the state value, the contribution of different nursing operations to the risk control can be quantitatively evaluated, providing data support for optimizing the nursing path. The real-time update of the risk state ensures the information synchronization between the nurse and the doctor, avoiding the information delay and break in the traditional mode. The time span from risk triggering to removal, the curve slope and the like can be used as key indicators for nursing quality monitoring, for continuously improving the clinical service level.

[0032] In one embodiment of the present application, the care pathway optimization module periodically extracts completed case data from the system, the completed case data containing a complete dynamic integrated care timeline and all feedback execution records. The care pathway optimization module selects multiple cases with the same or similar type of risk warning signal as the analysis sample set. For each analysis sample in the analysis sample set, the care pathway optimization module extracts all data segments from the dynamic integrated care timeline from the risk warning trigger time to the end of the predetermined observation time window. The care pathway optimization module identifies the sequence of care interventions recorded in the data segment, which is the sequence of care operation items recorded in the feedback execution record. The care pathway optimization module quantitatively analyzes the patient state indicators at the risk warning trigger time and at the end of the predetermined observation time window, calculates the change in state indicators or outcome classification. The care pathway optimization module uses statistical analysis methods to analyze the correlation between different sequences of care interventions and the change in patient state indicators or outcome classification, and determines the association strength between specific care operation modes and clinical outcomes. Through the above analysis, the care pathway optimization module identifies high-efficiency care operation modes that are positively correlated with significant improvement in patient state after execution, and inefficient or high-risk care operation modes that do not produce the expected effect or are associated with adverse events after execution. The care pathway optimization module revises the corresponding standard care pathway entries in the care pathway rule library based on the analysis results, including adding, deleting or modifying care operation items, adjusting operation specifications or priorities. The care pathway optimization module applies the revised care pathway rule library to the care management and recommendation generation process for subsequent new interventional surgery patients.

[0033] In a specific implementation, in combination with an example scenario, the care pathway optimization module extracts all cases that have triggered a "contrast-induced nephropathy" risk warning signal from discharged patients as the analysis sample set in each monthly analysis period, and the analysis sample set includes case A, case B, and case C. The care pathway optimization module extracts all data segments from the dynamic integrated care timeline of case A from the "contrast-induced nephropathy" warning trigger time (6 hours after surgery) to the end of the predetermined observation time window (72 hours after surgery). The care pathway optimization module identifies the sequence of care interventions recorded in the data segment, and the sequence of care interventions for case A is "increase intravenous infusion to 150 ml / hour", "monitor urine output every 2 hours", and "schedule postoperative 6-hour and 24-hour review of renal function". The care pathway optimization module quantitatively analyzes the patient state indicator serum creatinine at the risk warning trigger time, which is 110 micromoles / liter, and the patient state indicator serum creatinine at the end of the predetermined observation time window, which is 115 micromoles / liter, and calculates the change in state indicators, which is an increase of 5 micromoles / liter. The care pathway optimization module performs the same operation on all cases in the analysis sample set, as shown in Table 1.

[0034] Table 1: Correlation Analysis between Nursing Intervention Sequence and Changes in Patient Creatinine

[0035] In some embodiments, the nursing pathway optimization module uses statistical methods to analyze correlations and calculate the strength of the correlation between a specific nursing operation mode "use of acetylcysteine" and changes in creatinine. The nursing pathway optimization module may optionally employ a formula... Calculate a reference value for the association strength, where Reference values ​​representing the strength of the association between specific nursing care practices and positive outcomes. This represents the number of cases in the analyzed sample set that contain this operating mode. Representing the The standardized values ​​of patient status indicators that show a positive change in a particular case (e.g., creatinine decreases to a positive value). Representatives based their work on evidence-based medicine regarding the first... The nursing procedure execution quality score was calculated for each case. Based on data analysis, the nursing pathway optimization module identified the nursing procedure pattern "increase intravenous fluids; monitor urine output; re-examine renal function; administer acetylcysteine" as positively correlated with a decrease (improvement) in creatinine, indicating it was an efficient nursing procedure pattern; while the pattern "increase intravenous fluids; monitor urine output" was associated with an increase (worsening) in creatinine, indicating it was an inefficient nursing procedure pattern. Based on this analysis, the nursing pathway optimization module revised the standard nursing pathway entry corresponding to "contrast-induced nephropathy" in the nursing pathway rule base, adding the item "assess and administer acetylcysteine ​​(if no contraindications)" to the nursing procedure items and assigning it a higher execution priority. The revised nursing pathway rule base will be invoked when the next newly admitted patient at risk of contrast-induced nephropathy triggers an alert. In terms of data comparison, the old rule base entries did not include recommendations related to "acetylcysteine," while the new revised entries do.

