CN112906205B - Virtual learning method for total hip replacement surgery - Google Patents

Abstract

The invention discloses a virtual learning method for total hip replacement surgery, which comprises the steps of initializing a system, establishing communication connection between an input end and a controller, and completing mapping between a force feedback device and the controller in a real-time virtual scene; in the operation process, the input end interacts with the surgical instrument model in the virtual scene in a three-dimensional graphic input control mode, and simultaneously, an operator can timely see a training picture through image feedback and is provided with key information prompts in the training process. The invention can obtain vivid and visual perception, improve the understanding of the abstract principle, relieve the problems of expensive and tense medical and operation resources in various regions of China and reduce the operation risk.

Description

A virtual learning method for total hip replacement surgery

technical field

The invention relates to the technical field of total hip replacement, in particular to a virtual learning method for total hip replacement.

Background technique

The development of surgery has gone through three stages: open surgery, minimally invasive manual surgery, and robot-assisted surgery. Surgery is developing towards a more miniaturized, dexterous and intelligent direction.

Open surgery is traditional surgery, in which bones and internal organs are exposed by cutting from the surface of the body. The advantage of open surgery is that the doctor can fully understand the anatomical information of the lesion area directly through the naked eye, and have a good grasp of the operation. However, the disadvantages of this kind of surgery are large wound, heavy pain, heavy bleeding, long hospital stay and slow wound healing, and even potentially increase the risk of postoperative complications. Minimally invasive surgery, that is, minimally invasive surgery. It is mainly performed by using laparoscopy, thoracoscopic and other modern medical instruments and related equipment. Minimally invasive surgery has been widely used in surgical treatment due to its advantages of small wound, less blood loss and short hospitalization period. However, manual minimally invasive surgery has problems such as low positioning accuracy, space and environmental constraints, intraoperative fatigue, and hand tremors caused by long working hours. Robot-assisted surgery has become one of the international frontier research hotspots.

With the development of automation and robotics, surgical robots have begun to penetrate into all aspects of surgical planning, minimally invasive positioning, and non-invasive treatment. The surgical robot improves the surgeon's accuracy by filtering out hand tremors, scaling hand movements into smaller movements, and controlling the manipulator to complete a series of surgical operations. However, in robot-assisted surgery, the doctor's operating habits are different from traditional surgery, mainly because the master-slave control mode is not as flexible and convenient as pure manual operation. Therefore, the physician's operating skills need repeated training before he can proficiently control the robot. Common training modes include machine-computer interaction mode and virtual learning mode. The human-computer interaction mode is to directly control the joystick to manipulate the robot to complete the specified action, which requires high cost and cannot meet the characteristics of diversification of surgical objects in a single system, so it cannot be widely used. The virtual learning system is to improve the proficiency in advance, and then transition to the human-computer interaction mode to complete the follow-up operations. The virtual learning system is based on computer software and hardware, assisted by related technical means, and through the simulation of the known or unknown world, people can truly feel an advanced computer application technology. The virtual trainer provides a multi-source information fusion and interactive 3D dynamic visualization and physical behavior interaction system simulation, which is of great significance to the surgeon's surgical training, surgical effect prediction, surgical plan formulation and navigation.

Currently available technologies:

1) The digital virtual simulation dental training method is used for oral clinical skill training. It is a novel and effective dental experiment teaching method to evaluate the effect through system operation completion time, target completion degree, error scoring, and comprehensive scoring.

2) The virtual learning method of mandibular surgery can realize multiple operation modes such as grasping, cutting, drilling, and suturing at the user operation end. For cutting and drilling operations, the system renders the best planned path in the virtual scene in real time and displays the progress of the operation as a percentage. When the error rate exceeds 10%, track the cutting line in time and declare failure.

3) The learning method of cardiomyocyte surgery based on tactile feedback plays a key role in training surgeons' hand-eye coordination and ability to perform high-precision injection tasks.

4) The virtual orthopedic surgery simulator learning method enables surgeons to perform fracture replacement surgery training on a personal computer without using any expensive hardware equipment.

