Virtual simulation critical nursing practical training system for diabetic ketoacidosis

Abstract

The invention provides a virtual simulation critical care practical training system for diabetic ketoacidosis. The virtual simulation critical care practical training system comprises a setting module, a dynamic illness state simulation module, a medical care support module, a time acceleration module, a virtual case module, a scene triggering module, a training module, a training nurse data acquisition module, a training analysis module and a cross-platform interface module. According to the invention, through dynamic illness state simulation, multi-modal interaction and intelligent evaluation, core pain points of lack of authenticity, strong evaluation subjectivity, insufficient special scene coverage and the like in traditional DKA nursing training are systematically solved. The innovation value of the method not only reflects the breakthrough of the nursing training technology, but also reconstructs the methodology of nursing ability cultivation of acute and critical diseases, and provides a reusable technical normal form for deep fusion of clinical practice and medical education.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality, in particular to a virtual simulation critical care training system for diabetic ketoacidosis. Background Art

[0002] Virtual simulation systems can simulate real-world scenarios in a virtual environment, such as emergency situations (e.g., cardiac arrest, asphyxia, and trauma), training nurses in cardiopulmonary resuscitation (CPR), hemostasis, and AED use. The system supports AI guidance, providing real-time feedback and assessment. Some schools abroad use virtual simulation systems to provide students with opportunities to repeatedly practice venipuncture, improving their skills by simulating the vascular conditions of different patients. In these virtual emergency scenarios, nursing students work closely with doctors to treat patients, enhancing teamwork. Most nursing schools in China have introduced virtual simulation systems for emergency nursing care, but lack simulation training systems for critical care nursing for common diseases. A domestic study found that surveys of newly diagnosed children with type 1 diabetes in some regions revealed that approximately half of them had ketoacidosis at the time of diagnosis, some with severe DKA. The prevalence of ketoacidosis in adults with type 2 diabetes is also significant. A multicenter retrospective study included 643 patients with diabetic ketoacidosis, of whom 45.7% had type 2 diabetes and 47.9% had type 1 diabetes.

[0003] Chinese patent application CN115376377A discloses a virtual simulation device for comprehensive pre-hospital rescue of traffic accidents, which includes a simulation training room, somatosensory virtual equipment (such as VR glasses, hand sensors), simulation training equipment (cardiopulmonary resuscitation simulator, trauma care model) and a control computer, and supports multi-scenario first aid process simulation. Accident data can be generated according to the set vehicle speed and casualties, and key treatment nodes (such as bleeding point positioning, airway management priority) can be marked through a labeling system. Real-time scoring is performed based on processing time and operation accuracy, which is suitable for multidisciplinary collaborative training. This system is mainly used in first aid systems. The hardware system can be adopted for critical care training of diabetic ketoacidosis, but the software system is not applicable.

[0004] Chinese patent application CN113257095A discloses an intelligent acupuncture model and an intelligent acupuncture training system, which integrates sensors on the traditional acupuncture bronze man to achieve accurate mapping of the acupuncture position and the virtual anatomical structure (acupoints, nerves). The training nurse's operation results (such as acupuncture depth and angle) are linked to the teaching software through the system communication module to generate operation scores and correction suggestions. It solves the problem that rigid materials cannot truly simulate the feel of puncture, and is extended to nursing operation (such as venipuncture) training scenarios. This system is suitable for the venipuncture part of the critical care nursing training for diabetic acidosis, and does not involve the overall connotation of critical care nursing training.

[0005] Chinese patent application CN118315007A discloses an interactive method, system, and device for diabetic patients, designed to achieve more precise management of diabetic patients. Chinese patent application CN114242253A discloses a health management method and system for early warning of diabetic ketoacidosis, combining artificial intelligence and big data analysis to monitor patients' blood sugar, blood ketones, and other indicators in real time and predict DKA risk. However, none of these systems are suitable for critical care training for diabetic acidosis. Summary of the Invention

[0006] In order to solve the above problems, the purpose of the present invention is to provide a virtual simulation critical care training system for diabetic ketoacidosis.

[0007] The present invention provides a virtual simulation critical care training system for diabetic ketoacidosis, the system comprising:

[0008] The dynamic condition simulation module dynamically simulates a virtual emergency case of a diabetic ketoacidosis patient by combining the dynamic ECG monitoring display unit, the patient dynamic status display unit, the dynamic infusion display unit, the dynamic urine bag display unit, and the dynamic finger blood glucose display unit;

[0009] Virtual case module, used for hierarchical management of virtual cases of emergency treatment for patients with diabetic ketoacidosis;

[0010] A scenario triggering module is used to simulate multiple types of virtual scenarios during virtual case training, including virtual disease scenarios, virtual treatment scenarios, virtual nursing scenarios, and virtual humanities scenarios;

[0011] The training module is used to guide nurses to perform operations during virtual case training.

[0012] Optionally, a Holter monitor display screen unit is used to continuously display the patient's Holter monitor according to a predetermined virtual case setting, and to display the monitor on a split screen in certain circumstances;

[0013] The dynamic status monitoring display unit is used to display the patient's whole body or local status. The local status includes: head: facial expression, complexion, pupils, nasal breathing, oral cavity, oxygen inhalation status, ventilator status, endotracheal intubation status, nasogastric feeding tube status; pupil: click the light source to project the pupil special effect: pupil dilation, click to remove the light source to display the pupil special effect; chest: shortness of breath, obvious chest rise and fall; abdomen; limbs: local skin status; perineum; sacrum;

[0014] The dynamic infusion display unit virtually displays the patient's dynamic infusion status and syringe pump status according to the infusion speed and syringe pump speed adjusted by the trained nurse, and adjusts the speed;

[0015] Dynamic urine bag display unit, used to display the urine color, properties, empty and full status, and urine volume in the urine bag;

[0016] Dynamic finger blood glucose display unit, used to display finger blood glucose value.

[0017] Optionally, the levels of virtual cases managed in the virtual case module include elementary, intermediate and advanced levels, each virtual case is bound through a setting module, and the corresponding virtual case is output during training;

[0018] Primary: Virtual Case of Typical Diabetic Ketoacidosis without Complications in Adults and Children;

[0019] Intermediate: Virtual case of diabetic ketoacidosis combined with heart failure in the elderly, virtual case of diabetic ketoacidosis in pregnancy; virtual case of diabetic ketoacidosis in pregnancy, virtual case of diabetic ketoacidosis using SGLT-2 inhibitors;

[0020] Advanced: Virtual case of cardiac arrest caused by hypokalemia in diabetic ketoacidosis, and virtual case of diabetic ketoacidosis combined with shock.

[0021] Optionally, the scene triggering module includes:

[0022] Condition triggering units include blood glucose change triggering subunit, hypokalemia ECG monitoring triggering subunit, cerebral edema triggering subunit, acute pulmonary edema triggering subunit, shock triggering subunit, emergency cesarean section triggering subunit, and DKA symptom relief triggering subunit;

[0023] Treatment trigger unit, including fluid therapy plan doctor order change trigger sub-unit, gastrointestinal fluid therapy doctor order change trigger sub-unit, insulin injection pump therapy doctor order change trigger sub-unit, potassium supplementation doctor order change trigger sub-unit, acidosis correction doctor order change trigger sub-unit, infection control doctor order trigger sub-unit, and complication doctor order trigger sub-unit;

[0024] The humanistic trigger unit includes the trigger sub-unit caused by family members’ dissatisfaction with nursing care, the trigger sub-unit caused by worsening of the disease and dissatisfaction with treatment, and the trigger sub-unit caused by family members’ concerns about the disease.

[0025] Optionally, the training module includes:

[0026] The introduction unit is used to guide nurses in training by putting on VR goggles while playing predetermined virtual case scenarios;

[0027] In the reception unit, AI support personnel and trained nurses conduct shift handovers, while voice message reminders and patient status displays are also provided;

[0028] Rescue unit, used to assist in training nurses in rescue scene operations;

[0029] The outcome unit is used to prompt the trained nurse to put on the VR goggles again to check the patient's outcome.

[0030] Optionally, the system further comprises a medical care support module for simulating a clinical medical care support system; the medical care support module comprises:

[0031] The medical support unit includes a medical order support subunit, an examination support subunit, and a test support subunit. By pre-setting virtual cases, the medical order support subunit gradually displays the medical order as virtual time advances. The examination support subunit is used to print out examination sheets and view the examinations and reports that the virtual patient has undergone. The test support subunit is used to bind blood collection tubes and view the tests and reports that the virtual patient has undergone.

[0032] The nursing support unit includes a nursing assessment subunit and a nursing document subunit; the nursing assessment subunit is used to conduct fall and bed fall assessment, pressure ulcer risk assessment, self-care ability assessment, Glasgow coma assessment, tube slip assessment, mood thermometer assessment, pain assessment, nutritional risk assessment, restraint needs assessment, and deep vein thrombosis assessment; the nursing document subunit is used to manage admission assessment, nursing diagnosis, and non-surgical nursing record sheets.

[0033] Optionally, the system further comprises a time acceleration module, which is used to accelerate or stop time upon receiving an instruction from the training nurse during the virtual case training process.

