ar-based intraoperative anesthetic crisis intelligent emergency guidance system

Abstract

The present application relates to the field of medical auxiliary and augmented reality technology, in particular to an intraoperative anesthesia emergency intelligent emergency guidance system based on AR, which comprises an emergency identification module, a situation awareness module, a drug tracking module and an AR presentation module; the emergency identification module completes the grading and accurate determination of anesthesia emergencies and the triggering of emergency procedures based on the membership function and sliding window smoothing processing; the situation awareness module collects the available state of rescue resources in the operating room in real time; the drug tracking module generates a standardized emergency drug record; the AR presentation module is built-in with a minimum action set dynamic guidance engine and a drug trajectory marking engine for collaborative work, which can dynamically generate and render the minimum set of rescue actions currently to be performed, while generating a spatially anchored drug visual mark. The present application can realize the whole-process intelligent emergency guidance of intraoperative anesthesia emergencies, and improve the standardization, timeliness and team collaboration efficiency of rescue disposal.

Description

Technical Field

[0001] This invention relates to the field of medical assistance and augmented reality technology, and more specifically, to an AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies. Background Technology

[0002] Intraoperative anesthetic emergencies are high-risk events that occur suddenly during clinical anesthesia. They are characterized by rapid onset, rapid progression, and high mortality. They are one of the core causes of perioperative patient mortality. The standardization and timeliness of emergency treatment directly determine the patient's prognosis and are a key area of ​​control in the fields of anesthesiology and clinical emergency medicine.

[0003] Current emergency response solutions for intraoperative anesthesia emergencies include attempts to use AR glasses to overlay vital signs / crisis lists. However, these solutions suffer from several technical shortcomings, such as the inability to dynamically minimize steps based on resources and dependencies, and the lack of spatial tracking and dosage risk linkage for medication administration. Firstly, emergency identification is susceptible to environmental interference from intraoperative electrosurgical units and sensor malfunctions, resulting in a high false trigger rate and hindering the ability to tiered assessment and process adaptation from suspected to confirmed cases. Secondly, existing guidance solutions primarily present standardized, complete emergency procedures, failing to dynamically adjust based on real-time personnel, equipment, and medication resources in the operating room, leading to information overload and interference with the anesthesiologist's core resuscitation actions. Thirdly, the lack of spatialized, visualized, and end-to-end tracking of emergency medication administration makes it difficult to intuitively present medication history and cumulative dosage risks, potentially causing information asynchrony and duplicate medication during team resuscitation. Fourthly, the inability to quickly synchronize the status of newly admitted resuscitation personnel can delay crucial resuscitation time.

[0004] To address the numerous shortcomings of the existing technologies, this invention provides an AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of this invention is to provide an AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies.

[0006] To achieve the above objectives, the present invention provides the following technical solution: An AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies includes an emergency identification module, a situational awareness module, a drug tracking module, and an AR presentation module. The emergency identification module is used to collect the patient's vital signs parameters and image data during the operation in real time. Based on the pre-configured feature patterns corresponding to various anesthetic emergencies, it calculates the confidence level of the corresponding anesthetic emergency event and completes the determination of anesthetic emergencies and the triggering of emergency guidance process according to the preset confidence level threshold. The situational awareness module is used to collect data in real time on the identity and availability of medical staff, the location and operating status of emergency equipment, and the storage location and inventory status of emergency medicines in the operating room scene, and generate a dataset of the availability status of rescue resources in the current scene. The drug tracking module is used to collect and record drug information, administration information and operator information for intraoperative emergency medications in real time, and generate standardized medication records. The AR presentation module is communicatively connected to each of the aforementioned modules and includes a minimum action set dynamic guidance engine and a drug trajectory imprinting engine that work together. The minimum action set dynamic guidance engine is used to dynamically generate and render only the minimum set of emergency actions that need to be executed at the moment, based on the determined emergency type and the available status dataset of emergency resources. The drug trajectory imprinting engine is used to generate spatially anchored drug visual imprints based on standardized medication records and complete the rendering.

[0007] Furthermore, the minimum action set dynamic guidance engine is pre-configured with a complete resuscitation action library corresponding to various anesthesia emergencies. Each action in the action library is configured with personnel requirement parameters, equipment requirement parameters, execution attribute parameters, and prerequisite dependencies. The engine calculates the current executability of each action based on the resuscitation resource availability dataset, generates a set of actions that can be executed immediately through topological sorting based on the prerequisite dependencies between actions, and prioritizes the actions in the set, rendering only the highest priority actions as the minimum resuscitation action set.

[0008] Furthermore, the minimum action set dynamic guidance engine calculates the comprehensive priority score of an action based on the prognostic relevance weight of the action itself and the blocking effect of the action on subsequent dependent actions, and completes the priority classification of the action according to the comprehensive priority score.

[0009] Furthermore, the drug trajectory imprinting engine is pre-configured with visual coding rules corresponding to different emergency drug categories. The engine matches the corresponding visual code according to the standardized medication record, generates a visual imprint containing the core information of the drug, anchors the visual imprint at the three-dimensional spatial coordinates of the drug administration site through visual SLAM technology, and performs dynamic decay processing on the visual imprint over time.

[0010] Furthermore, the drug trajectory imprinting engine generates efficacy assessment markers at the corresponding drug visual imprints based on the patient's vital sign changes after medication; at the same time, the engine counts the cumulative doses of drugs of the same category, and triggers the warning rendering of the corresponding visual imprints when the cumulative dose approaches or exceeds the preset safety limit.