However, there is currently no virtual learning method applied to total hip arthroplasty under master-slave control mode, resulting in the failure of the operation if there is a mistake in the real operation situation.

Contents of the invention

The purpose of the present invention is to overcome the above problems in the prior art and provide a virtual learning method for total hip replacement surgery.

In order to achieve the above object, the technical scheme provided by the present invention is:

A virtual learning method for total hip replacement surgery. First, the system is initialized, the communication connection between the input terminal and the controller is established, and the mapping between the force feedback device and the controller is completed in the real-time virtual scene; during the operation , the input terminal interacts with the surgical instrument model in the virtual scene through the control mode of three-dimensional graphic input, and at the same time allows the operator to see the training screen in time through image feedback, and is equipped with key information prompts during the training process.

Further, the specific process is as follows:

S1. Connect the force feedback device and the controller;

S2. Install "Unity 5Haptic Plugin for GeomagicOpenHaptics 3.3" in Unity5.2.3 or above to communicate with the force feedback device;

S3. Call the programming interface of "Haptic Plugin for Geomagic OpenHaptics 3.3" to obtain the relevant parameters of the force feedback device;

S4. Realize the real-time interaction between the force feedback device and the virtual device set in Unity through 3D graphic input;

S5. Select the attitude angle input method to complete data exchange with the device;

S6. Initialize the workspace and work mode of the plug-in;

S7. Continuously update the working space according to the position of the camera, and connect the virtual force feedback device with the obtained value to complete real-time communication;

S8, establishing the bone model in the virtual scene;

S9, creating a surgical scene and surgical instruments in the virtual scene;

S10. Setting virtual key points on the training object model;

S11. Design image feedback information;

S12, designing a camera focusing function;

S13. Start training and evaluation.

Further, in the step S3, the programming interface is divided into two parts: Haptic Device API and Haptic LibraryAPI;

Among them, the Haptic Device API provides the underlying interface of the device, and directly accesses the real-time parameters of various devices in the device status query table through the callback function;

The Haptic Library provides an upper-layer interface. During operation, the host computer program obtains the information of the force feedback device including position, attitude, joint angle, terminal velocity, and joint angular velocity at a frequency of 1000 Hz through the programming interface, and continuously sends the information Implement closed-loop control for the controller.

Further, in the step S5, in the process of communication based on 3D graphics input and force feedback, the device initialization is completed through the "hdInitDevice" function provided by the plug-in, wherein the default parameter is "HD_DEFAULT_DEVICE"; then use the "hdGetCurrentDevice" function to find the current device and The "hdGetDoublev" callback function reads device parameters, which include four types of current translation position information, current translation velocity and acceleration information, and current attitude angle; the calling form of "hdGetDoublev" is "(parameter name, return value type) "; Then use "hdBeginFrame" and "hdEndFrame" as the starting point and end point of data access to ensure the consistency of access, so as to realize the synchronous update of the data of the force feedback device and the controller; in the test, for the position and speed including translation Information including acceleration is directly read through the access interface provided by the Haptic Device API.

Further, for the acquisition of attitude angle information, the 16-element terminal attitude array is pre-obtained through the Haptic Device API access interface, and then converted to the speed information of each axis in the equivalent axis-angle coordinate system; expressed in the equivalent axis-angle coordinate system The described 16-element terminal pose array is shown in the following formula (1):

Among them, R _K (θ) is the identity matrix, and the elements of the matrix are called 16 elements, each of which is represented by t[], K is the attitude angle of the current attitude relative to the base coordinate system, and the attitude angle can be approximated by difference at x, Velocity information in the three directions of y and z;

Formula 2 and Formula 3 can be obtained from 1:

θ is the corresponding rotation angle around the coordinate axis;

Further, the step S6 is specifically as follows:

The active space of the force feedback device in the real scene is approximated as a cube, and its size data is converted from the "float3" array to "IntPtr" through the "ConvertFloat3ToIntPtr" command in "SetHapticWorkSpace";

Then use "ConvertIntPtrToFloat3" in "GetHapticWorkSpace" to convert "IntPtr" into the "float 3" array in the Unity editor to determine the space size;

Next, use the "UpdateHapticWorkspace" function in the plug-in to update the workspace and the "IndicateMode" function to set the training interaction mode according to the position of the camera;

In the next step, set the state of the created object to "Touchable Face" in Unity3D, which is any one of "front", "back", "front and back";

In the process of setting force feedback, create and set related properties in the scripts of "Environment constantforce", "viscosity", "spring effect" and "friction effect", including amplitude, duration, and gain; to set the object, you need to obtain all objects with "Tochable ″ tag object array, and get the mesh properties of the object, then draw the geometry, read the geometry features and start the force feedback events of all the different objects.