[0034] Optionally, the system further includes a training nurse data acquisition module for acquiring data generated by the training nurses during the virtual case training process; the trainee data acquisition module includes:

[0035] A motion trajectory unit is used to record the motion trajectory data of the trainee nurse's limbs during the virtual case training through sensors set on the trainee nurse's limbs;

[0036] The wristband unit is used to record the physiological data of nurses in training at different time periods;

[0037] The audio and video recording unit is used to record the communication, operation data, facial expression data and eye tracking data of the training nurses during the virtual case training process.

[0038] Optionally, the system further comprises a training analysis module for analyzing the acquired training nurse data and generating an analysis report;

[0039] The training analysis module includes: an operation analysis unit for performing comparative analysis based on the motion trajectory data, operation steps and standard data;

[0040] The thinking analysis unit is used to analyze attention distribution based on eye tracking data; analyze physiological quality based on physiological data during rescue; analyze the psychological stress resistance of nurses in training based on the period of voice tremor in specific links and physiological data at that time; and analyze emotional and psychological states based on facial expression data;

[0041] A humanities analysis unit is used to extract communication data between the training nurses and their families, analyze the training nurses' psychological state, and screen and classify caring words based on the communication data, count the number of times they are used, and generate an analysis report;

[0042] The teamwork analysis unit is used to analyze communication efficiency based on internal team communication data, and to analyze the operation proportion and rescue efficiency of each team member.

[0043] Optionally, the system further comprises a setting module for setting user information and training information;

[0044] The cross-platform interface module is used to manage multiple cross-platform interfaces so as to utilize the cross-platform interfaces to perform platform docking with various hardware devices.

[0045] The virtual simulation critical care nursing training system for diabetic ketoacidosis of the present invention can simulate the real first aid scenario of diabetic ketoacidosis patients. At the same time, combined with actual operation training, it breaks the time and space limitations of traditional nursing training (the operation is disconnected from the patient's condition at the time), relies on real cases (there are problems such as high operation risks, non-repeatable scenarios, and high costs), improves the subjective evaluation system (skill assessment avoids relying solely on the instructor observation method for assessment, and derives the evaluation of students' psychological quality and humanistic care), and designs nursing deductions for special scenarios (complications such as acute pulmonary edema or cerebral edema).

[0046] The present invention's virtual simulation critical care nursing training system for diabetic ketoacidosis utilizes three key technical pillars: dynamic disease simulation, multimodal interaction, and intelligent assessment. It systematically addresses the core pain points of traditional DKA nursing training, including a lack of authenticity, subjective assessments, and insufficient coverage of specific scenarios. Its innovative value lies not only in its breakthroughs in nursing training technology but also in reconstructing the methodology for developing critical care nursing skills, providing a reusable technical paradigm for the deep integration of clinical practice and medical education.

[0047] Based on the following detailed description of specific embodiments of the present invention in conjunction with the accompanying drawings, those skilled in the art will become more aware of the above and other objects, advantages and features of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0049] Figure 1 This is a structural diagram of a virtual simulation critical care training system for diabetic ketoacidosis according to an embodiment of the present invention;

[0050] Figure 2 The configuration module of the embodiment of the present invention shows the configuration module structure and workflow diagram;

[0051] Figure 3 This is a dynamic condition simulation module of an embodiment of the present invention: a module structure and workflow diagram for dynamically displaying the condition simulation status of a patient under a specific scenario setting;

[0052] Figure 4 This is a schematic diagram of the structure of the medical care support module according to an embodiment of the present invention;

[0053] Figure 5 This is a schematic diagram of the structure and workflow of the time acceleration module according to an embodiment of the present invention;

[0054] Figure 6 It is a schematic diagram of the structure and workflow of the virtual case module of an embodiment of the present invention;

[0055] Figure 7 Schematic diagram of the structure of the scene trigger module according to an embodiment of the present invention;

[0056] Figure 8 It is a schematic diagram of the structure and workflow of the training module of an embodiment of the present invention;

[0057] Figure 9 This is a schematic diagram of the structure of a student data acquisition module according to an embodiment of the present invention;

[0058] Figure 10 This is a schematic diagram of the structure of the training analysis module according to an embodiment of the present invention;

[0059] Figure 11 The cross-platform interface module of an embodiment of the present invention shows a schematic diagram of the cross-platform interface module structure. DETAILED DESCRIPTION

[0060] The embodiments of the present invention are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to illustrate the present invention and are not intended to limit the present invention.

[0061] The embodiment of the present invention provides a virtual simulation critical care training system for diabetic ketoacidosis, such as Figure 1 As shown, the system includes ten modules: setup module, dynamic condition simulation module, medical care support module, time acceleration module, virtual case module, scenario trigger module, training module, nurse training data acquisition module, training analysis module, and cross-platform interface module. The following describes the functions of each module in detail.

[0062] 1. Setting up the module

[0063] The setting module is mainly used to set user information and training information. User information can include teacher information and student information, and training information can include training tasks, training instructions, and other information. Specifically, the setting module can set the list of bound teachers and training nurses, set training tasks for professors, train nurses to train binding, train nurses to train instructions, etc. The specific process is as follows Figure 2 shown.

[0064] Login interface: (Professor) Enter the work number and click the login button to enter the training nurse entry interface. Enter the training nurse's student number and automatically bind the training nurse's name. After entering, click the save button to enter the next interface. If there is any modification, click the modify button to modify it. After saving, enter the next interface. The training content selects the elementary typical diabetic ketoacidosis without complications case, which is divided into adult version and pediatric version. Intermediate: ① Case of elderly diabetic ketoacidosis with heart failure; ② Case of gestational diabetic ketoacidosis; ③ Case of diabetic ketoacidosis with cerebral edema; ④ Case of diabetic ketoacidosis using SGLT-2 inhibitors; ⑤ Random. Advanced: Case of cardiac arrest caused by hypokalemia in diabetic ketoacidosis; Case of diabetic ketoacidosis with shock; Random. After saving, enter the next interface. If there is any modification, click the modify button to modify it. After saving, enter the next interface.

[0065] Select the training method, with or without an AI instructor. Multiple roles are available, including AI doctors, AI head nurses, AI nurses, AI patients, and AI family members. Training nurses are recommended based on the selected case. Entry-level cases default to single-nurse training, while intermediate and advanced cases are set up for two or three people (if only one person is training, AI training nurses B and C can be set up). Save the file and proceed to the next screen. If you need to make any changes, click the edit button to make them, then save and proceed to the next screen. Training Instructions: Example: This training is entry-level training, with a single person and AI instructor, featuring an AI doctor, AI patient, AI family member, and AI support staff. Multiple flat-panel displays (all touchscreen-enabled): The Holter monitor display is located above the monitor, the patient's dynamic status display is located above the mannequin's head, the ambulatory infusion system display is located on the IV stand, the ambulatory urine bag display is located near the bag's hook, and the ambulatory fingerstick glucose display is located near the left finger. The Medical Care Support Unit is installed on both the nurses' station computer and the nurses' mobile work tablets. The Medical Support Unit primarily processes medical orders and reviews examination and test results, while the Nursing Support Unit primarily performs various assessments and records. The printing equipment is located at the nurses' station, with three printers capable of printing medical order execution sheets, medication patches, and patient wristbands. The system prompts: Do you want to accelerate time? How will the training be conducted? If the nurse agrees, she can answer: "Accelerate time." Time will be accelerated and stopped 2 minutes before the corresponding task time. The training venue has multiple cameras and recording functions. Before training, please wear wristbands A, B, C... (select the corresponding options). Before training, please wear motion sensor devices A, B, C... for both upper limbs (mid-forearm) and lower limbs (ankles). Before training, please wear vests A, B, C..., which have cameras and voice recording devices. Before training, please wear VR glasses and adjust the tightness. Following the system's voice prompts, remove the VR glasses for practical training. When all operations are completed, a voice prompt, such as "ECG monitoring operation completed," will be given. This facilitates the system's advancement of virtual case progress.

[0066] 2. Dynamic disease simulation module

[0067] The dynamic condition simulation module includes five display screen units: dynamic electrocardiogram monitoring display screen unit, patient dynamic status display screen unit, dynamic infusion display screen unit, dynamic urine bag display screen unit and dynamic finger blood glucose display screen unit. Figure 3 shown.

[0068] Holter monitor display unit: Located above the monitor. Activation conditions: The trainer voice prompts: "Complete ECG monitoring operation." The system voice prompts: "Please turn off the ECG monitor." The tablet above the monitor manually triggers the Holter monitor to activate. The tablet displays the patient's Holter monitor continuously based on the pre-defined virtual case settings. In special cases, such as pregnant patients with DKA, a split screen may be used. If the system voice prompts for fetal heart rate monitoring, the tablet will split into two screens: the left side displays the Holter monitor and the right side displays the fetal heart rate.