[0011] Furthermore, the emergency identification module configures corresponding membership functions for various signs and characteristics of different types of anesthetic emergencies, calculates the instantaneous confidence of the emergency event based on the membership of each sign parameter, obtains the stable confidence through a sliding window smoothing process, and completes the suspected emergency determination, confirmed diagnosis determination, and guides the exit process based on the comparison results of the stable confidence and a preset threshold.

[0012] Furthermore, the situational awareness module calculates the number of available rescue personnel based on the type and availability of medical staff, filters the set of available emergency medical equipment based on the accessibility and operating status of the equipment, calculates the patient risk index based on the patient's vital signs parameters, and generates the dataset of available rescue resources.

[0013] Furthermore, the AR presentation module performs spatial grid division and density detection on the rendered content of the two engines to complete the anti-occlusion collaborative layout of the rendered elements; when a new medical staff member is detected entering the operating room scene, the AR presentation module automatically triggers the highlighting of the visual imprint and the prominent rendering of the minimum rescue action set to complete the rapid synchronization of the rescue status.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves accurate and interference-resistant classification of intraoperative anesthetic emergencies by combining an emergency identification module with a confidence calculation method based on membership functions and sliding window smoothing. Simultaneously, through a minimum action set dynamic guidance engine, based on emergency type, real-time resuscitation resource status, and action pre-dependencies, it dynamically generates and renders only the highest priority resuscitation action currently required by topological sorting and comprehensive priority calculation. This technical solution effectively avoids the problem of misjudgment of emergencies caused by intraoperative environmental interference, while also solving the information overload defect of existing fixed emergency guidelines. This allows anesthesiologists to focus on core resuscitation actions throughout the process, significantly improving the timeliness and standardization of emergency treatment for anesthetic emergencies. 2. This invention achieves spatial anchoring visualization of emergency medication through the collaborative operation of a drug tracking module and a drug trajectory imprinting engine, based on visual SLAM technology. It generates visual drug imprints through pre-configured visual coding rules and combines them with a time decay model, efficacy assessment markers, and a cumulative dosage safety warning mechanism to intuitively present information about the entire medication process. At the same time, through AR-rendered spatial anti-occlusion collaborative layout and a rapid synchronization mechanism for the rescue status of newly arrived personnel, it achieves seamless information synchronization of the rescue team, effectively avoiding medical safety risks such as duplicate medication and dosage overdose, and significantly improving the coordination efficiency of multi-person collaborative rescue. Attached Figure Description

[0015] Figure 1 This is a block diagram of an AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies. Figure 2This is a flowchart illustrating the implementation of the minimum action set dynamic guidance engine of this invention. Figure 3 This is a flowchart illustrating the implementation of the drug trajectory imprinting engine of the present invention. Detailed Implementation