Further, said step S8 establishes the bone model in the virtual scene, which is divided into two steps of CT image segmentation and three-dimensional modeling;

Wherein, the CT image segmentation is based on the threshold method of mimics, and the image segmentation of the bone CT data is completed by region growing; after importing the patient's CT atlas in the minics interface, the threshold is selected for binarization, and the gray value of the CT atlas The pixels within the threshold range are retained; threshold segmentation mainly uses the difference in grayscale characteristics between the target area to be extracted in the image and its background, and divides the CT image into the target area and the background area to generate a corresponding binary image; then use the area The grow tool splits the binarized image into multiple blocks while removing floating pixels.

Further, in the step S9, the surgical environment is inserted in the form of textures, and the force feedback model is combined with the simple three-dimensional model that comes with Unity3D; a main camera is created in the "Hierarchy" panel for the shooting operation training process See the running picture, add the "Skybox" component in the "Inspector" to design the texture style, and switch the "Clear Flags" in the Camera component to the "Skybox" mode; the operation environment is divided into front, back, left, right, The three-dimensional space composed of the upper and lower six maps, and the virtual surgical tool is assembled from a sphere and a capsule, and the center of the sphere serves as the positioning coordinates of the entire surgical tool.

Further, in the step S12, the "Field of view" in the Camera component is set through a script in the camera, and the focus is adjusted in the form of a mouse button during operation.

Further, in the step S13, the experiment is mainly divided into two steps. During the operation of the Game panel, the focus is first adjusted so that the training screen is switched to the most comfortable angle for the operator; then the force feedback device is controlled to approach the operator at the correct angle. The key point is not completed until both key points turn blue; during the training process, the familiarity of the operator's training is evaluated by the time spent on the Game panel.

Compared with the existing technology, the principle and advantages of this scheme are as follows:

(1) Through the "OpenHaptics" toolkit, developers can apply force or haptic feedback devices to a wide range of fields, such as surgical simulation and medical training, aerospace and military training, assistive technology for the blind or visually impaired, and game entertainment .

(2) The coordinated operation of the force feedback device and the virtual device is completed based on 3D graphics input. Compared with the method based on keyboard character input, the input signal of the force feedback device can be flexibly collected and the interaction process can be completed quickly.

(3) Choose the attitude angle input method to complete the data exchange with the device. Compared with the translation input method, make full use of the advantages of the six degrees of freedom of the force feedback device to complete the data input.

(4) The quality of surgical instruments and the touch of virtual objects can be simulated in the learning system, so as to experience the surgical scene more realistically.

(5) The establishment of the virtual femoral head model is strictly based on the patient's CT image and the key point information on the bone is created in accordance with the current requirements of total hip replacement surgery. The entire training model is authentic.

(6) The training scene can be oriented to a variety of surgical environments, and the training complexity can also be easily set in Unity according to the difficulty in the actual process. The real-time display of time and distance in the image feedback link can give some path guidance during the user's operation, thereby assisting in the completion of training.

(7) In the virtual scene, the camera focus can be adjusted through the mouse button, which can easily change the picture information of the scene during the training process.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the services that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only For some embodiments of the present invention, those skilled in the art can also obtain other drawings based on these drawings without creative work.

1 is a schematic diagram of a virtual learning system for total hip replacement surgery;

Figure 2 is a schematic diagram of a scene based on mimics virtual bone model;

Fig. 3 is a schematic diagram of a virtual simulation training scene.