[0069] Dynamic Status Monitoring Display Unit: Located above the mannequin's head. Activation Condition: When a nurse enters the virtual emergency room, the screen continuously displays the patient's global and local status. A system voice prompt states: The virtual patient is above the mannequin's head. To view local status, simply touch the corresponding area. The tablet display is divided into two areas: global and local. The global area continuously displays the patient's entire body scene based on the pre-defined virtual case settings. The local screens are as follows: Head: Facial expression, complexion, pupils, nasal breathing, oral cavity, oxygenation status, ventilator status, endotracheal intubation status, nasogastric tube status, etc. Pupils: Clicking the light source will display pupil effects on the tablet: dilated pupils; clicking the light source will remove the pupil effects. Chest: Shortness of breath, noticeable chest rise and fall (based on the virtual case settings). Abdomen: Abdominal pain, such as abdominal bulge during pregnancy or pancreatitis, manifests as abdominal curling. Extremities: Local skin conditions, such as dryness, and intravenous puncture sites, such as extravasation and blood reflux. Perineum: Conditions, such as urinary incontinence. Sacrococcyx: Pressure ulcers, such as skin flushing.

[0070] Dynamic Infusion Display Unit: Located on the IV stand. Triggering conditions: The trainer voice-activated: Infusion operation completed (activates the infusion display); Syringe pump operation completed (activates the syringe pump display); System voice prompts: ① Turn off the infusion plunger and manually trigger the left side of the tablet on the IV stand to activate dynamic infusion; ② Turn off the syringe pump power and manually trigger the right side of the tablet on the IV stand to activate the dynamic syringe pump. The tablet displays: ① The patient's dynamic infusion status and syringe pump status are displayed virtually, based on the infusion and syringe pump speeds adjusted by the trainer. ② The tablet display area is divided into two areas: the dynamic infusion status display area and the dynamic syringe pump status display area. Clicking the infusion drop count will pop up a dialog box. Touch the screen to adjust the infusion drop count using the +, -, or OK buttons. Clicking Pump Speed 1 (or Pump Speed 2, etc.) will pop up a dialog box. Touch the screen to adjust Pump Speed 1 (or Pump Speed 2, etc.). ③ Touch the screen + to increase the number, - to decrease the number, and ok to confirm (the corresponding display screen speed is synchronized); ④ Scan the QR code of the patient's wristband and infusion bag, and the dynamic virtual status of the infusion will show the infusion of this bag. Similarly, the injection pump will display the running drug; ⑤ The dynamic virtual status of the infusion and injection pump will show empty infusion bag, no dripping of infusion, injection pump alarm (such as pressure alarm, which will make a dripping sound. Clicking the alarm area will silence the sound and readjust the pump speed to resume normal operation) under specific settings.

[0071] Dynamic urine bag display unit: Located where the urine bag is hanging. Activation condition: The trainer voice-commands: "Catheterization completed." System voice prompts: The tablet is divided into upper and lower display areas: the upper area displays the color, properties, and empty / full status of the urine bag after the virtual patient's catheterization, while the lower area displays the urine volume. To empty the bag, the training nurse voice-commands: "Empty urine bag." The tablet displays: ① The maximum display size for the urine bag is 1000ml; ② The training nurse voice-commands: "Empty the bag," the bag displays empty status, and the urine volume shows 0ml.

[0072] Dynamic finger glucose display unit: Located near the left finger. Activation condition: Trainer voice message: "Finger glucose operation completed." System voice prompt: "Finger glucose measurement completed. Check the tablet to see the latest finger glucose value." Tablet display: Displays the patient's current finger glucose value according to the pre-set virtual case settings.

[0073] 3. Medical care support module

[0074] The medical care support module is used to simulate the clinical medical care support system, which includes a medical support unit and a nursing support unit. The diabetic ketoacidosis virtual simulation critical care training system of this embodiment is connected to the hospital medical care support unit through the HIS system to synchronously simulate the clinical medical care support system, and is installed in the nurse station computer and the nurse mobile platform work board. The medical support unit includes a medical order support subunit, an examination support subunit, and a test support subunit. By pre-setting virtual cases, the medical orders are gradually displayed in the medical order support subunit as the virtual time advances, and the training nurses process the medical orders after logging in; in the examination support subunit, the examination list can be printed out and the examinations and reports that the virtual patient has done can be viewed; in the test support subunit, blood collection tubes (barcodes) can be bound and the tests and reports that the virtual patient has done can be viewed. To access the training system, training nurses must first turn on the computer at the nurse station, click on the medical support unit, and log in to start the medical support unit. Three printers are also configured: one dot matrix printer, which can print medical order execution forms when connected, two thermal printers, which can simultaneously print thermal medical order labels (for affixing to infusion bags, syringes, or oral medication bags), and another printer to print patient wristbands. Nurses can log in to the medical support unit on their mobile work tablet to view medical orders, examinations, and laboratory tests, which are used when verifying medical orders. The nursing support unit is on the mobile tablet, and training nurses can open it and log in to use it. It includes a nursing assessment subunit and a nursing document subunit. The nursing assessment subunit includes fall and bed assessment, pressure ulcer risk assessment, self-care ability assessment, Glasgow coma assessment, tube slip assessment, mood thermometer assessment, pain assessment, nutritional risk assessment, restraint needs assessment, and deep vein thrombosis assessment. The nursing document subunit includes the admission assessment homepage, nursing diagnosis, and non-surgical nursing record sheet. Nurses are trained to receive virtual patients and perform various assessments and write the homepage and records. The writing method is touch and handwriting (some records will be automatically entered by the system, such as various assessments, infusion volume, etc.). When leaving the nurse station, they should take the nurse's mobile work tablet to the emergency room (the system will give a voice reminder). Different hospitals use different systems, and various assessments will vary. Figure 4 .

[0075] 4. Time Acceleration Module

[0076] Time Acceleration Module: The system's voice reminders to the training nurse are not included in the training time. The system determines whether time can be accelerated based on the training nurse's operational milestones and the progress of the virtual case. During the virtual case training process, the system receives instructions from the training nurse to accelerate or stop time. Specifically, the training nurse can be prompted by voice, "Do you want to accelerate time?" If the nurse replies, "Accelerate time." The time acceleration module runs, advancing time at a rate of 120 minutes per 1 minute. If a task is advanced by 2 minutes during the specified time period or the urine bag volume reaches 990 ml, the time acceleration module stops. If the task is not completed within 2 minutes of the task point, the training nurse will be prompted with a "Task xx is not completed." The training nurse can complete the task according to the reminder. To continue accelerating time, wait for the next voice reminder. For example, if the remaining infusion volume is 200 ml and the infusion rate is 160 drops / minute, the system will stop when 16 ml of infusion remains (calculation method: 20 drops = 1 ml, 16 ml = 320 drops ÷ 160 drops / minute = 2 minutes). The urine volume is set to 30% of the total fluid volume, that is, the total fluid volume is 6000ml and the urine volume is 1800ml. If the total fluid volume reaches 3334ml and the urine volume is about 1000ml, the fast forward will stop. If the nurse finds that the urine bag is emptied in time, the fast forward will stop when the urine volume reaches 990ml again. The system is set to stop fast forward when a new condition changes within the specified time or 2 minutes before the scene is imported. Figure 5 .

