A VR-based nursing training system

Abstract

The present invention relates to the field of VR training technology, and in particular to a VR-based nursing training system, comprising: a central processing module, a virtual environment modeling module, a pain quantification module, a care resistance simulation module, an OpenPose posture recognition module and a VR component, wherein the pain quantification module constructs a human pain sensitivity atlas set, and the care resistance simulation module constructs a care resistance action set. When a user operation error occurs, the pain quantification module generates a pain index according to the human pain sensitivity atlas set and scene and interaction data, and transmits the pain index to the care resistance simulation module. When the pain index is transmitted, the care resistance simulation module randomly selects a corresponding care resistance action from the care resistance action set. The present invention helps users learn how to correctly care for patients with cognitive impairment who have care resistance behavior by organically combining VR technology, biosimulation technology and medical evaluation.

Description

Technical Field

[0001] The present invention relates to the field of VR training technology, and in particular to a VR-based nursing training system. Background Art

[0002] The nursing training system is a platform that uses modern information technology to provide training, learning and practice for nursing students. It aims to help learners master nursing knowledge and skills and improve nursing quality and patient safety by simulating various scenarios in real medical environments. Existing nursing training systems are often conducted in simulated environments, only scoring the accuracy of nurses' operations, and lack simulation of patients' feelings and reactions. In addition, in the actual nursing process, nurses may encounter care resistance behaviors of patients with cognitive impairment. Existing training systems fail to effectively simulate and train to deal with these situations, resulting in excessive restraint and excessive drug intervention in the nursing practice of patients with cognitive impairment. Summary of the invention

[0003] The present invention overcomes the shortcomings of the prior art and provides a VR-based nursing training system. By constructing a human pain sensitivity atlas set and a care resistance action set, the system simulates the possible care resistance reactions of patients with cognitive impairment during the care process. The present invention organically combines VR technology, biosimulation technology and medical evaluation to help users learn how to correctly care for patients with cognitive impairment who have care resistance behavior.

[0004] A VR-based nursing training system, comprising: a central processing module, a virtual environment modeling module, a pain quantification module, a care resistance simulation module, an OpenPose posture recognition module and a VR component;

[0005] The VR component is connected in communication with the virtual environment modeling module, the OpenPose posture recognition module, and the central processing module; the virtual environment modeling module is connected in communication with the central processing module, the pain quantification module, the care resistance simulation module, and the OpenPose posture recognition module; and the pain quantification module is connected in communication with the care resistance simulation module;

[0006] The virtual environment modeling module uses Unity3D to construct a simulation training space, and inputs the simulation training space data generated in the simulation training space into the VR component and the pain quantification module, so that users learn how to properly care for cognitively impaired patients with care-resistant behaviors in the simulation training space;

[0007] The pain quantification module uses the geodesic square inverse distance interpolation method to construct a human pain sensitivity atlas set, and the pain quantification module generates a pain index based on the human pain sensitivity atlas set and the simulation training space data;

[0008] The care resistance simulation module constructs a care resistance action set. When the pain index is input, the care resistance simulation module randomly selects the corresponding care resistance action from the care resistance action set.

[0009] Further, the VR component captures the user's hand posture data and the user interaction data, the VR component receives the simulation training space data from the virtual environment modeling module and presents it to the user and transmits the hand posture data and the user interaction data to the OpenPose posture recognition module;

[0010] The OpenPose posture recognition module receives hand posture data and user interaction data in real time, collects user videos in real time, samples the user videos using the OpenPose library and extracts skeleton key points, generates skeleton key point data, and refines the skeleton key points in combination with the hand posture data and user interaction data, and transmits the skeleton key point data and user interaction data to the virtual environment modeling module.