Detailed ways

The present invention will be further described below in conjunction with specific embodiment:

Referring to Figures 1-3, a virtual learning method for total hip replacement surgery. First, the system is initialized, the communication connection between the input terminal and the controller is established, and the force feedback device and the controller are completed in the real-time virtual scene. Mapping of the controller; during the operation, the input terminal interacts with the surgical instrument model in the virtual scene in the control mode of three-dimensional graphics input, and at the same time allows the operator to see the training screen in time through image feedback, and during the training process Equipped with key information prompts.

The communication between the input terminal and the controller and the process of mapping the virtual scene mainly use the OpenHaptics interface to access the data of the force feedback device, package it into a plug-in and pass it to the Unity3D thread, so as to complete the Unity interface control.

The specific process is as follows:

S1. Connect the force feedback device and the controller; use the provided USB interface to connect the force feedback device and the controller, and then connect the power cord to the socket and the connector on the back of the force feedback device for power supply. After the connection is completed, if the blue LED flashes twice quickly, it means that the controller and the force feedback device are successfully connected.

S2. Install "Unity 5Haptic Plugin for GeomagicOpenHaptics 3.3" on Unity5.2.3 or above to communicate with force feedback devices.

S3. Call the programming interface of "Haptic Plugin for Geomagic OpenHaptics 3.3" to obtain the relevant parameters of the force feedback device. The programming interface is divided into two parts: Haptic Device API (HDAPI) and Haptic Library API (HLAPI).

HDAPI provides the underlying interface of the device, and can directly access the real-time parameters of various devices in the device status query table through the callback function;

HLAPI provides an upper interface for programmers familiar with OpenGL. During operation, the host computer program obtains information such as the position, attitude, joint angle, terminal velocity, and joint angular velocity of the force feedback device at a frequency of 1000 Hz through the programming interface, and continuously sends the information to the controller to achieve closed-loop control.

S4. The real-time interaction between the force feedback device and the virtual device set in Unity is realized through 3D graphic input.

S5. Select the attitude angle input method to complete data exchange with the device; specifically, in the process of 3D graphic input and force feedback communication, complete device initialization through the "hdInitDevice" function provided by the plug-in, where the default parameter is "HD_DEFAULT_DEVICE"; then use The "hdGetCurrentDevice" function finds the current device and the "hdGetDoublev" callback function reads the device parameters. The parameters include current translation position information, current translation velocity and acceleration information, and current attitude angle; the calling form of "hdGetDoublev" is " (parameter name, return value type)"; then use "hdBeginFrame" and "hdEndFrame" as the starting point and end point of data access to ensure the consistency of access, so as to realize the synchronous update of the data of the force feedback device and the controller; in the test, The information including the position, velocity and acceleration of the translation is directly read through the access interface provided by the Haptic Device API.

For the acquisition of attitude angle information, the 16-element terminal attitude array is pre-obtained through the Haptic Device API access interface, and then converted to the velocity information of each axis in the equivalent axis-angle coordinate system; The element end pose array is shown in the following formula (1):

Among them, R _K (θ) is the identity matrix, and the elements of the matrix are called 16 elements, each of which is represented by t[], K is the attitude angle of the current attitude relative to the base coordinate system, and the attitude angle can be approximated by difference at x, Velocity information in the three directions of y and z;

Formula 2 and Formula 3 can be obtained from 1:

θ is the corresponding rotation angle around the coordinate axis;

S6. Initialize the workspace and work mode of the plug-in, the specific process is as follows:

Approximate the active space of the force feedback device in the real scene as a cube, and convert its size data (for example, the length, width and height are 20cm) from the "float 3" array to "IntPtr" through the "ConvertFloat3ToIntPtr" command in "SetHapticWorkSpace" ;

Then use "ConvertIntPtrToFloat3" in "GetHapticWorkSpace" to convert "IntPtr" into the "float 3" array in the Unity editor to determine the space size;

Next, use the "UpdateHapticWorkspace" function in the plug-in to update the workspace and the "IndicateMode" function to set the training interaction mode according to the position of the camera;

In the next step, set the state of the created object to "Touchable Face" in Unity3D, which is any one of "front", "back", "front and back";

In the process of setting force feedback, create and set related properties in the scripts of "Environment constantforce", "viscosity", "spring effect" and "friction effect", including amplitude, duration, and gain; to set the object, you need to obtain all objects with "Tochable ″ tag object array, and get the mesh properties of the object, then draw the geometry, read the geometry features and start the force feedback events of all the different objects.