[0077] 5. Virtual Case Module

[0078] The virtual case module manages virtual cases at different levels, including elementary, intermediate and advanced. Figure 6. Each virtual case is bound by setting up a module, and the corresponding virtual case is output during training. Primary: ① A typical adult version of a virtual case of diabetic ketoacidosis without complications. The patient presents with hyperglycemia, high blood ketones or positive urine ketones, metabolic acidosis, and Kussmaul breathing with the smell of rotten apples. Weight loss, fatigue, loss of appetite, nausea and vomiting, dry skin and mucous membranes, sunken eyeballs, decreased urine volume, tachycardia, decreased blood pressure, drowsiness in the later stages of the disease, laboratory characteristics include high potassium first and then low, high sodium first and then low, leukocytosis, and slightly elevated blood urea nitrogen / creatinine. Infection and other inducements are common. Treatment mainly includes fluid resuscitation, insulin therapy, correction of electrolytes, acidosis management, monitoring and transition, etc. ② A hypothetical case of a typical uncomplicated pediatric diabetic ketoacidosis. Compared to the adult version, DKA presents with a more rapid onset and more severe metabolic disturbances. Symptoms of DKA in children are atypical, typically presenting with lethargy, loss of appetite, abdominal pain, and vomiting. Abdominal pain and gastrointestinal symptoms are prominent, and blood pressure remains normal or elevated even in dehydrated patients. Treatment requires a more cautious, low-dose insulin regimen, as treatment for DKA in children can vary rapidly. Intermediate: ① A hypothetical case of elderly patients with diabetic ketoacidosis and heart failure. DKA in the elderly often lacks typical symptoms (such as polyuria and thirst) and is instead characterized by nonspecific manifestations such as confusion, loss of appetite, chest tightness, abdominal pain, or coma. Due to decreased cardiac and renal function, elderly patients are more susceptible to developing heart failure during fluid rehydration, primarily manifesting as symptoms of acute pulmonary edema. Treatment must consider both underlying conditions and complications, such as acute pulmonary edema. ② A virtual case of gestational diabetic ketoacidosis. The cause of DKA is closely related to changes in hormone levels during pregnancy, such as increased insulin antagonist hormones such as placental lactogen and prolactin, which lead to worsening insulin resistance. In addition, hyperemesis gravidarum, infection, prolonged fasting, or the use of glucocorticoids are also common triggers. Gestational DKA occurs in the second and third trimesters. The disease progresses more quickly. Abdominal pain may induce uterine contractions and lead to fetal distress. Fluid replacement: Avoid excessive glucose restriction. In late pregnancy, 10% glucose supplementation is required to meet fetal needs. Insulin dosage: It needs to be adjusted carefully to prevent hypoglycemia from affecting the fetus. Fetal monitoring: Continuous fetal heart monitoring is required to assess fetal condition. ③ A virtual case study of diabetic ketoacidosis complicated by cerebral edema. DKA is a common condition in children, especially those with newly diagnosed diabetes. Symptoms of cerebral edema include: Early symptoms: headache, restlessness, recurrent vomiting, unexpected decrease in heart rate, and elevated blood pressure; Progressive symptoms: worsening level of consciousness (drowsiness, coma), abnormal breathing patterns (such as apnea), cranial nerve palsies (oculomotor nerve involvement, anisocoria); Critical manifestations: rapid progression to brain herniation. Treatment requires prompt use of osmotic dehydrating agents such as mannitol, and early monitoring is crucial. ④ A virtual case study of diabetic ketoacidosis using SGLT-2 inhibitors. This often occurs in elderly patients taking SGLT-2 inhibitors. DKA is not typically associated with hyperglycemia, so SGLT-2 inhibitors should be discontinued. Other treatment options are the same as for elderly patients with DKA.Advanced: ① A virtual case of diabetic ketoacidosis with hypokalemia leading to cardiac arrest. This case involved a variety of factors leading to delayed potassium supplementation in the early stages of DKA treatment, and the nurse failed to promptly detect hypokalemia on ECG monitoring, resulting in cardiac arrest. Treatment is similar to cardiopulmonary resuscitation. ② A virtual case of diabetic ketoacidosis combined with shock was seen in an elderly patient taking SGLT-2 inhibitors. Diagnosis was delayed due to the absence of hyperglycemia, leading to severe inadequate fluid replacement and the development of hypovolemic shock. Characteristic symptoms include cold extremities and decreased blood pressure. Treatment involves rapid fluid replacement combined with circulatory resuscitation, while hourly urine output is monitored to adjust the fluid replacement rate and prevent heart failure. If blood pressure remains low despite adequate fluid replacement, consider using vasoactive medications.

[0079] 6. Scene trigger module

[0080] The scenario trigger module includes four aspects: condition, treatment, nursing and humanities. Figure 7 .

[0081] Condition Trigger Unit: 1) Blood Glucose Change Trigger Subunit: Fluid Replenishment Achieves Targeted Volume [Adults: Estimate total fluid volume based on 10% of the patient's original body weight. For example, a 60kg patient should receive approximately 6000ml of fluid over 24 hours. The average adult should receive 4000-6000ml, while those with severe dehydration may receive up to 6000-8000ml. First Hour: 15-20ml / kg (approximately 1.0-1.5L of normal saline for an adult). First 2 Hours: 1000-2000ml, subsequently adjusted to 250-500ml / hour based on blood pressure, urine output, and other parameters. Children: Rapid rehydration: 10-20ml / kg normal saline (infusion over 30-60 minutes), maintenance rehydration: infuse at an even rate for the remaining 24 hours, for the elderly, the total rehydration volume is estimated at 7% of the patient's original body weight. The blood sugar decrease rate for virtual case primary ① decreases by 4mmol / l per hour; virtual case primary ② decreases by 3mmol / l per hour; virtual case intermediate ① decreases by 3mmol / l per hour; virtual case intermediate ② decreases by 3mmol / l per hour; virtual case intermediate ③ decreases by 5.5mmol / l per hour before cerebral edema occurs, and decreases by 3mmol / l per hour after reducing the insulin dose (reducing the pump speed); virtual case intermediate ④ decreases by 0.1mmol / l per hour; virtual case advanced ① decreases by 4mmol / l per hour; virtual case advanced ② decreases by 0.1mmol / l per hour. 2) Hypokalemia ECG Trigger Subunit: In the hypothetical case, Advanced ①, the patient's blood potassium level was 4.0 mmol / l. After 30 minutes without potassium supplementation, ECG monitoring showed signs of hypokalemia, including flattened or inverted T waves, present or elevated U waves, and prolonged QT interval. 3) Cardiac Arrest Trigger Subunit: In the hypothetical case, Advanced ①, after 10 minutes of persistent hypokalemia on ECG monitoring, an arrhythmia (ventricular tachycardia) and an ECG alarm sounded. Two minutes after the alarm, cardiac arrest occurred. 4) Cerebral Edema Trigger Subunit: In the hypothetical case, Intermediate ③, the patient's fingerstick blood glucose level decreased by 5.5 mmol / l for the second time per hour, and 10 minutes after that, showed early signs of cerebral edema. 30 minutes after not reducing insulin dosage (or pump speed), symptoms of moderate to advanced cerebral edema developed. 60 minutes after not reducing insulin dosage (or pump speed), symptoms of critical cerebral edema developed. The patient gradually improved after 1-2 hours of dehydration. 5) Acute Pulmonary Edema Triggering Subunit: In the Intermediate Case ① scenario, the total volume of fluid infusion reached 2000ml, triggering acute pulmonary edema. Symptoms included: restlessness, profuse sweating, extreme dyspnea, a significantly increased respiratory rate (30-40 breaths / minute), and a feeling of suffocation. Coughing produced pink, frothy sputum, cyanosis of the lips, nail beds, and skin, increased heart rate, fluctuating blood pressure, crackles on lung auscultation, and cold, clammy extremities. After 20 minutes of acute pulmonary edema rescue treatment, symptoms resolved.6) Shock Trigger Subunit: In the virtual case Advanced II, early shock symptoms appear after blood pressure remains at 90 / 60 mmHg for 30 minutes: decreased or anuria, increased heart rate (100-120 beats / min), anxiety, restlessness, pale complexion, and cold, clammy extremities. Ten minutes after the fluid infusion rate is not increased, progressive shock symptoms appear: apathy, unresponsiveness, decreased blood pressure (systolic blood pressure <90 mmHg), rapid or shallow breathing, and cyanosis (bluish lips and nail beds). Severe shock symptoms appear after systolic blood pressure remains below 90 mmHg for 20 minutes without an increase in the fluid infusion rate: continued decrease in blood pressure (ECG alarm) and coma. Improvement occurs within 1-2 hours after emergency treatment. 7) Emergency Cesarean Section Trigger Subunit: In the Intermediate ② virtual case, in the late stages of overall treatment, the fetal heart rate will show: <110 beats / minute, or >160 beats / minute for 10 minutes. The system will display an obstetric consultation voice message: Emergency Cesarean Section Notice, Make Preoperative Preparations. 8) DKA Symptom Relief Trigger Subunit: Fluid infusion reaches 50% of the daily total according to schedule, fingerstick blood glucose <11.1 mmol / l, blood potassium is normal, blood pH >7.3, the patient regains consciousness, thirst disappears, skin elasticity is restored, and nausea, vomiting, and abdominal pain disappear.