[0011] Furthermore, the process in which the user learns how to properly care for a cognitively impaired patient with care-resistant behavior in a simulation training space includes the following steps:

[0012] Step S1: The user sets the simulation training space parameters, including scene settings, patient types, medical devices, nursing processes, and care resistance response measures;

[0013] Step S2: the virtual environment modeling module constructs a simulation training space according to the simulation training space parameters;

[0014] Step S3: The user uses medical devices to care for the patient in the simulation training space. When the user makes an error in operation, the pain quantification module generates a pain index based on the human pain sensitivity atlas and the simulation training space data, and transmits the pain index to the care resistance simulation module;

[0015] Step S4: when the pain index is input, the care resistance simulation module randomly selects a corresponding care resistance action from the care resistance action set;

[0016] Step S5: the user uses the care resistance coping means to respond to the care resistance, and provides care for the patient in the simulation training space under the care resistance environment;

[0017] Step S6: The central processing module calculates the RTC-DAT score according to the RTC-DAT scale and calculates the total score according to the pain index, RTC-DAT score, care resistance coping methods and the correctness of the nursing process.

[0018] Furthermore, the process of constructing a human pain sensitivity atlas by the pain quantification module includes the following steps:

[0019] Step P1: Collect QST test data of different types of patients, extract the pain location, the total number of pain locations in the QST test data and the pain threshold from the QST test data, and normalize the pain threshold to 0-1 and map it to pain sensitivity;

[0020] Step P2: construct a three-dimensional human body model, map the pain location and pain sensitivity to the surface of the three-dimensional human body model, apply the geodesic square inverse distance interpolation method to calculate the pain sensitivity of each point on the surface of the three-dimensional human body model, generate a human body pain sensitivity map, and repeat this step until each type of patient has a corresponding human body pain sensitivity map;

[0021] Step P3: Summarize the human pain sensitivity atlases of different types of patients into a human pain sensitivity atlas set.

[0022] Furthermore, the process of calculating the pain sensitivity of each point on the surface of the three-dimensional human body model by applying the geodesic square inverse distance interpolation method to generate a human body pain sensitivity map specifically includes the following steps:

[0023] Step P21: randomly selecting a point from the surface of the human body three-dimensional model as an interpolation point;

[0024] Step P22: extracting the geodesic distance from each pain location to the interpolation point;

[0025] Step P23: Calculate the interpolation weight of each pain location for the interpolation point. The calculation formula of the interpolation weight of each pain location for the interpolation point is as follows:

[0026] ;

[0027] in, Representative The interpolation weight of the pain location for the interpolation point, Representative The geodesic distance from the pain location to the interpolation point, n represents the total number of pain locations in the QST test data;

[0028] Step P24: Apply the geodesic square inverse distance interpolation method to calculate the pain sensitivity of each point on the surface of the three-dimensional human body model. The formula of the geodesic square inverse distance interpolation method is as follows:

[0029] ;

[0030] in, represents the pain sensitivity at the interpolation point, Representative pain sensitivity at each pain site, represents the pain sensitivity of the pain location closest to the interpolation point;

[0031] Step P25: Repeat steps P21 to P24 until each point on the surface of the three-dimensional human body model has corresponding pain sensitivity.

[0032] Furthermore, the process of constructing the care resistance action set by the care resistance simulation module includes the following steps:

[0033] Step R1: Collect videos of different care resistance actions, and use the OpenPose library to extract the skeleton key point data of different care resistance actions from the videos of different care resistance actions;

[0034] Step R2: According to the RTC-DAT scale, the care resistance intensity and care resistance duration of the skeletal key point data of different care resistance actions are marked, the correct care resistance response measures are marked, and the mapping relationship between the pain index and the care resistance intensity is set, and summarized into a care resistance action set.

[0035] Furthermore, the process of generating a pain index based on the human pain sensitivity atlas and the simulation training space data specifically includes the following steps:

[0036] Step S31: extracting the error impulse and the error impulse action area from the simulation training space data, and extracting the human pain sensitivity map corresponding to the patient type from the human pain sensitivity map set;

[0037] Step S32: Calculate the pain index according to the human pain sensitivity map corresponding to the error impulse, the error impulse action area and the patient type. The calculation formula of the pain index is as follows:

[0038] ;

[0039] in, represents the pain index, Represents the total number of points within the error impulse action area, Represents the first The pain sensitivity of each point, Represents error impulse.