[0016] Example, refer to Figure 1 The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies in this embodiment includes an emergency identification module, a situational awareness module, a drug tracking module, and an AR presentation module. These modules work together to achieve dynamic guidance and visualization of medication history in intraoperative anesthesia emergency scenarios. The specific implementation of the emergency identification module involves real-time acquisition of the patient's vital signs parameters, including heart rate (HR), systolic blood pressure (SBP), diastolic blood pressure (DBP), mean arterial pressure (MAP), and blood oxygen saturation, via a medical device interface. End-tidal carbon dioxide Sampling frequency for vital signs such as body temperature (T) is no less than 100Hz. The data is determined based on the fastest change cycle of intraoperative vital signs being no less than 10ms. 100Hz sampling can completely capture the details of vital sign changes and avoid the loss of key features. In view of the special nature of intraoperative anesthetic emergencies, the system has pre-set judgment logic for various anesthetic emergencies. For intraoperative anaphylactic shock, when SBP < 90 mmHg and HR > 100 bpm are simultaneously met, and the airway pressure suddenly increases by more than 10 cm above the baseline value, it is considered a anaphylactic shock. When the above symptoms are accompanied by signs of urticaria on the skin, it is considered a suspected case of anaphylactic shock; for malignant hyperthermia, when... Sustained high fever lasting more than 5 minutes with a T-ratio rate exceeding 0.5℃ / 15min and the presence of muscle rigidity is considered a suspected case of malignant hyperthermia. For local anesthetic poisoning, altered consciousness, arrhythmia, epileptic-like movements, and a recent history of local anesthetic injection are considered a suspected case of local anesthetic poisoning. For cardiac arrest, the absence of a valid heart rate for three consecutive cardiac cycles and the disappearance of the invasive arterial pressure waveform are considered a suspected case of cardiac arrest. For severe bradycardia, a sustained heart rate below 40 bpm accompanied by a mean arterial pressure below 60 mmHg for more than 5 seconds is considered a suspected case of severe bradycardia. To handle noise and fluctuations in vital sign data and avoid false triggers caused by transient interference such as intraoperative electrosurgical interruptions or sensor malfunctions, this module employs a confidence calculation method based on membership functions. For each type of anesthetic emergency E, the system pre-defines its corresponding feature pattern, represented by a set of membership functions: for intraoperative anaphylactic shock, the membership function for low blood pressure is... Defined as: ; In the formula, x is the real-time collected systolic blood pressure (SBP) value of the patient, in mmHg; the value of this function ranges from 0 to 1, and the higher the value, the more the systolic blood pressure matches the hypotension characteristics of anaphylactic shock. Membership function of fast heart rate Defined as: ; In the formula, x is the real-time heart rate (HR) value of the patient, in bpm; the value of this function ranges from 0 to 1, and the higher the value, the more the heart rate matches the tachycardia characteristics of anaphylactic shock. Membership function of airway pressure elevation Defined as: ; In the formula, x is the difference between the current airway pressure and the baseline value, in cm. The function takes values ​​from 0 to 1. A higher value indicates that the increase in airway pressure is more consistent with the airway spasm characteristics of anaphylactic shock. Membership function of signs of urticaria Defined as: ; In the formula, x is the confidence level of urticaria signs output by AR camera image recognition, with a value range of 0 to 1, and is output by the image recognition model; the value range of this function is 0 to 1, and the higher the value, the more the skin signs are consistent with the characteristics of anaphylactic shock; For malignant hyperthermia, the membership function of myotonia is... Defined as: ; In the formula, x is the confidence level of masseter muscle tension or limb stiffness output by AR camera image recognition, with a value range of 0 to 1, and is output by the image recognition model; the value range of this function is 0 to 1, and the higher the value, the more the muscle rigidity signs are consistent with the characteristics of malignant hyperthermia. For local anesthetic poisoning, the membership function of altered consciousness Defined as: ; In the formula, x represents the confidence level of the patient's response to the call command, as monitored by the AR device, and its value ranges from 0 to 1. A higher value indicates a better state of consciousness for the patient. The function's value ranges from 0 to 1, and a higher value indicates that the change in consciousness is more consistent with the characteristics of local anesthetic poisoning. Membership function of epileptic-like movements Defined as: ; In the formula, x is the confidence level of the epileptic-like action output by the AR camera image recognition, with a value range of 0 to 1, and is output by the action recognition model; the value of this function ranges from 0 to 1, and the higher the value, the more the action characteristics are consistent with the manifestation of local anesthetic poisoning; Instantaneous confidence level for anesthetic emergency type E at the current moment Calculated as a weighted sum of the membership degrees of each parameter: ; In the formula, Here are the weighting coefficients for each parameter, satisfying... The value is determined based on the diagnostic specificity of the corresponding parameter for anesthetic emergency type E, and the weight is positively correlated with the specificity. Let be the membership function value corresponding to the k-th trait parameter at time t. Let be the real-time collected value of the kth vital sign parameter at time t; In this embodiment, the weighting coefficients for various anesthetic emergencies are as follows: Anaphylactic shock: Blood pressure weighting Take 0.3, heart rate weight Take 0.2, airway pressure weight Take 0.3, skin sign weight Take 0.2; Malignant hyperthermia: The weights are set to 0.4 for body temperature, 0.3 for muscle rigidity, and 0.3 for other signs. Local anesthetic poisoning: altered consciousness weighted at 0.4, arrhythmia weighted at 0.2, epileptic-like movements weighted at 0.3, and history of local anesthetic injection weighted at 0.1; Cardiac arrest: The weight for loss of effective heart rate is 0.6, and the weight for loss of invasive arterial pressure waveform is 0.4; Severe bradycardia: Heart rate weighted at 0.6, mean arterial pressure weighted at 0.4; To prevent false alarms caused by single noise fluctuations, the system uses a sliding window averaging method to calculate the stability confidence score. Let the time window length be W; in this implementation, W = 5 seconds. This value is chosen because the duration of intraoperative interference signals typically does not exceed 3 seconds, and a 5-second window can effectively filter out single interferences while ensuring the timeliness of emergency assessment. Therefore, the stability confidence score is... The average instantaneous confidence level over the past W seconds: ; In the formula, t is the integration time variable, in seconds; t is the current time, in seconds; W is the sliding window length, in seconds. In practical discrete sampling systems, a weighted moving average is used to achieve this: ; In the formula, The smoothing factor is set to 0.3. The value is determined by balancing the response speed and smoothing effect of the confidence level. 0.3 can take into account both the timeliness of emergency identification and the ability to resist interference. The sampling interval is 0.01 seconds, matching the 100Hz sampling frequency; t represents the current time in seconds. when Exceeding the threshold When the value is 0.7, the system determines it as a suspected anesthetic emergency and triggers the guidance process; when Exceeding the threshold When the value is 0.9, the system automatically confirms the emergency event and upgrades the urgency level of the guidance; when Falling back to If the following conditions are met and the condition persists for more than 10 seconds, the system will automatically exit the emergency guidance mode. The above threshold values ​​are based on the confidence requirements for the diagnosis of clinical anesthesia emergencies. 0.7 is the minimum confidence level for a suspected diagnosis, and 0.9 is the confidence level standard for a confirmed diagnosis.