[0082] Treatment Trigger Unit: 1) Fluid Therapy Plan Change Trigger Subunit: In Virtual Cases Elementary ①, Intermediate ①, Intermediate ②, and Advanced ①, if finger blood glucose is ≤13.9 mmol / L, the saline infusion should be discontinued and replaced with 5% dextrose solution or dextrose saline solution, while continuing insulin therapy. The dosage is typically 1 unit of insulin per 2-4 grams of dextrose (e.g., 500 ml of 5% dextrose + 6-12 units of insulin), with specific adjustments based on blood glucose monitoring. In Virtual Cases Elementary ②, Intermediate ③, Intermediate ④, and Advanced ②, if finger blood glucose is ≤11.1 mmol / L, the saline infusion should be discontinued and replaced with 5% dextrose solution or dextrose saline solution. If shock occurs, follow the shock order change. 2) Gastrointestinal Fluid Infusion Order Trigger Subunit: In Virtual Case Intermediate ①, after treatment of acute pulmonary edema improves, a gastrointestinal fluid order is initiated: insert a gastric tube, administering 1 / 2 of the total fluid volume, at a constant rate every hour, while simultaneously slowing the intravenous fluid rate. 3) Insulin pump therapy order change trigger subunit: Adult version: 50 units of insulin + 50 ml of 0.9% NS; Pediatric version: 1 unit / kg of insulin + 50 ml of 0.9% NS. Infusion pump rate: 4 ml / h for fingerstick blood sugar > 20 mmol / l; 2 ml / h for fingerstick blood sugar 14-20 mmol / l; 1 ml / h for fingerstick blood sugar 10-13.9 mmol / l; 0.5 ml / h for fingerstick blood sugar < 9.9 mmol / l. 4) Potassium supplementation order change trigger subunit: Blood potassium < 3.3 mmol / L: Prioritize potassium supplementation, withhold insulin therapy, and resume insulin therapy only after blood potassium rises to ≥ 3.5 mmol / L. Blood potassium < 5.2 mmol / L and normal urine output (> 40 ml / h): Administer intravenous potassium concurrently with fluid and insulin therapy. Normal blood potassium but urine output > 40 ml / h: Even if blood potassium is normal, potassium supplementation is still necessary. Normal blood potassium but urine output <30ml / h: Potassium supplementation is resumed after urine output recovers. Blood potassium ≥5.3mmol / L: Stop potassium supplementation. Concentration: Generally, 1.5-3.0g (40-60mmol / L) of potassium chloride is added per liter of fluid, and at least 40mmol / L is recommended for children. Rate: Initially, 13-20mmol of potassium is supplemented per hour (equivalent to 1.0-1.5g / h of potassium chloride). In severe hypokalemia, it can reach 20-60mmol / h, but ECG monitoring is required and the rate must be controlled to ≤1.5g / h. Total amount for 24 hours: Usually 6-10g of potassium chloride is required, which can be combined with oral potassium supplementation. 5) Correction of acidosis doctor's order change trigger subunit: Use when pH <6.9 (severe acidosis) or life-threatening hyperkalemia (blood potassium ≥6.5mmol / L) or shock, 1.7-3.4ml / kg of 5% sodium bicarbonate, and discontinue use when pH ≥7.0. 6) Infection control order trigger subunit, white blood cell count ≥ 10x109 / l, accompanied by infection symptoms, and intravenous antibiotics are prescribed.7) Complications medical order triggering sub-unit: Acute pulmonary edema: Virtual case intermediate ① The nurse calls the doctor to determine that it is acute pulmonary edema, and prescribes a series of medical orders for rescuing acute pulmonary edema: body position adjustment, slow down the infusion rate, high-flow oxygen therapy, drug treatment such as diuretics, vasodilators, cardiotonics, sedation, hormones, etc.; Acute cerebral edema: Virtual case intermediate ③ The nurse calls the doctor to determine that it is acute cerebral edema, and prescribes a series of medical orders for rescuing acute pulmonary edema: raise the head 30 degrees, administer mannitol 0.5-1.0g / kg rapid intravenous drip, slow down the fluid infusion rate, reduce the insulin pump speed, etc.

[0083] Nursing Trigger Unit: 1) Infusion Progress Trigger Subunit, which calculates the total patient volume. For example, for an elderly patient weighing 60 kg, the total fluid volume is 4200 ml. If 2000 ml is infused and acute pulmonary edema develops, gastrointestinal fluid replacement is prescribed. Of the remaining 2200 ml, 1100 ml is intravenous fluid and 1100 ml is gastrointestinal fluid. Fluid replacement volumes for adults and children are calculated using the corresponding formulas. Fluid replacement is calculated based on the nurse's adjusted drip rate: 20 drops / min = 1 ml. If the drip rate is 320 drops / min, 1000 ml of fluid will take 62.5 minutes to deliver. 2) Urine production rate triggering subunit: advance according to the amount of fluid replacement: in virtual case primary ①, virtual case primary ② and virtual case intermediate ②, within the first hour, the urine volume is calculated as 2% of the fluid replacement volume, that is, 1000 ml of fluid is replaced in the first hour, and the urine volume is 20 ml; within the second hour, the urine volume is produced at 10% of the fluid replacement volume, that is, 1500 ml of fluid is replaced in 2 hours, and the total urine volume is 150 ml. After the third hour, the urine volume is produced at 30% of the fluid replacement volume, that is, 2500 ml of fluid is replaced in 6 hours, and the total urine volume is 750 ml. In Virtual Case Intermediate ①, Virtual Case Intermediate ③, Virtual Case Intermediate ④, and Virtual Case Advanced ①, within the first hour, urine output is calculated as 1% of the amount of fluid administered. That is, if 1000 ml of fluid is administered in the first hour, urine output is 10 ml. Within the second to third hours, urine output is calculated as 5% of the amount of fluid administered. That is, if 1500 ml of fluid is administered in two hours, urine output is 75 ml. After the third hour, urine output is calculated as 30% of the amount of fluid administered. If 2500 ml of fluid is administered in the sixth hour, urine output is 750 ml. In Virtual Case Advanced ②, during shock, the shock manifestations are displayed. After shock is resolved, urine output is calculated as 5% of the amount of fluid administered in the first two to three hours. After the fourth hour, urine output is calculated as 30% of the amount of fluid administered. 3) Re-blood sampling trigger subunit: When a nurse fails to properly shake the blood sample tube during collection, the laboratory reports 20 minutes later: The blood sample is coagulated and the blood sample needs to be re-drawn. 4) Syringe Pump Alarm Trigger Subunit: 5 minutes after the completion of an infusion, if venous blood backflows and coagulation occurs, the syringe will not be able to deliver the fluid properly due to increased pressure, and a pressure alarm will sound. 5) Infusion Access Reestablishment Trigger Subunit: If blood backflow into the venous access indicates coagulation, the infusion will not drip. In this case, the nurse needs to re-establish the access through venipuncture. If the time exceeds 5 minutes, the syringe pump will sound a pressure alarm to remind you.

[0084] Humanistic triggering unit: 1) Family members’ dissatisfaction with nursing triggering sub-unit: For example, when a nurse draws blood twice in a short period of time (blood drawing is ordered once and executed twice), it is triggered, and the AI family members complain to the nurse; 2) The worsening condition and dissatisfaction with treatment triggering sub-unit: ① It is triggered after the patient complains of abdominal pain three times, ② It is triggered after complications occur and the rescue is completed, and the AI family members complain to the doctor; 3) Family members’ concerns about the condition triggering sub-unit: ① The child is repeatedly hospitalized due to DKA, ② The patient forgets to take insulin, which causes DKA, and it is triggered after the nursing operation is completed after admission, and the AI family members complain to the patient.

[0085] 7. Training Module

[0086] The training module consists of an introduction unit, a reception unit, a rescue unit, and an outcome unit. Each unit follows a set virtual case, and the training module is introduced using the virtual case Elementary ① Typical diabetic ketoacidosis without complications adult version as an example. Figure 8 .

[0087] During the introduction session, the system prompted the training nurse to sit at the nurse station, put on VR goggles, and play a pre-set virtual case scenario (a patient's journey from onset to emergency room emergency treatment): an 18-year-old male with type 1 diabetes for four years, using a three-short and one-long insulin regimen (6 units of aspart insulin before each meal and 14 units of glargine insulin before bed). The patient had just won the college entrance examination and was celebrating with classmates on a five-day short trip. He often forgot to take his insulin while dining out, but simply increased his glargine dose to 18 units at night. A friend reminded him to remember to take his pre-meal insulin, and he replied, "It's okay, I've increased the dose of long-acting insulin at night, so it's fine in the short term." While out, he got caught in the rain at a scenic spot and caught a cold, but he didn't take it seriously. That same night, he returned home and developed a high fever. His family gave him cold medicine, and he rested at home. He hadn't taken any insulin because of a poor appetite, fatigue, nausea, and vomiting, which had made him feel like eating. Three days later, the family gradually noticed that the patient loved to sleep and felt that something was wrong. They immediately called 120 and sent the patient to the emergency department at 20:10 at night. After the emergency doctor saw the patient, the nurse took the finger blood sugar test at 20:15: "hi", and put the patient on ECG monitoring and oxygen, and collected arterial and venous blood. The test results showed that the blood sugar was 35mmol / l, blood ketone bodies ≥3.2mmol / l, pH 7.25, and white blood cell count was 16.64x10 9 / l, serum potassium: 5.61mmol / l. The patient was drowsy, had an orbital pressure reflex, pupils of equal size and roundness, 3mm in diameter, and were responsive to light. T: 38.9°C, HR 120 beats / min, R: 30 beats / min, with a "rotten apple odor." Blood pressure: 90 / 60mmHg. The patient had a flushed complexion, high skin temperature, dry skin on the limbs, and sunken eyeballs. At 20:30, the nurse established intravenous access according to the doctor's instructions and infused 1000ml of 0.9% NS. An insulin intravenous pump (0.9% NS 50ml + insulin 50u, pumped in at 4ml / h) was installed. The male doctor performed urinary catheterization, but failed to drain out approximately 1ml of dark yellow urine (a section of urine was drained from the catheter, but there was no urine in the urine bag). At 8:50 PM, the emergency nurse contacted the ward and informed them of an 18-year-old male patient with lethargy, DKA, and oxygen, along with an IV, an insulin pump, and a urinary catheter. He would be transported to the ward in approximately 10 minutes. At the end of the call, the system prompted: "Remove the VR goggles and prepare supplies." The training nurse prepared the appropriate rescue equipment and a spare bed based on the emergency call. After the call, the training nurse announced: "Supplies prepared." The system prompted: "Return to the nurses' station and put on the VR goggles again." A virtual emergency patient admission scene was played: The patient, accompanied by support staff and family members, was admitted to the ward at 9:00 PM and brought to the bedside. The on-duty doctor and nurse immediately arrived and worked together to transfer the patient to the bed. At the end of the call, the system prompted: "Remove the VR goggles! The patient has arrived. Hurry and see him."