[0040] Furthermore, step S4 specifically includes the following steps:

[0041] Step S41: generating the current care resistance intensity according to the mapping relationship between the pain index and the care resistance intensity;

[0042] Step S42: screening the care resistance actions whose care resistance strength is equal to the current care resistance strength from the care resistance action set, and randomly selecting a care resistance action as the current care resistance action, and recording the care resistance strength and care resistance duration of the current care resistance action;

[0043] Step S43: Map the skeleton key point data of the current care resistance action to the patient in the simulation training space.

[0044] Furthermore, step S6 specifically includes the following steps:

[0045] Step S61: Formulate a nursing process accuracy scoring rule, and calculate the user's nursing process accuracy score according to the nursing process accuracy scoring rule;

[0046] Step S62: Calculate the RTC-DAT score of each care resistance action according to the RTC-DAT scale, add up the RTC-DAT scores of the care resistance actions for which the user takes correct care resistance coping measures as the correct coping score, and add up the RTC-DAT scores of the care resistance actions for which the user fails to take correct care resistance coping measures as the incorrect coping deduction score;

[0047] Step S63: Calculate the total score using the total score formula, which is as follows:

[0048] ;

[0049] in, Represents the total score, Represents the correctness score of the nursing process. Represents the number of times user operation errors occurred. Representative The pain index caused by the user's operation error, Represents the correct response score, Points will be deducted for incorrect responses.

[0050] The beneficial effects of the present invention are as follows:

[0051] (1) The present invention provides a nursing training system based on VR, comprising: a central processing module, a virtual environment modeling module, a pain quantification module, a care resistance simulation module, an OpenPose posture recognition module and a VR component. The pain quantification module constructs a human pain sensitivity atlas set, and the care resistance simulation module constructs a care resistance action set. If a user operation error occurs, the pain quantification module generates a pain index based on the human pain sensitivity atlas set and the simulation training space data, and transmits the pain index to the care resistance simulation module. When the pain index is transmitted, the care resistance simulation module randomly selects a corresponding care resistance action from the care resistance action set. The present invention helps users learn how to correctly care for patients with cognitive impairment who have care resistance behavior by organically combining VR technology, biosimulation technology and medical evaluation.

[0052] (2) The geodesic square inverse distance interpolation method provided by the present invention is based on the texture and curvature characteristics of the human body surface, and uses the geodesic distance to represent the distance between points on the body surface. Based on medical prior knowledge, it overcomes the technical bias of the inverse distance interpolation method using the Euclidean distance. Inspired by the inverse square law, the interpolation weights are formulated according to the inverse square of the geodesic distance to represent the nonlinear attenuation of the reference value of the pain position. By allocating the remaining interpolation weights to the nearest pain position, it is achieved that while retaining the characteristic that the sum of the interpolation weights of the original inverse distance interpolation formula is 1, the influence of the local pain position on the interpolation point is enhanced, so that the human pain sensitivity map is closer to the actual human pain sensitivity distribution, thereby improving the authenticity of the pain simulation.

[0053] (3) The human pain sensitivity atlas proposed in the present invention is based on the fact that different diseases may distort the patient's perception of pain. For example, diabetic peripheral neuropathy caused by diabetes mellitus is a type of diabetic neuropathy. The corresponding human pain sensitivity atlas is customized according to the QST test data of different types of patients, thereby improving the authenticity of pain simulation and helping users to learn how to reduce unnecessary pain as much as possible during the care process for different types of patients.

[0054] (4) The calculation formula of the pain index proposed in the present invention is based on the error impulse, the total number of points in the error impulse action area and the pain sensitivity of each point in the error impulse action area. Since the impulse is the product of force and force action time, and the number of points in the three-dimensional model can be used to characterize the size of the area, the calculation formula of the pain index is actually used to characterize the product of pressure, pressure action time and pain sensitivity of each point. The impulse is used here because the collision detection mechanism of Unity3D uses relative linear velocity and impulse to characterize the interaction between contact pairs. This formula compromises the physical simulation mechanism of Unity3D to better simulate the pain occurrence mechanism in reality and help users learn how to reduce stimulation to patients during the nursing process.