[0017] The situational awareness module is implemented by collecting real-time environmental and resource information about the operating room, with a particular focus on key resources related to intraoperative anesthesia emergencies. Specific data collected includes: the number and identification of medical personnel in the operating room, with a focus on the real-time location and activity status of anesthesiologists, anesthesia nurses, surgeons, and circulating nurses; the availability of anesthesia emergency equipment, including the location of the ambulance, the charging status of the defibrillator, the location of the difficult airway cart, and the working status of the suction device; and the inventory status of anesthesia emergency medications, including the remaining quantity of medications in each drawer of the ambulance and the quantity of spare medications in the anesthesia medication cabinet. The specific implementation method is as follows: the number and identity of personnel are counted through the facial recognition or employee badge recognition function of AR devices, with special marking of anesthesiology personnel; the location and status of equipment such as ambulances, defibrillators, ventilators, suction devices, and difficult airway carts are detected through Bluetooth beacons or RFID tags, and the status of defibrillators is read through Bluetooth to read their self-test status and charging completion status; the storage location and remaining quantity of emergency drugs are read through barcode scanning or RFID, and key drugs for anesthetic emergencies such as adrenaline, atropine, ephedrine, norepinephrine, and dantrolene are subject to key monitoring; This module updates the collected information in real time and uses it for the minimum action set dynamic guidance engine; the collected data includes: the number of available anesthesia-related personnel. Collection of available anesthesia and emergency medical equipment Patient's current vital signs and risk index wait; Number of available anesthesia-related personnel The calculation formula is: ; In the formula, The weights for the m-th medical staff member are as follows: anesthesiologists take a weight of 1.0, anesthesia nurses take a weight of 0.8, surgeons take a weight of 0.3, and circulating nurses take a weight of 0.2. The values ​​are based on the different types of staff members’ ability and authority to participate in emergency anesthesia care. Let be the availability status coefficient of the m-th medical staff member. It is 1.0 when the staff member is idle, 0.2 when the staff member is busy, and 0 when the staff member is unreachable. The availability status is determined by the location and action recognition of the AR device. When the staff member is performing an operation unrelated to the rescue, the staff member is determined to be busy. Collection of available anesthesia and emergency medical equipment The criteria for determining whether the equipment is usable are: the equipment is within the reach of the operating room, the self-test status is normal, and the conditions for emergency use are met; the defibrillator is determined to be usable when it is fully charged and has no faults in the self-test, the ventilator is determined to be usable when it is in standby mode and the supporting tubing is complete, the difficult airway equipment is determined to be usable when the complete set of instruments is complete and in a sterile standby state, and the suction device is determined to be usable when the negative pressure is normal and the tubing connection is intact. Patient's current vital signs risk index The calculation formula is: ; In the formula, n is the number of vital sign parameters involved in the calculation; This represents the real-time value of the k-th vital sign parameter; This is the midpoint of the normal range; The value is half the normal range; the index ranges from 0 to 1, and the higher the value, the more unstable the patient's vital signs are, and it is used for dynamic adjustment of the priority of subsequent actions.

[0018] The specific implementation of the drug tracking module includes: real-time recording of complete information on all intraoperative anesthesia emergency medications, with a focus on key drugs used in emergency anesthesia care; automatic identification of drug type when anesthesiologists or nurses scan drug barcodes with a barcode scanner or recognize drug packaging images using an AR camera; recording the three-dimensional coordinates of the administration site, such as intravenous catheter puncture points and central venous puncture points, using the spatial positioning function of the AR device; recording the injection time using the system clock; and recording the executor's identity using the user identification function of the AR device. Due to the specific nature of anesthetic emergencies, the system specifically marks the following drug categories: vasoactive drugs, including adrenaline, noradrenaline, ephedrine, dopamine, etc.; anticholinergic drugs, including atropine, glycopyrronium bromide, etc.; anesthetic antagonists, including naloxone, flumazenil, etc.; and emergency special drugs, including dantrolene, fat emulsions, etc. This information is packaged into a single medication record in the following format: ,in This field is used for subsequent color coding and dose accumulation; This module is also responsible for transmitting medication records in real time to the drug trajectory imprinting engine of the AR presentation module, which is used to generate visual imprints.

[0019] The AR presentation module is the core of this invention and comprises two collaborative subsystems: a minimal action set dynamic guidance engine and a drug trajectory imprinting engine. These two subsystems share the same AR visual rendering channel, dynamically adjusting the displayed content based on the current intraoperative anesthesia emergency scenario to ensure that only the most urgently needed information is presented in the anesthesiologist's field of vision.