[0088] In the admission unit, the AI support staff conducts a handover with the training nurse, accompanied by voice notifications and patient status displays. The training nurse arrives at the bedside. The AI support staff then hands over the patient's admission card, stating that the patient has been placed on the bedside table. The patient is carrying an IV bag containing 1000ml of 0.9% NS (500ml remaining) and a pump containing 50ml of 0.9% NS plus 50 units of insulin at 4ml / h. Both are placed on the bed. The patient is receiving emergency monitoring and oxygen, and has a urinary catheter. After reviewing the information, you provide a voice confirmation: "Handover Completed." System notifications are also provided: "Infusion Completed" triggers the dynamic infusion tablet system, "Injection Pump Completed" triggers the dynamic syringe pump tablet system, and "ECG Monitoring Completed" triggers the dynamic ECG display unit. The tablet above the patient's head displays the patient's dynamic status, showing both overall and local conditions. "Urinary Catheter Completed" triggers the dynamic urine bag display unit. "Finger Blood Glucose Completed" triggers the patient to touch the dynamic blood glucose tablet with their left hand. The system time is displayed next to the patient's bed number and advances rapidly, so remember to check it regularly. The training nurse hangs up the intravenous infusion, checks the local infusion situation (right lower limb), adjusts the drip rate of the infusion plate to 334 drops / min (the system setting is 20 drops / min = 1 ml), and the training nurse says "infusion operation completed" (the first infusion operation is not included in the operation). The system voice prompts: turn off the piston and touch the left plate on the infusion stand to adjust the infusion drop rate; install the syringe pump, and say "syringe pump operation completed" (the first syringe pump operation is not included in the operation). The system voice prompts: turn off the syringe pump and touch the right plate on the infusion stand to adjust the pump rate to 4 ml / h; connect oxygen at 3 l / min, and say "oxygen operation completed" (the first oxygen operation is not included in the operation), hang the drainage bag, turn on the drainage switch, and say "urine catheter operation completed" (the first urine catheter operation is not included in the operation). The system voice prompts: there is a dynamic finger blood glucose display unit on the urine bag hanging place. The urine volume can be seen by touching the screen. The maximum volume is 1000 ml. When the urine is full of 1000 ml, the nurse needs to give a voice prompt: "Empty the urine bag." The ECG monitor is mounted on the headrest. A voice message reads, "ECG monitoring operation completed." The system prompts: Turn off the ECG monitor. Touch the tablet above the monitor to view the patient's Holter status. The ECG monitor displays: Heart rate 110-120 beats / min, Blood pressure 90-95 / 60-65 mmHg, R: 28-30 beats / min, SpO2 97-98% (with oxygen). Simultaneously, the system prompts: The tablet above the patient's head can be used to view the patient's overall and local condition. The training nurse completes the check and a voice message reads, "Handover completed." Intermittently, the AI doctor can be heard asking questions about the patient's condition and the AI family member responding. The AI doctor verbally orders a finger glucose test. The system prompts: The Holter tablet is on the patient's left hand.The training nurse performs a finger glucose test on the patient's left finger. Upon completion, the voice prompt "Finger glucose test completed" is heard. The nurse then touches the tablet on the patient's left hand, displaying the current finger glucose level. The nurse informs the AI doctor: 21:30 finger glucose level: 29.3mmol / l. The AI family member reminds the patient: "The infusion is complete." The AI doctor then responds: "Enter 1000ml of 0.9% NS." The training nurse repeats the test once, gets confirmation, and then returns to the treatment room to retrieve the medication for infusion, adjusting the drip rate to 334 drops / min. The training nurse then voice prompts: "The infusion is complete." The AI doctor tells the training nurse to write a medical order. The system reminds the nurse: Patient information must be entered as soon as possible. The training nurse returns to the nurse's station and enters the virtual patient's information: Bed 19, Zhang xx, 18 years old, DKA. A printed patient wristband is brought to the bedside and placed on the patient's left wrist.

[0089] Rescue Unit: Trained nurses perform admission care, measure vital signs, conduct nursing examinations, and take body temperatures. The trained nurse will voice: "Temperature measurement complete," and the system will automatically respond with 38.7°C. Trained nurses will then perform fall and bed fall assessments, pressure ulcer risk assessments, self-care assessments, Glasgow coma assessments, tube slippage assessments, nutritional risk assessments, and deep vein thrombosis assessments based on the patient's condition directly on the mobile tablet nursing support unit. Assessment results are directly imported into the nursing record. Upon inquiry from family members, the nurses will complete the admission care homepage and non-surgical nursing record form. Patient weight: 60 kg. The system will remind them at 10:00 PM: "Medical order has arrived." The trained nurse will voice back: "Received." The system will no longer remind them. The nurse will return to the nurses' station to process the order: verify, save, and proofread. After entering the password, the nurse will print out the order execution form and medication label. The nurse will then enter the inspection system to prepare blood collection tubes (binding with barcodes) and write the bed number and name. The doctor ordered a fingerstick blood glucose test every 1 hour. The patient brought medications, blood collection tubes, and a mobile tablet and wheeled the treatment cart to the bedside. The patient scanned the patient's wristband barcode or QR code, then the medication QR code, using the medical support unit on the mobile tablet. The tablet then announced, "Order verification successful," and the infusion medication and volume were automatically entered into the nursing record corresponding to the virtual time. Before venous and arterial blood collection, the patient's QR code was scanned, followed by the barcode on the tube. The tablet system displayed a check mark for each item (the system automatically recorded the collection time). The trained nurse then checked that all check marks were checked, then proceeded to draw blood (the trained nurse announced, "Venous blood collection completed," "Arterial blood collection completed") and notified the support system by phone to send the blood for testing. If the nurse failed to shake the tube properly, they would receive a call from the AI laboratory 5 minutes after sending it for testing, stating that the blood test had clotted and the item had been returned and the blood collection was repeated. The trained nurse then reattached the tube and collected blood. During the patient's blood collection, the AI family member was dissatisfied with the second blood draw, leading to an argument. The trained nurse reassured the patient. Based on the drip count set by the trained nurse, the system should complete the infusion by 22:30. If the nurse fails to promptly detect the patient's infusion, the infusion will not drip for more than 5 minutes. If this continues for more than 5 minutes, the syringe pump will trigger a pressure alarm due to increased infusion resistance caused by blood coagulation at the puncture site. After the trained nurse clears the alarm, she will re-insert the infusion needle. Upon completion, the trained nurse will announce, "Intravenous infusion completed." Based on the patient's weight, the total infusion volume was set at 6000 ml. 1000 ml of NS was infused in the first hour. At 21:30, 10 ml of urine was produced. Another 1000 ml of NS was infused in the second hour. At 22:30, fingerstick blood sugar was 25.3 mmol / l, and urine output was 200 ml. The ECG monitor showed a heart rate of 100-110 beats / min, blood pressure of 100-120 / 60-70 mmHg, R: 25-28 beats / min, and Spo2 of 97-98% (with oxygen).The patient's eyeballs recovered. At 10:00 PM, the system indicated the test results were available. The training nurse opened the mobile tablet to view the results: serum potassium: 5.1 mmol. She notified the AI doctor, who replied, "Received." About a minute later, the system notified her of a new order (1500 ml of 0.9% NS + 45 ml of 10% KCl, or 4.5 g of potassium, to be administered intravenously; another intravenous blood draw at 10:30 PM). The training nurse replied, "Received." She returned to the nurses' station to process the order, dispense medication, and change the infusion at the designated location, adjusting the drip rate to 167 ml / h. (The training nurse announced, "Change infusion completed.") Around 10:00 PM, the patient's temperature was taken. The training nurse announced, "Temperature measurement completed," and the system automatically returned a reading of 38.7°C. At 11:00 PM, the system indicated the test results were available. The training nurse opened the mobile tablet to view the results: serum potassium: 4.5 mmol, white blood cell count: 11.64 x 10. 9 / l, notifying the AI doctor, who responded with a response. About 1 minute later, the system prompted a new order (0.9% NS 100ml + cephalexin 2.0g, intravenous drip, with another venous and arterial blood draw at 0:30). The training nurse replied, "Received." The AI family member was asked if they had any drug allergies. After receiving the AI family member's response: "No," the patient returned to the treatment room to get medication and prepare for venous blood draw. At 23:30, fingerstick blood sugar: 21.3mmol / l, urine volume: 750ml. The training nurse used the downtime to observe changes in the patient's condition and write nursing records. Based on the training nurse's operational nodes and the progress of the virtual case, the system prompted whether time acceleration was needed. Upon confirmation, the time acceleration module was activated. At 00:00, the AI family member requested another temperature measurement. The training nurse took the patient's temperature, and the training nurse announced, "Temperature measurement complete." The system automatically responded with a temperature reading of 38.1°C. The ECG monitor indicated a heart rate of 90-100 beats / min, blood pressure of 110-120 / 60-70 mmHg, R: 22-26 beats / min, and Spo2 of 97-98% (with oxygen). At 0:30, the infusion was changed (to a cephalosporin), and the infusion rate was adjusted to 100 drops / min. (The training nurse announced, "Infusion change complete.") The infusion was expected to be completed by 0:50. At 0:30, the fingerstick blood glucose reading at 4 hours was 17.3 mmol / l, and the urine output was 900 ml. After notifying the doctor about the blood glucose reading, the doctor ordered an adjustment to the insulin intravenous pump rate of 2 ml / h. The procedure was complete, and the training nurse announced, "Injection pump operation complete." Venous and arterial blood samples were drawn and sent for testing. At 1:00, the system indicated the test results were available. The training nurse opened the mobile tablet to view the results: serum potassium: 4.35 mmol, serum ketones: 1.7 mmol / l, and pH: 7.4. At 1:30 a.m. on the 5th hour, the fingerstick blood sugar level was 13.3 mmol / l, and the urine volume was 990 ml. If the nurse noticed this, they would immediately empty the urine bag and record the urine volume in the nursing record. If not, the system would remind the training nurse 2 minutes later that the urine bag was almost full. The patient regained consciousness. After notifying the doctor about the blood sugar level, the doctor ordered an adjustment to the insulin intravenous pump rate: 1 ml / h. The saline + potassium infusion was suspended and the order was revised to 1000 ml of 5% GS + 6 units of insulin + 30 ml of 10% KCl, or 3 g of potassium, via intravenous drip, adjusting the infusion rate to 60 drops / min. The training nurse discarded approximately 300 ml of the saline + potassium solution and replaced it with a sugar + insulin + potassium solution. At 2:30 a.m., the fingerstick blood sugar level was 9.3 mmol / l, and the urine volume was 1080 ml (due to the emptying of the urine bag, the urine volume tablet showed 90 ml). At this time, the ECG monitor shows: the patient's heart rate is 90-100 beats / min, blood pressure is 100-120 / 60-70 mmHg, R: 20-22 beats / min, Spo297-98% (under oxygen inhalation). System voice: rescue is over.