[0055] (5) The total score formula proposed in the present invention incentivizes the correctness of the nursing process and the correct response to nursing resistance, allowing users to remedy operational errors by correctly responding to nursing resistance. This overcomes the technical bias of the nursing training system that only incentivizes the correctness of the nursing process and encourages users to learn how to correctly respond to nursing resistance. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a module diagram of a VR-based nursing training system proposed by the present invention;

[0057] Figure 2 A flowchart for users to learn how to properly care for cognitively impaired patients with care-resistant behaviors in a simulation training space;

[0058] Figure 3 This is a schematic diagram of the geodesic square inverse distance interpolation method proposed in the present invention.

[0059] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the examples given are only used to explain the present invention, and are not used to limit the scope of the present invention.

[0061] Example 1, see Figure 1 , a VR-based nursing training system, comprising: a central processing module, a virtual environment modeling module, a pain quantification module, a care resistance simulation module, an OpenPose posture recognition module and a VR component;

[0062] The VR component is connected in communication with the virtual environment modeling module, the OpenPose posture recognition module, and the central processing module; the virtual environment modeling module is connected in communication with the central processing module, the pain quantification module, the care resistance simulation module, and the OpenPose posture recognition module; and the pain quantification module is connected in communication with the care resistance simulation module;

[0063] The virtual environment modeling module uses Unity3D to construct a simulation training space, and inputs the simulation training space data generated in the simulation training space into the VR component and the pain quantification module, so that users learn how to properly care for cognitively impaired patients with care-resistant behaviors in the simulation training space;

[0064] The pain quantification module uses the geodesic square inverse distance interpolation method to construct a human pain sensitivity atlas set, and the pain quantification module generates a pain index based on the human pain sensitivity atlas set and the simulation training space data;

[0065] The care resistance simulation module constructs a care resistance action set. When the pain index is input, the care resistance simulation module randomly selects the corresponding care resistance action from the care resistance action set.

[0066] Embodiment 2: This embodiment is based on the above embodiment. The VR component captures the user's hand posture data and user interaction data. The VR component receives the simulation training space data from the virtual environment modeling module and presents it to the user, and transmits the hand posture data and user interaction data to the OpenPose posture recognition module.

[0067] The OpenPose posture recognition module receives hand posture data and user interaction data in real time, collects user videos in real time, samples the user videos using the OpenPose library and extracts skeleton key points, generates skeleton key point data, and refines the skeleton key points in combination with the hand posture data and user interaction data, and transmits the skeleton key point data and user interaction data to the virtual environment modeling module.

[0068] Example 3, see Figure 2 Based on the above embodiment, this embodiment includes the following steps in which a user learns how to properly care for a cognitively impaired patient with care-resistant behavior in a simulation training space:

[0069] Step S1: The user sets the simulation training space parameters, including scene settings, patient types, medical devices, nursing processes, and care resistance coping measures. In this embodiment, the scene is set as a ward, the patient type is diabetes, the medical device is a disposable blood collection bag, the nursing process is blood sample collection, and the care resistance coping measures include comfort toys, restraint belts, and anti-anxiety drugs;

[0070] Step S2: the virtual environment modeling module constructs a simulation training space according to the simulation training space parameters;

[0071] Step S3: The user uses medical devices to care for the patient in the simulation training space. When the user makes an error in operation, the pain quantification module generates a pain index based on the human pain sensitivity atlas and the simulation training space data, and transmits the pain index to the care resistance simulation module;

[0072] Step S4: when the pain index is input, the care resistance simulation module randomly selects a corresponding care resistance action from the care resistance action set;

[0073] Step S5: the user uses the care resistance coping means to respond to the care resistance, and provides care for the patient in the simulation training space under the care resistance environment;

[0074] Step S6: The central processing module calculates the RTC-DAT score according to the RTC-DAT scale and calculates the total score according to the pain index, RTC-DAT score, care resistance coping methods and the correctness of the nursing process.