[0020] S51, Minimal Action Set Dynamic Guidance Engine: This engine does not show the anesthesiologist the complete standard emergency procedure, but dynamically calculates and displays only the minimum action set that must be performed at the moment based on the intraoperative conditions, including personnel, equipment, and patient status. like Figure 2 As shown, firstly, the system predefines a complete action library for each type of anesthetic emergency; let the type of anesthetic emergency be E, and its corresponding complete action library be... Each of the actions It includes the following attributes: action name, required number and type of personnel (e.g., whether a second anesthesiologist is needed), required equipment set, execution attribute parameters, correlation weight with patient prognosis, and set of prerequisite dependent actions; for example, for intraoperative anaphylactic shock, the complete action library includes: Stop administering suspected drugs Inject adrenaline, Provide 100% oxygen Call an additional anesthesiologist. Establish a second intravenous access 500ml rapid intravenous infusion Prepare endotracheal intubation equipment, etc.; the execution attribute parameters include, but are not limited to: the estimated execution time of the action, the required skill level (such as attending physician, resident physician, nurse), and the urgency of the action (such as immediate execution or execution later); for example, for the injection of adrenaline, the estimated execution time is 10 seconds, the required skill level is an anesthesiologist, and the urgency level is the highest. These parameters are pre-defined by anesthesiology experts according to clinical guidelines and are used for subsequent feasibility calculations and priority ranking. Secondly, based on real-time data collected by the situational awareness module, the system calculates the current executability index for each action; for each action... The required number of personnel is Among them, the number of anesthesiologists has a higher weighting, and the required equipment set is: Then the action Current executability index The calculation is as follows: ; In the formula, Let be the weighting coefficient, satisfying In this embodiment, the default value is... The value is determined based on the fact that personnel and equipment are the core prerequisites for action execution, while time cost is a secondary influencing factor; for time-sensitive emergencies such as severe high fever, the value can be increased. A value up to 0.5 increases the priority of actions that execute quickly; if ,but Taking 1 directly indicates that the personnel requirement is fully met, because no personnel are needed. The cardinality of a set is the number of elements in the set. For action Estimated execution time, in seconds. This represents the maximum estimated execution time for all actions in the action library, in seconds. All three terms in this formula are dimensionless values, ranging from 0 to 1. The value ranges from 0 to 1, with a higher value indicating that the action is easier to execute under the current conditions. Next, the system constructs an action execution sequence diagram based on the logical dependencies between actions; for each action... and ,like Must If executed previously, then defined for The system considers prerequisite actions; for example, establishing intravenous access must be performed before rapid infusion, and stopping suspected medication must be performed before administering an antagonist. Based on these dependencies, the system uses a topological sorting algorithm to generate a set of actions that can be executed immediately. The specific implementation steps are as follows: 1. Traverse all actions in the action library, count the number of pre-dependent actions for each action, and denote it as the in-degree; 2. Add actions with an in-degree of 0 to the execution queue; 3. Take an action from the queue to be executed and add it. Set the set, and decrement the in-degree of all subsequent dependent actions of this action by 1; 4. If the in-degree of a subsequent dependent action decreases to 0, add it to the pending execution queue; 5. Repeat steps 3 to 4 until the queue to be executed is empty; 6. If there are still actions not added after the traversal is completed. For a set, if a circular dependency exists, ignore the actions that are circularly dependent and only add actions that are independent. gather; Finally, the system... The actions in the process are prioritized and divided into three levels: Level 1 actions, which must be executed immediately; Level 2 actions, which are recommended to be executed; and Level 3 actions, which are optional to be executed. The priority ranking is based on the prognostic relevance weight of the actions. And the degree to which an action blocks subsequent actions, specifically, for actions... Its overall priority score The calculation is as follows: ; In the formula, For action The prognostic relevance weight, ranging from 0 to 1, is pre-defined by anesthesiologists according to clinical guidelines and is positively correlated with the degree of influence of the action on the patient's prognosis; for example, in anaphylactic shock, the effect of administering adrenaline... Take 1.0 and administer 100% oxygen. Take 0.9 and prepare the endotracheal intubation equipment. Take 0.5; Indicates dependence The set of all subsequent actions that can be executed; This is the damping coefficient, with a value of 0.5. The value is chosen to balance the importance of the current action itself and its blocking effect on subsequent actions, so as to avoid over-amplifying the priority of the preceding action. The rules for classifying action levels are as follows: 1. When When the number of actions in a set is ≥3, the 1-2 actions with the highest overall priority score are Level 1 actions, the 2-3 actions with the second highest scores are Level 2 actions, and the rest are Level 3 actions; 2. When When there are 2 actions in a set, the action with the highest score is a level 1 action, and the other one is a level 2 action; 3. When When the number of actions in a set is 1, the action is a level 1 action, and there are no level 2 or level 3 actions; The method for determining whether an action has been completed is as follows: 1. For medication-related actions, the medication tracking module receives the corresponding medication record, which determines that the action has been completed. 2. For equipment operation actions, the status change signal of the corresponding equipment is received through the situational awareness module, and the action is determined to be completed. 3. For actions that do not receive feedback from equipment, such as personnel calls or pathway establishment, the action is considered complete by the anesthesiologist's voice confirmation command or by the AR camera recognizing the image features of the corresponding action. In anesthesia emergencies, the AR presentation module only renders and displays the primary action at the current moment, i.e., the minimum action set. For example, in the case of anaphylactic shock, if only one anesthesiologist is present and the ambulance has not yet arrived, the minimum action set may only show stopping the infusion of the suspected drug and administering 100% oxygen. After the nurse delivers adrenaline, the system automatically updates the administration of adrenaline to the primary action and displays it in the AR field of view. Secondary and tertiary actions are completely hidden and are only shown as very faint prompts when the anesthesiologist actively queries them or at the edge of the AR field of view. After the primary action is completed, the system re-executes the above calculation process and dynamically updates the minimum action set. S52. Drug Tracking Engine: This engine presents the history of each intraoperative anesthetic emergency medication in the AR field of view as a spatially anchored visual imprint, allowing anesthesiologists to intuitively see what happened in the past. It is specifically optimized for the characteristics of repeated and combined medication in anesthetic emergencies. like Figure 3 As shown, firstly, the system defines visual coding rules for each category of anesthetic drugs. Drug categories are divided according to their pharmacological effects in intraoperative emergency rescue: vasoactive drugs (red, RGB value 255,0,0); anticholinergic drugs (green, RGB value 0,255,0); sedative-analgesics (blue, RGB value 0,0,255); muscle relaxants (yellow, RGB value 255,255,0); anesthetic antagonists (purple, RGB value 128,0,128); and emergency special drugs (orange, RGB value 255,165,0). The color codes are loaded during system initialization and stored