S7. Continuously update the workspace in the "void Update" loop. The working space is constantly updated according to the position of the camera, and the virtual force feedback device is connected with the obtained values to complete real-time communication.

S8, establishing the bone model in the virtual scene, specifically divided into two steps of CT image segmentation and three-dimensional modeling;

Wherein, the CT image segmentation is based on the threshold method of mimics, and the image segmentation of the bone CT data is completed by region growing; after importing the patient's CT atlas in the minics interface, the threshold is selected for binarization, and the gray value of the CT atlas The pixels within the threshold range are retained; threshold segmentation mainly uses the difference in grayscale characteristics between the target area to be extracted in the image and its background, and divides the CT image into the target area and the background area to generate a corresponding binary image; then use the area The grow tool splits the binarized image into multiple blocks while removing floating pixels.

In the next step, computational processing is performed on the segmented serial CT images to restore their three-dimensional structures. Commonly used 3D reconstruction algorithms are divided into two categories: surface rendering and volume rendering. The surface rendering method mainly extracts one or more tissue contours of interest on the image, and generates a grid model for subsequent processing with the help of intermediate geometric plane elements and algorithms. Algorithms for surface rendering mainly include fault surface reconstruction, moving cube method, etc. Among them, the fault surface reconstruction algorithm, as a classic algorithm for surface rendering, is the most widely used. Volume rendering method: directly use the collected voxels to resample through the algorithm, and then complete the 3D reconstruction. Commonly used volume rendering methods mainly include ray casting method, snowball throwing method, staggered transformation method and texture mapping method. The advantages of volume rendering can generate every detail of the 3D model, and has the characteristics of high image quality and easy parallel processing. The disadvantage is that the amount of data to be processed is large, the calculation time is long, and the algorithm is complex, which affects the efficiency of 3D reconstruction. Surface rendering only needs to process a small part of all data, so the calculation speed is fast; however, surface rendering cuts off the connection between the structure outline and the overall structure, and key point information may be lost. The virtual learning system proposed in this paper has relatively high requirements for the details of three-dimensional objects, so the bone model is modeled by volume rendering based on the ray-casting method. The training object model proposed in this patent is mainly composed of a virtual bone model created by mimics and a translucent human body model created by Solid work. Save the two models in "stl" format, and convert them into "obj" format in the 3D builder software and import them into the Unity3D scene.

S9, creating a surgical scene and surgical instruments in the virtual scene;

The surgical environment is inserted in the form of textures, and the force feedback model is combined using the simple three-dimensional model that comes with Unity3D; create a main camera in the "Hierarchy" panel to shoot the running picture seen during the operation training process, in the " Add the "Skybox" component in "Inspector" to design the texture style, and switch the "ClearFlags" in the Camera component to "Skybox" mode; the surgical environment is made up of six textures: front, back, left, right, up and down three-dimensional space, and the virtual surgical tool is assembled from a sphere and a capsule, wherein the center of the sphere serves as the positioning coordinates of the entire surgical tool.

S10. Setting virtual key points on the training object model;

In total hip replacement surgery, it is usually necessary to install a metal acetabular cup in the patient's acetabulum and fix it with 2-3 screws to connect with the femoral head prosthesis. However, the position of the acetabular cup and screws is the key to the whole operation, which directly affects the success or failure of the operation. According to surgical experience, Kirschner wires are usually inserted vertically at 2cm above the upper edge of the acetabulum in the direction of 12 o'clock (the bone plate is thicker here, and the Kirschner wires are not easy to shift) to position the acetabular cup. Therefore, two red sockets are designed in this embodiment. The virtual keypoints of are used to simulate Kirschner wire positioning. Since the key point is relatively small and difficult to observe in the Game panel, a sphere with this point as the center and a radius of 1cm is used for simulation, and the coordinates of the center of the circle are the position coordinates of the key point.