[0090] Ending Unit: The system prompts you to return to the nurses' station and put on the VR goggles again to view the patient's outcome: The patient is out of critical condition, switched to a subcutaneous pump on the second day, and her temperature returns to normal on the fifth day. On the sixth day, the pump is removed and the patient is switched back to the "Three Short, One Long" regimen. The diabetes specialist nurse begins systematic health education on the third day, and the patient is discharged on the seventh day. After viewing the virtual case's ending, the system prompts you to remove the VR goggles and discuss your feelings and reflections on the training. After the training nurse finishes speaking, the system reminds you: "Training is over, clean up!" The training nurse returns her supplies to their designated locations.

[0091] 8. Training nurse data acquisition module

[0092] The training nurse data acquisition module consists of four components: motion trajectory sensors, wristbands, a recording unit, and direct data extraction from the training system. The motion trajectory sensors are attached to the mid-forearms of both upper limbs and ankles of both lower limbs, recording limb motion trajectory data, such as movement route and speed, while each nurse performs a virtual case. The wristbands record physiological data from each nurse at different times, particularly heart rate and blood pressure data during emergency situations. The recording unit, equipped with a camera and recording system, is installed in training areas such as the nurse station and emergency room. It records facial expressions, procedure steps, duration, and eye tracking (recording the nurse's observation path and frequency) during specific situations, such as emergency situations and arguments. The recording unit also records the number of voice tremors, use of humanistic vocabulary, effectiveness of communication content, and the maintenance of reflection logs. During the entire training, the training system will record relevant data, which can be directly extracted: such as the number of times the system reminds xx operation, the time from the issuance of medical orders to the execution of medical orders on patients, the number of nursing errors, the recording of rescue and training time, the recording of operation items and time of different training nurses, and the extraction of nursing assessment data, nursing diagnosis data and nursing record content data from nursing documents. For details, see Figure 9 .

[0093] 9. Training Analysis Module

[0094] The training analysis module includes operation analysis unit, thinking analysis unit, humanities analysis unit and teamwork analysis unit. Figure 10 .

[0095] The operation analysis unit includes: ① Motion trajectory analysis system, which uses sensors to record the three-dimensional data of the training nurse's limb movement trajectory, movement speed, etc. in real time, and compares it with the pre-stored standard movement trajectory of nursing experts, analyzes in detail the incorrect operations of the training nurse, and marks the operation time. Pay special attention to whether there is any disorder in the movement trajectory, and analyze it in combination with the physiological data of the training nurse at that time; ② Operation connotation analysis system: compare with the pre-stored standard nursing expert operation steps to determine whether there are any omissions in the operation steps, whether the operation sequence is correct, and mark the relevant problems; Manual operation scoring: focus on the system operation standard area, and deduct points according to the hospital nursing department's operation process regulations (the full score is 100 points). If there are identical operations, calculate according to the average score, the total score of all operations ÷ the number of operations = the average score of the operation (that is, the score of this training operation).

[0096] Thinking Analysis Unit: ① Eye Tracking Analysis System: Utilizing an eye-tracking camera to monitor the trainee nurse's level and frequency of observation of the five tablets in the dynamic condition simulation module, this is compared with the eye tracking trajectory of the nursing expert to analyze whether the trainee nurse's attention is distributed appropriately. This reflects whether the trainee nurse's observation timing and key points are grasped correctly. If the trainee nurse's attention to the tablets in the patient's dynamic condition display unit is insufficient, it can be inferred that the nurse is not paying enough attention to the patient's overall and local conditions. ② Physiological Data Analysis System: Observes fluctuations in the trainee nurse's heart rate and blood pressure during the rescue operation, noting the periods of fluctuation. Whether these fluctuations correspond to periods of operational errors or irregular movement patterns reflects the trainee nurse's psychological quality and the need for improvement in their emergency response capabilities. ③ Analysis system for training reminders, errors, order delays, and emergency training duration: This system counts the number and types of system reminders issued to nurse trainees during training; the number and types of nursing errors by nurse trainees; and a delay of more than 5 minutes between order issuance (the time the system notifies the nurse trainees of an order and begins recording it) and order processing (the time it takes the nurse trainees to process the order on their computer). This system also counts the number of order delays, emergency training duration (the time it begins recording when a patient experiences an emergency condition such as cerebral edema, hypokalemia, or pulmonary edema, and ends when the patient is successfully or unsuccessfully rescued), and total training duration (from the start of training to the end of training). ④ Voice tremor analysis system: This system captures periods of voice tremor in nurse trainees during specific sessions and analyzes the nurse trainees' psychological stress tolerance in conjunction with physiological data at that time. ⑤ Facial expression analysis system: This system extracts facial expressions during emergency training and infers the nurse trainees' emotional state by analyzing facial features such as eye and mouth movements and changes in facial muscles. A deep learning-based convolutional neural network (CNN) model was used to extract facial expression features from images and analyze the psychological state of the training nurses. 6. Nursing Document Analysis System: This system identifies whether nursing assessments and diagnoses are consistent with the nursing expert's assessments, whether there are errors in the nursing record (reflecting the training nurse's meticulousness), and whether the nursing record is comprehensive compared to the nursing expert's. Omissions are also annotated. The annotated information is then reported to a manual platform for verification. 7. Reflection Log Analysis System: This system uses noise reduction to convert the training nurses' recordings into text. The system performs structured annotation: cognitive level, emotional level, and practical relevance. Analysis steps: 1) Cognitive Development Analysis: Reflection Level = 0.4 x Concept Reference + 0.3 x Self-Criticism + 0.3 x Improvement Strategy. 2) Emotional Evolution Tracking: This system uses the BERT sentiment analysis model to generate a phased emotional fluctuation map. 3) Practice Transformation Evaluation: See Table 1.

[0097] Table 1 ARCS motivation model mapping

[0098] Dimensions Typical sentences Quantitative indicators attention "This failure made me start to pay attention to..." Frequency of mentions of key concepts Relevance "This reminds me of Chapter 5 of the textbook..." Theory-practice conjunction density Self-confidence "Next time I encounter a similar situation I will..." The proportion of first-person active sentences Satisfaction "Once you've mastered this assessment..." Frequency of positive emotion words

[0099] Finally, the system writes a reflection report, and the score is the reflection level score × 20 points. For example, the reflection level score is 0.4 × 20 points = 8 points, and the highest score is 20 points.

[0100] Humanistic Analysis Unit: ① Facial Expression Analysis System: This system extracts communication data between trainee nurses and family members to analyze the trainee nurses' psychological states. For example, this system extracts and analyzes facial expressions during arguments with family members, analyzing the trainee nurses' psychological states in specific environments. ② Humanistic Vocabulary Analysis System: This system extracts the voice of trainee nurses during humanistic communication with family members, processes it through noise reduction, converts it into text, and extracts and categorizes core care vocabulary. The core care vocabulary classification is shown in Table 2.