in the drug category-color mapping table. For dantrolene, a specific drug for malignant hyperthermia, a white flashing halo is added to the orange color, with a flashing frequency of 2Hz, to indicate its specialness and scarcity. When the drug tracking module records a new intraoperative medication event, the system generates a visual imprint based on the event information; the imprint generation process includes the following steps: Step 1: Obtain the medication event data packet, including the drug name. Drug categories ,dose Injection time Executor ID 3D coordinates of the drug administration site For anesthetic emergencies, the administration site is usually the location of an established intravenous access. Step 2: According to drug category Look up the corresponding base color from the drug category-color map table. ; Step 3: Calculate the initial brightness of the imprint. The initial brightness is set to the maximum value of 1.0; Step 4: Generate the visual content of the imprint; the imprint uses a floating label format, and the label content includes: abbreviated drug name, such as EPI for adrenaline, ATR for atropine, dosage value, injection time (only hours and minutes are displayed), and the abbreviation of the executor's surname; the label background color is [color missing]. The text color is white. The label is surrounded by a […]. A faint halo of the same color, with its radius slowly pulsating over time at a frequency of 0.5 Hz, simulates the rhythmicity of vital signs, and the pulsation amplitude is ±20% of the halo's base radius; Step 5: Anchor the generated imprint to the three-dimensional coordinates of the drug administration site. The system employs visual SLAM technology for anchoring, combined with pre-set static markers within the surgical scene for position correction. These static markers are fixed to equipment with fixed positions, such as the operating table and anesthesia machine. When the patient's position changes, the system uses an AR camera to track the position of the intravenous access puncture point in real time, dynamically correcting the anchoring coordinates of the imprint. This ensures that even if the anesthesiologist moves their head or the patient's position is slightly adjusted, the imprint always remains in the correct spatial position, precisely corresponding to the patient's intravenous access location. The system performs dynamic decay processing on existing imprints. For imprint j, its generation time is... The current time is The current brightness of the imprint The calculation is as follows: ; In the formula, The initial brightness is set to 1.0. The decay time constant is set to 600 seconds in this embodiment. The value is based on the fact that the golden time window for emergency anesthesia rescue is usually 10 to 15 minutes. 600 seconds corresponds to 10 minutes, which is consistent with the short-term memory pattern of anesthesiologists for medication events in clinical rescue. Every 10 minutes, the imprint brightness decays to 36.8% of the original value. It is a natural exponential function; and The units are all seconds; when When the imprint is below the threshold of 0.1, it completely disappears from the field of vision. This exponential decay model simulates the short-term memory pattern of anesthesiologists for intraoperative medications, making the newest medications the most prominent and the earlier medications gradually fade out, forming an intuitive time gradient. At the same time, the size of the imprint also decreases linearly with brightness: ,in The base size is 5cm x 2cm, which is the actual physical size, to ensure that it does not obstruct the key puncture point or surgical field. The actual display size of the imprint at time t, in cm; The system also generates a synchronous imprint for the storage location of each medication on the anesthesia ambulance, with the anchor position being the coordinates of the medication's storage location. When an anesthesiologist or nurse retrieves a medication from the ambulance, the imprint for that medication's storage location will flash briefly at a frequency of 2Hz for 2 seconds, indicating that the medication has been retrieved. This feature effectively prevents duplicate medication retrieval during chaotic anesthesia emergencies. New anesthesia personnel can determine that a medication has been used and its current inventory may be empty if they see a decaying imprint for a medication's location on the ambulance. Based on the physiological response data after injection, the system automatically generates a pharmacodynamic assessment marker next to the imprint. Considering the characteristics of anesthetic emergencies, it monitors changes in key physiological parameters within 30 seconds after injection and calculates the deviation between the actual and expected responses. Specific assessment rules and thresholds are as follows: 1. After administering vasopressors, including adrenaline, noradrenaline, ephedrine, and dopamine, monitor changes in SBP and HR. If the SBP increases by less than 10 mmHg, it is considered that the actual response is significantly lower than expected, and a flashing red warning halo is generated around the mark, indicating that the drug is ineffective, and consideration should be given to changing the drug or adjusting the dose. If the SBP increases to the expected range, a brief green confirmation halo is generated, which disappears after 2 seconds. 2. After atropine injection, monitor changes in heart rate (HR). If the HR rises by less than 10 bpm, it is considered that the actual response is significantly lower than expected, and a red warning halo is generated. If the HR rises to 60 bpm or more, a green confirmation halo is generated. 3. Monitor after administration of dantrolene. And the trend of body temperature changes, if If the decrease in blood pressure is less than 5 mmHg and the rate of temperature rise does not decrease, it is considered a significantly lower-than-expected actual response, generating a red warning halo; if The temperature continues to drop and the rate of increase in body temperature slows down, generating a green confirmation halo; The system automatically accumulates the total dose of drugs of the same category and triggers an alert when the accumulated dose approaches the safety limit; for drug category c, let its accumulated dose be... The safety limit is Cumulative dose The calculation method is as follows: sum the doses with a brightness greater than 0.1 for all valid imprints of this category; Drug safety upper limit The calculation formula is: ; In the formula, This represents the upper limit of the recommended dose per kilogram of body weight for representative drugs in this category, based on the clinical anesthesia guidelines and expert consensus published by the Chinese Society of Anesthesiology; W represents the patient's actual weight in kg; for example, the single bolus dose of adrenaline is 0.1~0.5 mg. Take 0.03 mg / kg, with a cumulative dose generally not exceeding 1 mg; atropine Take 0.04 mg / kg, with a single maximum dose of 2 mg and a cumulative maximum dose of 5 mg; when Exceed At that time, a yellow warning halo is generated around all imprints of that category, with the halo brightness being equal to... Proportional; when Exceed When the yellow halo turns red and the pulsation frequency of the halo increases from 1Hz to 3Hz, it indicates an absolute risk of overdose. For example, if two 0.1mg doses of adrenaline have already been injected, with a cumulative dose of 0.2mg, and if the patient's upper limit for adrenaline is 0.3mg, then when the cumulative dose reaches 0.24mg, i.e., the 80% threshold, a yellow halo will appear around the two existing adrenaline imprints on the patient's arm simultaneously. When the anesthesiologist prepares to inject the third dose, both imprints will flash simultaneously, indicating that the cumulative dose has reached the warning line, and other vasoactive drugs should be considered. To avoid visual interference caused by an excessive number of valid imprints during prolonged resuscitation, the system defaults to rendering and displaying only valid imprints generated within the last 30 minutes in the field of vision; imprints older than 30 minutes and whose brightness has not dropped to the disappearance threshold are hidden by default, with only background data records retained, ensuring that only medication information strongly related to the current resuscitation is retained in the anesthesiologist's field of vision. S53. Dual-mode collaborative workflow: The collaborative workflow between the minimum action set dynamic guidance engine and the drug trajectory imprinting engine is as follows: Step S1: Triggering and initializing anesthesia emergencies.