[0036] See Figure 4 This is a correlation analysis chart showing the relationship between nursing procedure execution quality scores and positive outcome rates. It reveals a strong positive correlation between nursing procedure execution quality and patient positive outcome rates, serving as a key visualization tool for assessing the effectiveness of clinical interventions. The equation of the red trend line in the chart is y = 1.44x. The slope of the curve is 52.09, indicating that for every 1 point improvement in nursing procedure quality, the positive outcome rate for patients increases by an average of approximately 1.44 percentage points. All data points closely follow the trend line, demonstrating a strong correlation between the two, indicating that nursing quality is a key factor influencing patient outcomes. This provides quantitative evidence for the conclusion that "improving nursing procedure quality can significantly improve patient prognosis," which can be directly used to support nursing pathway optimization and resource allocation decisions. Nursing procedure quality scores can be used as a core KPI, and monitoring changes in these scores can predict patient outcome trends, enabling proactive quality management.

[0037] In one embodiment of the present application, the home rehabilitation guidance module is triggered when the interventional surgery patient is about to be discharged or transferred to the rehabilitation stage. The home rehabilitation guidance module extracts the key rehabilitation indicators, uncompleted care targets and discharge medication information related to the interventional surgery patient from the dynamic comprehensive care timeline. Based on the extracted key rehabilitation indicators, uncompleted care targets and discharge medication information, the home rehabilitation guidance module automatically generates a structured home rehabilitation plan, which includes daily monitoring items, rehabilitation training content, medication taking reminders and re-visit notes. The home rehabilitation guidance module pushes the structured home rehabilitation plan to the interventional surgery patient or his / her family members through the patient mobile terminal application, and provides a plan execution record and symptom feedback input interface in the patient mobile terminal application. The home rehabilitation guidance module collects the execution record and symptom feedback data from the patient mobile terminal application, and supplements the execution record and symptom feedback data as a new data source to the continuity of care record of the interventional surgery patient, forming an extended care timeline after discharge.

[0038] In specific implementation, the home rehabilitation guidance module designs a mobile terminal application interface for the patient, in which the daily monitoring items to be completed, rehabilitation training task list and medication reminders are clearly displayed. The home rehabilitation guidance module provides interactive functions such as checking completion, recording specific values or uploading pictures and videos for each task in the mobile terminal application interface, which are used to record the plan execution. The home rehabilitation guidance module sets a symptom feedback special area in the mobile terminal application interface, through which the interventional surgery patient can report discomfort symptoms through pre-set symptom classification or free text description. The home rehabilitation guidance module synchronizes the plan execution record and symptom feedback data submitted by the interventional surgery patient to the hospital's nursing information management system in real time or at a fixed time through a secure data transmission protocol. In the system interaction session interface at the hospital end, the home rehabilitation guidance module provides a dedicated view for the nursing staff to view the home rehabilitation execution and symptom feedback of the interventional surgery patient, and supports the nursing staff to send guidance information or reminders to the interventional surgery patient through the dedicated view.