[0075] Example 4, see Figure 3 This embodiment is based on the above embodiment, and the process of constructing a human pain sensitivity atlas set by the pain quantification module includes the following steps:

[0076] Step P1: Collect QST test data of different types of patients, extract the pain location, the total number of pain locations in the QST test data and the pain threshold from the QST test data, and normalize the pain threshold to 0-1 and map it to pain sensitivity;

[0077] Step P2: construct a three-dimensional human body model, map the pain location and pain sensitivity to the surface of the three-dimensional human body model, apply the geodesic square inverse distance interpolation method to calculate the pain sensitivity of each point on the surface of the three-dimensional human body model, generate a human body pain sensitivity map, and repeat this step until each type of patient has a corresponding human body pain sensitivity map;

[0078] Step P3: Summarize the human pain sensitivity atlases of different types of patients into a human pain sensitivity atlas set.

[0079] Embodiment 5: Based on the above embodiment, this embodiment uses the geodesic square inverse distance interpolation method to calculate the pain sensitivity of each point on the surface of the three-dimensional human body model, and the process of generating a human body pain sensitivity map specifically includes the following steps:

[0080] Step P21: randomly selecting a point from the surface of the human body three-dimensional model as an interpolation point;

[0081] Step P22: extracting the geodesic distance from each pain location to the interpolation point;

[0082] Step P23: Calculate the interpolation weight of each pain location for the interpolation point. The calculation formula of the interpolation weight of each pain location for the interpolation point is as follows:

[0083] ;

[0084] in, Representative The interpolation weight of the pain location for the interpolation point, Representative The geodesic distance from the pain location to the interpolation point, n represents the total number of pain locations in the QST test data;

[0085] Step P24: Apply the geodesic square inverse distance interpolation method to calculate the pain sensitivity of each point on the surface of the three-dimensional human body model. The formula of the geodesic square inverse distance interpolation method is as follows:

[0086] ;

[0087] in, represents the pain sensitivity at the interpolation point, Representative pain sensitivity at each pain site, represents the pain sensitivity of the pain location closest to the interpolation point. In this embodiment, the pain sensitivity of the first pain location is 0.5, the pain sensitivity of the second pain location is 0.55, the pain sensitivity of the third pain location is 0.3, and the pain sensitivity of the fourth pain location is 0.3. The pain sensitivity of the interpolation point is finally 0.4635;

[0088] Step P25: Repeat steps P21 to P24 until each point on the surface of the three-dimensional human body model has corresponding pain sensitivity.

[0089] Embodiment 6: This embodiment is based on the above embodiment. The process of the care resistance simulation module constructing the care resistance action set includes the following steps:

[0090] Step R1: Collect videos of different care resistance actions, and use the OpenPose library to extract the skeleton key point data of different care resistance actions from the videos of different care resistance actions;

[0091] Step R2: According to the RTC-DAT scale, mark the care resistance intensity and care resistance duration of the skeletal key point data of different care resistance actions, mark the correct care resistance response measures, set the mapping relationship between the pain index and the care resistance intensity, and summarize them into a care resistance action set. In this embodiment, a pain index below 10 corresponds to mild care resistance, a pain index between 10-30 corresponds to moderate care resistance, and a pain index above 30 corresponds to severe care resistance.

[0092] Embodiment 7, based on the above embodiment, this embodiment generates a pain index according to a human pain sensitivity atlas and simulation training space data, including the following steps:

[0093] Step S31: extracting the error impulse and the error impulse action area from the simulation training space data, and extracting the human pain sensitivity map corresponding to the patient type from the human pain sensitivity map set. In this embodiment, the unit of the error impulse is Newton-second;

[0094] Step S32: Calculate the pain index according to the human pain sensitivity map corresponding to the error impulse, the error impulse action area and the patient type. The calculation formula of the pain index is as follows:

[0095] ;

[0096] in, represents the pain index, Represents the total number of points within the error impulse action area, Represents the first The pain sensitivity of each point, Represents error impulse.