[0021] The emergency identification module detects an intraoperative anesthetic emergency event, such as anaphylactic shock or malignant hyperthermia, and sends the emergency type identifier E and current vital signs data to the AR presentation module; the AR presentation module activates the dual-mode collaborative mode, clears the previously displayed non-emergency information, and enters the anesthetic emergency guidance state; Step S2: Calculate the minimum action set.

[0022] The minimum action set dynamic guidance engine calculates the primary action set for the current moment based on the emergency type E and the current anesthesia personnel and emergency equipment data provided by the situational awareness module. The result is sent to the rendering queue; Step S3: Drug Imprinting.

[0023] The drug trajectory imprinting engine reads all anesthetic medication records of the current patient in the last 30 minutes from the drug tracking module database, paying special attention to critical emergency drugs such as vasoactive drugs and antagonists, generating corresponding visual imprints, calculating the current brightness and performing spatial anchoring, and sending all valid imprints into the rendering queue. Step S4: Spatial conflict detection and collaborative layout.

[0024] The spatial layout of primary motion guidance elements and drug trajectory imprints in the rendering queue needs to be optimized to avoid mutual occlusion. The anesthesiologist's current field of vision is divided into a 10×10 grid area, and the density of existing rendering elements in each grid is calculated. For newly added rendering elements, the nearest area with the lowest density is selected for placement. For motion guidance elements, they are placed near the anesthesiologist's gaze point without obstructing key operation areas, such as the location of intravenous puncture points and airway operation areas. For drug imprints, their original anchoring position at the drug administration site is maintained, but the orientation of the imprint can be adjusted to ensure that it always faces the doctor's line of sight. Step S5: Dual-mode collaborative rendering.

[0025] Based on the optimized layout, the AR device simultaneously renders primary action guidance and drug imprints in the anesthesiologist's field of vision. Primary action guidance is presented in a prominent manner, such as dynamically flashing arrows pointing to the target operation location, such as the adrenaline drawer, oxygen knob, and intravenous access in an ambulance, accompanied by a concise action description. Drug imprints are presented as faded, semi-transparent labels at the corresponding drug administration sites.

[0026] When a primary action instruction and a drug imprint coincide in space, the system triggers a collaborative display mode. For example, when the minimum action set engine prompts for an immediate injection of epinephrine, if the patient already has two epinephrine imprints on their arm, representing the previous two doses, these two imprints will be highlighted simultaneously, and a connecting line will be generated to the end of the arrow of the current primary action instruction, signifying that based on the existing medication history, 0.2mg has been administered, and the current action should be performed. When the anesthesiologist receives the execution instruction, they can directly see the complete medication history and cumulative dosage, avoiding mechanical execution of instructions and ignoring the risk of drug accumulation. Step S6: Dynamic updates and real-time refresh.

[0027] As the emergency anesthesia treatment progresses, the system continuously monitors new medication events, changes in vital signs, and changes in the status of personnel and equipment, and updates them in real time: a new imprint is generated immediately when a new medication event occurs; the minimum action set engine recalculates the primary action set every 0.5 seconds, and updates the guidance immediately if there are changes, such as adding a primary action after emergency medications arrive; all drug imprints update their brightness values ​​every frame to achieve smooth visual decay. Step S7: Rapid status synchronization when a new anesthesiologist joins.

[0028] When the situational awareness module detects a new anesthesiologist or support nurse entering the operating room, the system automatically generates a quick status snapshot view for that person: in their AR field of vision, all currently effective drug imprints are briefly highlighted, with the brightness increasing to 1.0 for 3 seconds. At the same time, the current primary action guidance is displayed in a larger size in the center of the field of vision, allowing newly joined anesthesiologists to grasp the complete medication history and the core actions that must be performed in a short time without having to verbally ask what drugs were used or what operations were performed. This enables zero-delay team collaboration, which is crucial for time-sensitive anesthetic emergencies such as malignant hyperthermia.