S11. Design image feedback information;

The image feedback information consists of the time display of the operation and the real-time distance of the surgical tool to the key point. The time display is updated at a fixed frequency in "fixedupdate", and the integer form is converted into components and seconds; when the time exceeds 60 seconds, the color will change from white to red to remind the user that the time exceeds expectations. The real-time distance prompt is obtained by the difference between the position coordinates of the surgical tool and the key points in "Position" in "Transform" during the movement process. When the distance between the virtual surgical instrument and the coordinates of the center of the circle is less than 3cm, let the displayed value change from red to green to prompt the operator to reach the goal immediately; The color of the sphere changes from red to blue to indicate that the target position has been reached.

S12. Design the camera focus function; considering that the scene in the Game panel is often not the best observation position for training, so create a new script in the camera to set the "Field of view" in the Camera component, and use the mouse during operation Press the button to adjust the focus.

S13. Start training and evaluation; the experiment is mainly divided into two steps. During the operation of the Game panel, firstly adjust the focus to switch the training screen to the most comfortable angle for the operator; then control the force feedback device to approach the key point at the correct angle , until both keypoints are blue. During the training process, the operator's familiarity with the training is evaluated by the time spent on the Game panel.

The implementation examples described above are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention. Therefore, all changes made according to the shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims (8)

1. A virtual learning method for total hip replacement surgery is characterized in that a system is initialized, communication connection between an input end and a controller is established, and mapping of a force feedback device and the controller is completed in a real-time virtual scene; in the operation process, the input end interacts with a surgical instrument model in a virtual scene in a three-dimensional graphic input control mode, and simultaneously an operator can see a training picture in time through image feedback and is provided with key information prompts in the training process;

the specific process is as follows:

s1, connecting a force feedback device and a controller;

s2, installing 'Unity 5Haptic plug for Geomagic OpenHaptics 3.3' in the version of Unity5.2.3 or above to communicate with the force feedback equipment;

s3, calling a programming interface of 'Haptic plug for Geomagic OpenHaptics 3.3' to acquire relevant parameters of the force feedback equipment;

s4, realizing real-time interaction between the force feedback equipment and virtual equipment set in the Unity through 3D (three-dimensional) graphic input;

s5, selecting an attitude angle input mode to complete data communication with equipment;

s6, initializing a working space and a working mode of the plug-in;

s7, continuously updating a working space according to the position of the camera, and completing real-time communication connection between the virtual force feedback equipment and the obtained numerical value;

s8, establishing a bone model in a virtual scene;

s9, creating an operation scene and an operation instrument in the virtual scene;

s10, setting virtual key points on a training object model;

s11, designing image feedback information;

s12, designing a camera focusing function;

s13, starting training and evaluating;

in the step S3, a programming interface is divided into a Haptic Device API and a Haptic Library API;

the Haptic Device API provides a bottom-layer interface of the equipment, and directly accesses real-time parameters of various equipment in the equipment state query table through a callback function;

the Haptic Library provides an upper layer interface, and in the operation process, an upper computer program obtains information of the force feedback equipment including position, attitude, joint angle, terminal speed and joint angular speed at the frequency of 1000hz through a programming interface and continuously sends the information to the controller to realize closed-loop control.

2. The virtual learning method for total hip replacement surgery as claimed in claim 1, wherein step S5, in the 3D graphical input and force feedback based communication process, DEVICE initialization is performed by the "hdInitDevice" function provided by the plug-in, wherein the DEFAULT parameter is "HD _ DEFAULT _ DEVICE"; then, finding the current equipment and reading equipment parameters by using an hdGetCurrentdevice function and an hdGetDoublev callback function, wherein the parameters comprise current translation position information, current translation speed and acceleration information and a current attitude angle; wherein the invocation form of the hdGetDoublev is "(parameter name, return value type)"; then using the 'hdBeginFrame' and the 'hdEndFrame' as a starting point and an end point of data access to ensure the consistency of the access, thereby realizing the synchronous update of the data of the force feedback equipment and the controller; in the test, information including the position, velocity and acceleration of the translation is read directly through the access interface provided by the Haptic Device API.