[0097] Embodiment 8: This embodiment is based on the above embodiment, and step S4 specifically includes the following steps:

[0098] Step S41: generating the current care resistance intensity according to the mapping relationship between the pain index and the care resistance intensity;

[0099] Step S42: screening the care resistance actions whose care resistance strength is equal to the current care resistance strength from the care resistance action set, and randomly selecting a care resistance action as the current care resistance action, and recording the care resistance strength and care resistance duration of the current care resistance action;

[0100] Step S43: Mapping the skeleton key point data of the current care resistance action to the patient in the simulation training space. In this embodiment, the mapping of the skeleton key point data of the current care resistance action to the patient in the simulation training space is achieved through the Mecanim animation system of Unity3D.

[0101] Embodiment 9: This embodiment is based on the above embodiment, and step S6 specifically includes the following steps:

[0102] Step S61: Formulate a nursing process accuracy scoring rule, and calculate the user's nursing process accuracy score according to the nursing process accuracy scoring rule. In this embodiment, the nursing process accuracy scoring rule is formulated according to "Basic Nursing";

[0103] Step S62: Calculate the RTC-DAT score of each care resistance action according to the RTC-DAT scale, add up the RTC-DAT scores of the care resistance actions for which the user takes correct care resistance coping measures as the correct coping score, and add up the RTC-DAT scores of the care resistance actions for which the user fails to take correct care resistance coping measures as the incorrect coping deduction score;

[0104] Step S63: Calculate the total score using the total score formula, which is as follows:

[0105] ;

[0106] in, Represents the total score, Represents the correctness score of the nursing process. Represents the number of times user operation errors occurred. Representative The pain index caused by the user's operation error, Represents the correct response score, Points will be deducted for incorrect responses.

[0107] Embodiment 10: This embodiment is based on embodiment 9. In step S5, the user resists the patient's mild care and chooses to use a restraint belt. Due to excessive restraint, it is recorded as an incorrect response and points are deducted in step S63.

[0108] Embodiment 11: This embodiment is based on embodiment 9. In step S5, the user resists the mild care of the patient and chooses to use a comfort toy. Since the correct response is recorded as a correct response score in step S63.

[0109] Embodiment 12. This embodiment is based on the above embodiment and runs in the Windows operating system environment. The simulation training space is constructed by Unity3D. The central processing module, pain quantification module, care resistance simulation module, OpenPose posture recognition module and virtual environment modeling module are implemented by a The HP Omen 16 and its connected DCAM710 camera. The VR components include HTC Vive and NoitomHi52.0 motion capture gloves.

[0110] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

[0111] The present invention and its implementation methods are described above. Such description is not restrictive. What is shown in the accompanying drawings is only one of the implementation methods of the present invention. The actual structure is not limited thereto. In short, if ordinary technicians in this field are inspired by the present invention and do not deviate from the purpose of the invention, they can creatively design structural methods and embodiments similar to the technical scheme, which should all fall within the scope of protection of the present invention.