[0101] Table 2 Classification of core care vocabulary

[0102]

[0103]

[0104] The Humanities Analysis Unit will produce a paper report, including facial expression analysis and classification of core care vocabulary used by the trained nurses, along with a count of the number of times they were used. Finally, nursing experts will confirm the analysis results based on the video recordings.

[0105] The Team Collaboration Analysis Unit analyzes communication efficiency based on internal team communication data. ① Collaboration Communication Content Analysis System: Utilizes Natural Language Processing (NLP) technology to analyze team communication content. Communication efficiency analysis is shown in Table 3, with a total score of 60 points.

[0106] Table 3 Communication efficiency analysis

[0107]

[0108] As mentioned earlier, the nurse training data acquisition module records operational data during virtual case training. Therefore, when conducting group virtual case training, the lead rescue nurse can be analyzed and labeled. ② Analysis system for each member's operation ratio and rescue efficiency: See Table 4.

[0109] Table 4: Analysis of each member's operation ratio and rescue efficiency

[0110]

[0111]

[0112] ③ Each member's movement trajectory analysis system compares the movement trajectory with that of the nursing expert team. For example, if there are three nurses in training, the system automatically calculates the movement trajectory lengths of Nurse A (meters), Nurse B (meters), and Nurse C (meters). Scoring formula: Nurse A's movement trajectory length (meters) / Nursing expert A's movement trajectory length (meters) x 100%. Beginner level: ≥200%, Intermediate level: 150%-199%, Advanced level: ≤149%. Beginner level: 5 points, Intermediate level: 10 points, Advanced level: 15 points. See Table 5 for the total training score.

[0113] Table 5 Total training score

[0114]

[0115]

[0116] Training analysis graphic report: motion trajectory graphic analysis, eye tracking trajectory graphic analysis, physiological data trend chart, voice tremor sound wave graph and time period, facial expression analysis chart and humanistic care core vocabulary list, and intelligent analysis conclusion report.

[0117] 10. Cross-platform interface module

[0118] The cross-platform interface module is used to manage multiple cross-platform interfaces so that the virtual simulation critical care training system for diabetic ketoacidosis in the embodiment of the present invention can be connected to various hardware devices such as VR glasses, vests, bracelets, motion track sensors, computers, tablets, etc., and can also be connected to the hospital HIS system. Figure 11 Optionally, the cross-platform interface may be a wired interface such as USB, RS232 / RS485, or a wireless interface such as Bluetooth, WiFi, or Zigbee, which is not limited in the embodiment of the present invention.

[0119] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the concept of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A virtual simulation critical care training system for diabetic ketoacidosis, characterized by: The system comprises: The dynamic condition simulation module dynamically simulates a virtual emergency case of a diabetic ketoacidosis patient by combining the dynamic ECG monitoring display unit, the patient dynamic status display unit, the dynamic infusion display unit, the dynamic urine bag display unit, and the dynamic finger blood glucose display unit; Virtual case module, used for hierarchical management of virtual cases of emergency treatment for patients with diabetic ketoacidosis; A scenario triggering module is used to simulate multiple types of virtual scenarios during virtual case training, including virtual disease scenarios, virtual treatment scenarios, virtual nursing scenarios, and virtual humanities scenarios; The training module is used to guide nurses to perform operations during virtual case training.

2. The system according to claim 1, wherein: A Holter monitor display unit, configured to continuously display the patient's Holter monitor according to predetermined virtual case settings, and to display the monitor on a split screen in specific circumstances; The dynamic status monitoring display unit is used to display the patient's whole body or local status. The local status includes: head: facial expression, complexion, pupils, nasal breathing, oral cavity, oxygen inhalation status, ventilator status, endotracheal intubation status, nasogastric feeding tube status; pupil: click the light source to project the pupil special effect: pupil dilation, click to remove the light source to display the pupil special effect; chest: shortness of breath, obvious chest rise and fall; abdomen; limbs: local skin status; perineum; sacrum; The dynamic infusion display unit virtually displays the patient's dynamic infusion status and syringe pump status according to the infusion speed and syringe pump speed adjusted by the trained nurse, and adjusts the speed; Dynamic urine bag display unit, used to display the urine color, properties, empty and full status, and urine volume in the urine bag; Dynamic finger blood glucose display unit, used to display finger blood glucose value.

3. The system according to claim 1, wherein: The levels of virtual cases managed in the virtual case module include elementary, intermediate and advanced. Each virtual case is bound through the setting module, and the corresponding virtual case is output during training; Primary: Virtual Case of Typical Diabetic Ketoacidosis without Complications in Adults and Children; Intermediate: Virtual case of diabetic ketoacidosis combined with heart failure in the elderly, virtual case of diabetic ketoacidosis in pregnancy; virtual case of diabetic ketoacidosis in pregnancy, virtual case of diabetic ketoacidosis using SGLT-2 inhibitors; Advanced: Virtual case of cardiac arrest caused by hypokalemia in diabetic ketoacidosis, and virtual case of diabetic ketoacidosis combined with shock.

4. The system according to claim 1, wherein: The scene triggering module includes: Condition triggering units include blood glucose change triggering subunit, hypokalemia ECG monitoring triggering subunit, cerebral edema triggering subunit, acute pulmonary edema triggering subunit, shock triggering subunit, emergency cesarean section triggering subunit, and DKA symptom relief triggering subunit; Treatment trigger unit, including fluid therapy plan doctor order change trigger sub-unit, gastrointestinal fluid therapy doctor order change trigger sub-unit, insulin injection pump therapy doctor order change trigger sub-unit, potassium supplementation doctor order change trigger sub-unit, acidosis correction doctor order change trigger sub-unit, infection control doctor order trigger sub-unit, and complication doctor order trigger sub-unit; The humanistic trigger unit includes the trigger sub-unit caused by family members’ dissatisfaction with nursing care, the trigger sub-unit caused by worsening of the disease and dissatisfaction with treatment, and the trigger sub-unit caused by family members’ concerns about the disease.

5. The system according to claim 1, wherein: The training modules include: The introduction unit is used to guide nurses in training by putting on VR goggles while playing predetermined virtual case scenarios; In the reception unit, AI support personnel and trained nurses conduct shift handovers, while voice message reminders and patient status displays are also provided; Rescue unit, used to assist in training nurses in rescue scene operations; The outcome unit is used to prompt the trained nurse to put on the VR goggles again to check the patient's outcome.

6. The system according to any one of claims 1 to 5, characterized in that The system also includes a medical care support module for simulating a clinical medical care support system; The medical care support module includes: The medical support unit includes a medical order support subunit, an examination support subunit, and a test support subunit. By pre-setting virtual cases, the medical order support subunit gradually displays the medical order as virtual time advances. The examination support subunit is used to print out examination sheets and view the examinations and reports that the virtual patient has undergone. The test support subunit is used to bind blood collection tubes and view the tests and reports that the virtual patient has undergone. The nursing support unit includes a nursing assessment subunit and a nursing document subunit; the nursing assessment subunit is used to conduct fall and bed fall assessment, pressure ulcer risk assessment, self-care ability assessment, Glasgow coma assessment, tube slip assessment, mood thermometer assessment, pain assessment, nutritional risk assessment, restraint needs assessment, and deep vein thrombosis assessment; the nursing document subunit is used to manage admission assessment, nursing diagnosis, and non-surgical nursing record sheets.

7. The system according to any one of claims 1 to 5, characterized in that The system further comprises a time acceleration module, which is used to accelerate or stop time upon receiving an instruction from the training nurse during the virtual case training process.

8. The system according to any one of claims 1 to 5, characterized in that The system further includes a training nurse data acquisition module for acquiring data generated by the training nurses during the virtual case training process; the trainee data acquisition module includes: A motion trajectory unit is used to record the motion trajectory data of the trainee nurse's limbs during the virtual case training through sensors set on the trainee nurse's limbs; The wristband unit is used to record the physiological data of nurses in training at different time periods; The audio and video recording unit is used to record the communication, operation data, facial expression data and eye tracking data of the training nurses during the virtual case training process.

9. The system according to claim 8, characterized in that The system also includes a training analysis module for analyzing the acquired training nurse data and generating an analysis report; The training analysis module includes: an operation analysis unit for performing comparative analysis based on the motion trajectory data, operation steps and standard data; The thinking analysis unit is used to analyze attention distribution based on eye tracking data; analyze physiological quality based on physiological data during rescue; analyze the psychological stress resistance of nurses in training based on the period of voice tremor in specific links and physiological data at that time; and analyze emotional and psychological states based on facial expression data; A humanities analysis unit is used to extract communication data between the training nurses and their families, analyze the training nurses' psychological state, and screen and classify caring words based on the communication data, count the number of times they are used, and generate an analysis report; The teamwork analysis unit is used to analyze communication efficiency based on internal team communication data, and to analyze the operation proportion and rescue efficiency of each team member.

10. The system according to any one of claims 1 to 5, characterized in that The system further comprises a setting module for setting user information and training information; The cross-platform interface module is used to manage multiple cross-platform interfaces so as to utilize the cross-platform interfaces to perform platform docking with various hardware devices.

Images