3. The virtual learning method for total hip replacement surgery as claimed in claim 2, wherein for the acquisition of the attitude angle information, a 16-element end attitude array is obtained in advance through a Haptic Device API access interface, and then converted into the speed information of each axis in the equivalent axis angular coordinate system; the 16-element terminal attitude array described by the equivalent axis angular coordinate system is shown as the following formula (1):

wherein R is _K (θ) is a unit matrix, the elements of the matrix are called 16 elements, each using t [ 2 ]]Expressing that K is the attitude angle of the current attitude relative to a base coordinate system, and obtaining the speed information of the attitude angle in the x, y and z directions through difference;

formulae (2) and (3) are obtainable from (1):

theta is the corresponding rotation angle around the coordinate axis;

4. the virtual learning method for total hip replacement surgery as claimed in claim 1, wherein the step S6 is as follows:

approximating the activity space of the force feedback device in a real scene to a cube form, converting the size data thereof from the "float3" array to "IntPtr" through the "convertfoat 3to IntPtr" instruction in "sethapticWorkSpace";

then, converting the IntPtr into a float3 array in a Unity editor by using ConvertIntPtrToFloat3 in GetHapticWorkSpace to determine the space size;

next, according to the position of the camera, using an 'UpdateHapticWorkspace' function in the plug-in to update a working space and an 'indicateMode' function to set a training interaction mode;

next, setting the created object state to be "touchbleface" which is any one of "front", "back", "front and back" in Unity 3D;

in the process of setting force feedback, relevant attributes including amplitude, duration and gain are created and set in scripts of 'Environment constant force', 'vision', 'spring effect' and 'massage effect'; setting an object, wherein the object array with the 'desk' tag needs to be acquired, acquiring the grid attribute of the object, then drawing a geometric body, and reading the characteristics of the geometric body so as to start the force feedback events of all different objects.

5. The virtual learning method for total hip replacement surgery as claimed in claim 1, wherein the step S8 establishes a bone model in a virtual scene, and is divided into two steps of CT image segmentation and three-dimensional modeling;

wherein the CT image segmentation is based on a threshold method of mimics, and the region growth is used for completing the image segmentation of the bone CT data; after a CT atlas of a patient is imported into a minics interface, selecting a threshold value to carry out binarization processing, and reserving pixels of which the gray values are within the threshold value range in the CT atlas; the threshold segmentation mainly utilizes the difference of a target area to be extracted in an image and the background thereof on the gray characteristic to divide a CT image into the target area and the background area so as to generate a corresponding binary image; and then, segmenting the binary image into a plurality of blocks by using a region growing tool, and removing floating pixels.

6. The virtual learning method for total hip replacement surgery as claimed in claim 1, wherein in step S9, the surgical environment is inserted in a map form, and the force feedback model is combined by using a simple stereo model owned by Unity 3D; creating a main Camera in a 'Hierarchy' panel for shooting a running face painting seen in the operation training process, adding a 'Skybox' component in an 'observer' to design a chartlet style, and switching 'ClearFlags' in a Camera component to a 'Skybox' mode; the operation environment is a three-dimensional space formed by splicing front, back, left, right, upper and lower six sticking pictures, the virtual operation tool is formed by assembling a sphere and a capsule, and the circle center of the sphere is used as the positioning coordinate of the whole operation tool.

7. The virtual learning method for total hip replacement surgery as claimed in claim 1, wherein step S12, the Camera assembly is set up with "Field of view" by script in the Camera, and focusing is performed by mouse button during operation.

8. The virtual learning method for total hip replacement surgery according to claim 1, wherein in step S13, the experiment is mainly performed in two steps, and during the operation of the Game panel, focusing is performed first to switch the training screen to the angle most comfortable for the operator; then controlling the force feedback equipment to approach the key points at a correct angle until the two key points turn blue; the familiarity of the operator training is assessed by the time spent on the Game panel during the training process.

Images