Claims

1. A VR-based nursing training system, characterized by: It includes central processing module, virtual environment modeling module, pain quantification module, care resistance simulation module, OpenPose posture recognition module and VR component; The VR component is connected in communication with the virtual environment modeling module, the OpenPose posture recognition module, and the central processing module; the virtual environment modeling module is connected in communication with the central processing module, the pain quantification module, the care resistance simulation module, and the OpenPose posture recognition module; and the pain quantification module is connected in communication with the care resistance simulation module; The virtual environment modeling module uses Unity3D to build a simulation training space, and inputs the simulation training space data generated in the simulation training space into the VR component and the pain quantification module, so that the user learns how to care in the simulation training space; The pain quantification module uses the geodesic square inverse distance interpolation method to construct a human pain sensitivity atlas set, and the pain quantification module generates a pain index based on the human pain sensitivity atlas set and the simulation training space data; The care resistance simulation module constructs a care resistance action set. When the pain index is input, the care resistance simulation module randomly selects the corresponding care resistance action from the care resistance action set. The process of constructing a human pain sensitivity atlas set by the pain quantification module includes the following steps: Step P1: Collect QST test data of different types of patients, extract the pain location, the total number of pain locations in the QST test data and the pain threshold from the QST test data, and normalize the pain threshold to 0-1 and map it to pain sensitivity; Step P2: construct a three-dimensional human body model, map the pain location and pain sensitivity to the surface of the three-dimensional human body model, apply the geodesic square inverse distance interpolation method to calculate the pain sensitivity of each point on the surface of the three-dimensional human body model, generate a human body pain sensitivity map, and repeat this step until each type of patient has a corresponding human body pain sensitivity map; Step P3: Summarizing the human pain sensitivity atlases of different types of patients into a human pain sensitivity atlas set; The process of calculating the pain sensitivity of each point on the surface of the human body three-dimensional model by applying the geodesic square inverse distance interpolation method and generating the human body pain sensitivity map specifically includes the following steps: Step P21: randomly selecting a point from the surface of the human body three-dimensional model as an interpolation point; Step P22: extracting the geodesic distance from each pain location to the interpolation point; Step P23: Calculate the interpolation weight of each pain location for the interpolation point. The calculation formula of the interpolation weight of each pain location for the interpolation point is as follows: ； in, Representative The interpolation weight of the pain location for the interpolation point, Representative The geodesic distance from the pain location to the interpolation point, Represents the total number of pain locations in the QST test data; Step P24: Apply the geodesic square inverse distance interpolation method to calculate the pain sensitivity of each point on the surface of the three-dimensional human body model. The formula of the geodesic square inverse distance interpolation method is as follows: ； in, represents the pain sensitivity at the interpolation point, Representative pain sensitivity at each pain site, represents the pain sensitivity of the pain location closest to the interpolation point; Step P25: Repeat steps P21 to P24 until each point on the surface of the three-dimensional human body model has corresponding pain sensitivity.

2. A VR-based nursing training system according to claim 1, characterized in that: The process by which users learn how to properly care for patients with cognitive impairment and care-resistant behaviors in a simulation training space includes the following steps: Step S1: The user sets the simulation training space parameters, including scene settings, patient types, medical devices, nursing processes, and care resistance response measures; Step S2: the virtual environment modeling module constructs a simulation training space according to the simulation training space parameters; Step S3: The user uses medical devices to care for the patient in the simulation training space. When the user makes an error in operation, the pain quantification module generates a pain index based on the human pain sensitivity atlas and the simulation training space data, and transmits the pain index to the care resistance simulation module; Step S4: when the pain index is input, the care resistance simulation module randomly selects a corresponding care resistance action from the care resistance action set; Step S5: the user uses the care resistance coping means to respond to the care resistance, and provides care for the patient in the simulation training space under the care resistance environment; Step S6: The central processing module calculates the RTC-DAT score according to the RTC-DAT scale and calculates the total score according to the pain index, RTC-DAT score, care resistance coping methods and the correctness of the nursing process.

3. The VR-based nursing training system according to claim 1, characterized in that: The process of constructing the care resistance action set by the care resistance simulation module includes the following steps: Step R1: Collect videos of different care resistance actions, and use the OpenPose library to extract the skeleton key point data of different care resistance actions from the videos of different care resistance actions; Step R2: According to the RTC-DAT scale, the care resistance intensity and care resistance duration of the skeletal key point data of different care resistance actions are marked, the correct care resistance response measures are marked, and the mapping relationship between the pain index and the care resistance intensity is set, and summarized into a care resistance action set.

4. The VR-based nursing training system according to claim 2, characterized in that: The process of generating a pain index based on a human pain sensitivity atlas and simulated training space data specifically includes the following steps: Step S31: extracting the error impulse and the error impulse action area from the simulation training space data, and extracting the human pain sensitivity map corresponding to the patient type from the human pain sensitivity map set; Step S32: Calculate the pain index according to the human pain sensitivity map corresponding to the error impulse, the error impulse action area and the patient type. The calculation formula of the pain index is as follows: ； in, represents the pain index, Represents the total number of points within the error impulse action area, Represents the first The pain sensitivity of each point Represents error impulse.

Images