[0029] Through the detailed description of the above embodiments, the AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies of the present invention, through the collaborative cooperation of an emergency identification module, a situational awareness module, a drug tracking module, and an AR presentation module, constructs a complete emergency response system covering accurate identification of anesthesia emergencies, real-time perception of rescue resources, full-process tracking of emergency medication, and AR-visualized intelligent guidance. The system achieves anti-interference grading of emergencies based on membership functions and sliding window algorithms; it achieves dynamic and accurate push of rescue actions adapted to on-site resources through a minimum action set dynamic guidance engine; it achieves spatialized visual tracking of medication history and medication safety warnings through a drug trajectory imprinting engine; and it achieves efficient collaboration of the rescue team through AR-rendered collaborative layout and rapid personnel status synchronization mechanism. The present invention effectively solves the core problems in existing intraoperative anesthesia emergency management, such as poor anti-interference identification, insufficient guidance adaptability, unintuitive drug tracking, and low team collaboration efficiency, providing reliable technical support for perioperative anesthesia safety management.

[0030] The above formulas are all dimensionless calculations, and the preset parameters in the formulas should be set by those skilled in the art according to the actual situation.

[0031] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0032] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0033] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0034] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0035] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0036] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0037] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies, characterized in that, It includes an emergency identification module, a situational awareness module, a drug tracking module, and an AR presentation module; The emergency identification module is used to collect the patient's vital signs parameters and image data during the operation in real time. Based on the pre-configured feature patterns corresponding to various anesthetic emergencies, it calculates the confidence level of the corresponding anesthetic emergency event and completes the determination of anesthetic emergencies and the triggering of emergency guidance process according to the preset confidence level threshold. The situational awareness module is used to collect data in real time on the identity and availability of medical staff, the location and operating status of emergency equipment, and the storage location and inventory status of emergency medicines in the operating room scene, and generate a dataset of the availability status of rescue resources in the current scene. The drug tracking module is used to collect and record drug information, administration information and operator information for intraoperative emergency medications in real time, and generate standardized medication records. The AR presentation module is communicatively connected to each of the aforementioned modules and includes a minimum action set dynamic guidance engine and a drug trajectory imprinting engine that work together. The minimum action set dynamic guidance engine is used to dynamically generate and render only the minimum set of emergency actions that need to be executed at the moment, based on the determined emergency type and the available status dataset of emergency resources. The drug trajectory imprinting engine is used to generate spatially anchored drug visual imprints based on standardized medication records and complete the rendering.

2. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 1, characterized in that, The minimum action set dynamic guidance engine is pre-configured with a complete resuscitation action library corresponding to various anesthesia emergencies. Each action in the action library is configured with personnel requirement parameters, equipment requirement parameters, execution attribute parameters and prerequisite dependencies.

3. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 1, characterized in that, The engine calculates the current executability of each action based on the available status dataset of rescue resources, generates a set of actions that can be executed immediately through topological sorting based on the pre-dependencies between actions, and prioritizes the actions in the set, rendering only the highest priority actions as the minimum rescue action set.

4. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 2, characterized in that, The minimum action set dynamic guidance engine calculates the comprehensive priority score of an action based on the prognostic relevance weight of the action itself and the blocking effect of the action on subsequent dependent actions, and completes the priority classification of the action based on the comprehensive priority score.

5. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 1, characterized in that, The drug tracking engine is pre-configured with visual coding rules corresponding to different emergency drug categories. The engine matches the corresponding visual code according to the standardized medication record to generate a visual imprint containing the core information of the drug. The visual imprint is anchored to the three-dimensional spatial coordinates of the administration site through visual SLAM technology, and the visual imprint is dynamically decayed over time.

6. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 5, characterized in that, The drug trajectory imprinting engine generates efficacy assessment markers at the corresponding drug visual imprints based on the patient's vital sign changes after medication. At the same time, the engine counts the cumulative dose of drugs of the same category, and triggers the warning rendering of the corresponding visual imprint when the cumulative dose approaches or exceeds the preset safety limit.

7. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 1, characterized in that, The emergency identification module configures corresponding membership functions for various signs and characteristics of different types of anesthetic emergencies. Based on the membership of each sign parameter, it calculates the instantaneous confidence of the emergency event, obtains the stable confidence through a sliding window smoothing process, and completes the suspected emergency determination, confirmed diagnosis determination, and guides the exit process based on the comparison results of the stable confidence and a preset threshold.

8. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 1, characterized in that, The situational awareness module calculates the number of available rescue personnel based on the type and availability of medical staff, filters the set of available emergency medical equipment based on the accessibility and operating status of equipment, calculates the patient risk index based on the patient's vital signs parameters, and generates the dataset of available rescue resources.

9. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 1, characterized in that, The AR rendering module performs spatial grid division and density detection on the rendered content of the two engines to complete the anti-occlusion collaborative layout of the rendered elements.

10. The AR-based intelligent emergency guidance system for intraoperative anesthesia emergencies according to claim 1, characterized in that, When new medical staff are detected entering the operating room, the AR rendering module automatically triggers the highlighting of visual imprints and the prominent rendering of the minimum set of resuscitation actions, thus completing the rapid synchronization of the resuscitation status.

Images