[0039] With the example scenario, patient Y is about to be discharged after coronary intervention, and the dynamic integrated care timeline records its latest key rehabilitation indicators including serum creatinine 105 micromole / liter, morning blood pressure 135 / 85 millimeters of mercury, and unfinished care targets including "control systolic pressure below 130 millimeters of mercury", and discharge medication information including "aspirin enteric-coated tablets 100 milligrams once a day", "clopidogrel 75 milligrams once a day", and "metoprolol succinate tablets 47.5 milligrams once a day". The home rehabilitation guidance module extracts the above information and automatically generates a structured home rehabilitation plan. The structured home rehabilitation plan includes daily monitoring items "morning blood pressure measurement", "heart rate measurement", and "body weight measurement", rehabilitation training content "walk twice a day, 15-20 minutes each time", drug taking reminders "aspirin after breakfast", "clopidogrel at a fixed time every day", "metoprolol in the morning", and re-consultation matters "outpatient review one month after surgery, carry this record". The home rehabilitation guidance module pushes this plan through a special application installed on the patient Y's smart phone. In the application interface, patient Y sees the daily tasks clearly displayed in the form of a calendar and a list, such as "Monday: measure blood pressure (enter the value), walk (check completion), take aspirin (check completion)". Patient Y clicks the blood pressure monitoring task, enters the value "132 / 84", clicks the walking task, checks the completion, clicks the drug taking task, checks the completion and uploads the medicine box photo. The application interface has a symptom feedback area, and patient Y reports a slight discomfort through the preset classification "chest pain" on the third day after discharge, and supplements the description "slight chest tightness after walking, relieved after rest" in the text box. The home rehabilitation guidance module synchronizes the blood pressure value 132 / 84 submitted by patient Y, task completion check record, medicine box photo, and chest tightness symptom feedback data to the hospital's nursing information management system through an encrypted transmission protocol. These data are added as new data sources to patient Y's continuity of care record, forming a care timeline extending from the day of discharge, which includes "on the third day after surgery, patient reports slight chest tightness after walking, self-measured blood pressure 132 / 84". At the hospital end, the nursing staff responsible for patient Y logs in to the system and sees a "home rehabilitation" exclusive view in the system interaction session interface, which displays the completed blood pressure record, task check status, and reported "chest tightness" symptoms of patient Y in the form of a time stream. The nursing staff sends a guidance information "chest tightness symptoms have been known, please continue to observe, and contact in time if the pain is aggravated or does not subside" to patient Y through the exclusive view. In data comparison, the traditional way without this system is to issue paper discharge guidance, and patient compliance and symptom feedback are difficult to track; in this embodiment, patient Y's structured execution data and active symptom feedback are systematically recorded and integrated, and the nursing staff can provide remote guidance based on real-time data.

[0040] Reference Figure 5This is a radar chart of multidimensional indicators of postoperative rehabilitation of patients undergoing interventional surgery, from 1-3 days to 8-14 days after surgery. The overall trend of each rehabilitation indicator is continuously improving, especially blood pressure control, exercise capacity and symptom management, with the most significant improvement. The abstract rehabilitation indicators are converted into intuitive radar charts, allowing medical staff and patients to quickly grasp the overall picture of rehabilitation and identify strengths and weaknesses. The effectiveness of rehabilitation interventions at different stages is clearly demonstrated, verifying the effectiveness of the system in home rehabilitation guidance, medication adherence management, etc. By comparing the improvement rate of each dimension, the rehabilitation plan can be adjusted accordingly, such as strengthening intervention in exercise capacity and symptom management. The continuous improvement of each dimension provides quantitative evidence for the conclusion that the system effectively promotes postoperative recovery, and strengthens the innovation and clinical value of the technical solution.

[0041] The above is only a preferred embodiment of the present application, not other forms of limitations on the present application, any skilled in the art can use the above disclosed technical content to change or modify equivalent embodiments for other fields, but any simple modification, equivalent change and modification made to the above embodiments without departing from the technical solution content of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. An intelligent interactive management system for perioperative nursing information of interventional surgery patients, characterized in that, The system includes: The data acquisition module collects comprehensive perioperative physiological data and diagnostic and treatment event data from patients undergoing interventional surgery. The timeline generation module performs timeline alignment and structured encoding on the full-dimensional perioperative physiological data and the diagnostic and treatment event data to generate a dynamic comprehensive nursing timeline for the interventional surgery patient. The risk identification module, based on a preset perioperative risk knowledge graph, performs real-time scanning and analysis on the dynamic comprehensive nursing timeline, identifies abnormal data pattern sets related to high-risk nursing events, and generates risk warning signals that include risk type, associated data points, and confidence scores. The nursing intervention module, for the identified abnormal data pattern set, calls the matching entries in the nursing path rule base to automatically generate and push an individualized nursing intervention suggestion list, which includes the nursing operation items to be performed, the execution priority and reference basis; The interactive management module establishes a multi-role interaction channel between the interventional surgery patient, nurses, and doctors. Through the multi-role interaction channel, the risk warning signal and the individualized nursing intervention suggestion list are synchronized, and feedback execution records from nurses and doctors are collected and updated to the dynamic comprehensive nursing timeline.

2. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 1, characterized in that, The perioperative physiological data and the diagnostic and treatment event data are time-axis aligned and structured to generate a dynamic comprehensive nursing timeline for the interventional surgery patients, including: The comprehensive perioperative physiological data includes real-time vital sign waveforms, continuous test index trends, and drug infusion records; the diagnostic and treatment event data includes surgical progress markers, complication diagnosis records, and consultation opinion summaries. The dynamic integrated nursing timeline sorts and associates data items from different sources according to a unified timestamp; For each data point in the real-time vital signs waveform, continuous test index trend, and drug infusion record, append a second-level acquisition timestamp and data source identifier. For each event in the surgical process marker, complication diagnosis record, and consultation opinion summary, parse the document time or system entry time recorded therein, and use it as the event timestamp; Establish a unified timeline with the patient's surgery start time as zero, and map the full-dimensional perioperative physiological data with timestamps and the diagnostic and treatment event data onto the unified timeline in chronological order; For data items from different data sources at the same or adjacent times, the correlation is determined according to the data item type, and logical links are established for data items with clinical relevance to form the dynamic comprehensive nursing timeline with contextual relevance. The dynamic integrated nursing timeline is periodically compressed and summarized to extract key time nodes and data inflection points, forming a timeline summary view for quick overview.

3. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 2, characterized in that, The system, based on a pre-defined perioperative risk knowledge graph, performs real-time scanning and analysis of the dynamic comprehensive nursing timeline to identify a set of abnormal data patterns related to high-risk nursing events, including: Load the preset perioperative risk knowledge graph, which defines a variety of high-risk nursing events, and each event is associated with a series of physiological parameter threshold conditions and event logic relationships; The newly added data segments in the dynamic integrated nursing timeline are traversed using a sliding time window. For the data within the current time window, according to the rules in the perioperative risk knowledge graph, each physiological parameter is compared to see if it exceeds the threshold, and the logical relationship of the event is checked to see if it is satisfied. When all the associated conditions for a high-risk nursing event are met continuously or intermittently within a time window, an abnormal data pattern is identified. The risk warning signal is formed by recording the high-risk nursing event type corresponding to the identified abnormal data pattern, the specific set of data points that triggered the abnormal data pattern, and the confidence score calculated based on the completeness and duration of the condition.

4. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 3, characterized in that, For the identified set of abnormal data patterns, matching entries in the nursing pathway rule base are invoked to automatically generate and push a personalized list of nursing intervention suggestions, including: Based on the high-risk nursing event type identified in the risk warning signal, an index query is performed in the nursing pathway rule base; Retrieve standard nursing pathway entries corresponding to the high-risk nursing event type from the nursing pathway rule base. The standard nursing pathway entries include a suggested set of nursing operations, operation specifications, assessment frequency, and expected goals. Based on the individual information and current clinical status of the interventional surgery patients in the dynamic integrated nursing timeline, the nursing operation set in the standard nursing pathway items is adaptively adjusted. The adaptive adjustment includes adding or removing operation items and setting individualized execution parameters. Based on the urgency of the abnormal data pattern set and the confidence score of the risk warning signal, an execution priority is assigned to the adjusted nursing operation items; The nursing procedures that have been adaptively adjusted and assigned execution priorities, along with their corresponding operating procedures and the reference data, are formatted and packaged to generate the individualized nursing intervention suggestion list.

5. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 4, characterized in that, The establishment of a multi-role interaction channel between the interventional surgery patient, nursing staff, and doctor includes: Create system interaction sessions for the currently responsible nurse and the attending physician, and associate each system interaction session with a unique identifier for the interventional surgery patient; The generated risk warning signal and the individualized nursing intervention suggestion list will be pushed to the system interaction interface between relevant nursing staff and doctors in real time. In the system's interactive session interface, for each nursing operation item in the nursing intervention suggestion list, interactive options are provided to confirm execution, delay execution, or raise objections; When a nurse or doctor performs a nursing operation through the interactive options, the system records the operation type, operator identity, and operation timestamp to form the feedback execution record. The system's interactive sessions enable nurses and doctors to initiate real-time text or voice communications regarding risk warning signals or specific nursing procedures.

6. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 5, characterized in that, Updating the feedback execution record to the dynamic integrated nursing timeline includes: The feedback execution record is analyzed to extract the nursing operation items, execution status, operator information, and time information. In the dynamic integrated nursing timeline, the original data points or risk warning trigger points associated with the nursing operation items being performed are located; After the original data point or risk warning trigger point, add a new node representing the nursing intervention event. The content of the new node includes a description of the nursing operation item, the execution status, the operator information, and the corresponding operation timestamp. Update the status markers of relevant risk warning signals in the dynamic integrated nursing timeline according to the execution status of nursing operation items; The updated dynamic integrated care timeline is stored and made visible synchronously in all associated system interaction sessions.

7. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 6, characterized in that, Also includes: The nursing pathway optimization module periodically extracts completed case data from the system. The case data includes the complete dynamic integrated nursing timeline and all feedback execution records. The correlation between the triggering of the risk warning signal and the effect of subsequent nursing interventions was analyzed in the dynamic integrated nursing timeline, and the actual impact of different nursing interventions on the outcome of risk events was evaluated. Identify efficient nursing procedures that are positively correlated with significant improvement in patient condition after execution, and inefficient or high-risk nursing procedures that do not produce the expected results or are associated with adverse events after execution. Based on the analysis results, the corresponding standard nursing pathway entries in the nursing pathway rule base are revised. The revisions include adding, deleting, or modifying nursing operation items, and adjusting operation specifications or priorities. The revised nursing pathway rule base will be applied to the nursing management and suggestion generation process for subsequent new interventional surgery patients.

8. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 7, characterized in that, The analysis of the correlation between the triggering of the risk warning signal and the effectiveness of subsequent nursing interventions within the dynamic integrated nursing timeline specifically includes: Multiple cases with the same or similar types of risk warning signals were selected as the analysis sample set; For each analysis sample, extract all data segments from the time the risk warning is triggered to the end of the predetermined observation time window from the dynamic integrated nursing timeline; Identify the sequence of nursing interventions recorded in the data segment, i.e., the sequence of nursing procedures recorded in the feedback execution record; Quantitatively analyze the patient status indicators at the time the risk warning is triggered, and the patient status indicators at the end of the predetermined observation time window, and calculate the change in status indicators or outcome classification. Statistical analysis methods were used to analyze the correlation between different nursing intervention sequences and changes in patient status indicators or outcome categories, and to determine the strength of the association between specific nursing operation modes and clinical outcomes.

9. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 8, characterized in that, Also includes: The home rehabilitation guidance module extracts key rehabilitation indicators, unfulfilled nursing goals, and discharge medication information related to the patient from the dynamic comprehensive nursing timeline when the patient is about to be discharged or transferred to the rehabilitation stage. Based on the extracted information, a structured family rehabilitation plan is automatically generated, which includes daily monitoring items, rehabilitation training content, medication reminders, and precautions for follow-up visits. The structured family rehabilitation plan is pushed to the patient or their family through the patient's mobile terminal application, and the mobile terminal application provides an input interface for plan execution records and symptom feedback. Execution records and symptom feedback data from the patient's mobile terminal application are collected and used as a new data source to supplement the patient's continuing care record, forming an extended care timeline after discharge.

10. The intelligent interactive management system for perioperative nursing information of interventional surgery patients according to claim 9, characterized in that, The structured home rehabilitation plan is pushed to the patient or their family through a patient mobile terminal application, and the mobile terminal application provides an input interface for plan execution records and symptom feedback, including: Design a mobile application interface for patients, clearly displaying the daily monitoring items, rehabilitation training task list, and medication reminders. Each task is given interactive functions such as checking the completion box, recording specific values, or uploading pictures and videos to record the execution status of the plan; Set up a symptom feedback area where patients can report their discomfort symptoms through preset symptom categories or free text descriptions; The patient's submitted plan execution record and symptom feedback data are synchronized to the hospital's nursing information management system in real time or at regular intervals via a secure data transmission protocol. In the system interaction interface at the hospital, nursing staff are provided with a dedicated view to view the patient's home rehabilitation progress and symptom feedback, and nursing staff can send guidance information or reminders to patients through this dedicated view.

Images