A virtual reality-based medical procedure training system and method

Abstract

The application discloses a kind of medical operation training system and method based on virtual reality, comprising: training gloves, each finger joint outer wall is equipped with response module detection direction data, motion data and gravity data;Angular motion detection unit real-time detects angular motion data;Photographing unit real-time shoots eyeball movement picture;First processing unit is obtained according to angular motion data and eyeball movement picture first adjustment instruction and second adjustment instruction;Second processing unit is obtained according to each direction data, each motion data and each gravity data hand real-time motion model;Comparison unit compares hand real-time motion model with hand standard motion model and obtains deviation result;Adjustment unit is obtained according to deviation result prompting instruction, and the visual angle of operator is adjusted according to first adjustment instruction and second adjustment instruction;Stimulation circuit stimulates finger joint position according to prompting instruction and reminds.The application improves the precision of operator's medical operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical operation training equipment, and particularly relates to a medical operation training system and method based on virtual reality. BACKGROUND

[0002] Virtual reality technology includes computer, electronic information, simulation technology, and its basic implementation mode is mainly computer technology, which utilizes and integrates three-dimensional graphics technology, multimedia technology, simulation technology, display technology, servo technology and other latest developments of various high-techs, and generates a virtual world with realistic three-dimensional vision, touch, smell and other sensory experiences by means of computers and other devices, so that a person in the virtual world has a sense of being there. In the traditional medical operation training process, a surgical dummy, an animal carcass or a donor's body is generally used as a surgical training prop, but the surgical dummy, the animal carcass or the donor's body has a high cost and a small quantity, and cannot be used as a surgical training prop on a large scale. With the popularization of virtual reality technology, some medical operation training equipment integrating virtual reality has appeared, but the existing medical operation training equipment integrating virtual reality generally only feeds back the operation of the operator through a force feedback device, the feedback of the operation of the operator is limited, the operation precision of the operator cannot be further improved, and the medical operation level of the operator is limited. SUMMARY

[0003] In view of the defects in the prior art, the present application aims to provide a medical operation training system and method based on virtual reality, which can improve the operation precision of the operator.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical scheme: a medical operation training system based on virtual reality, comprising:

[0005] A training glove, each finger joint inner arm of the training glove is distributed with a stimulating circuit, and each finger joint outer wall of the training glove is provided with a sensing module, each sensing module corresponds to the stimulating circuit at the same finger joint position, and the sensing module is used for detecting the direction data, motion data and gravity data at each finger joint position in real time.

[0006] An intelligent glasses, which is internally provided with:

[0007] At least one angular motion detection unit, which is used for detecting the angular motion data of head motion in real time.

[0008] At least one shooting unit, which is used for shooting the eyeball movement pictures of eye motion in real time.

[0009] The first processing unit is respectively connected with the angle motion detection unit and the shooting unit, and is configured to obtain a first adjustment instruction and a second adjustment instruction according to the angle motion data and the eyeball movement picture respectively.

[0010] The data processing module is connected with the training glove and the smart glasses, and includes:

[0011] The second processing unit is configured to obtain a hand real-time motion model according to the direction data, the motion data and the gravity data.

[0012] The storage unit stores a plurality of surgical virtual scenes, and each of the surgical virtual scenes is configured with a corresponding hand standard motion model.

[0013] The comparison unit is respectively connected with the second processing unit and the storage unit, and is configured to compare the hand real-time motion model with the hand standard motion model to obtain a deviation result of each of the knuckles after the operator selects the surgical virtual scene.

[0014] The adjustment unit is connected with the comparison unit, and is configured to obtain a reminding instruction according to the deviation result, and to adjust a visual angle of the operator in the selected surgical virtual scene in real time according to the first adjustment instruction and the second adjustment instruction.

[0015] The stimulation circuit stimulates the corresponding knuckle position according to the reminding instruction.

[0016] Further, the angle motion detection unit and the shooting unit are both provided with two, the angle motion data and the eyeball movement picture can both be detected at the same time, and the first processing unit includes:

[0017] The first processing subunit is configured to convert two angle motion data into head rotation data according to a preset head posture algorithm, and to input two head rotation data into a preset first calculation formula to obtain comprehensive head rotation data.

[0018] The second processing subunit is configured to extract eyeball movement features in the eyeball movement picture according to a preset feature extraction algorithm, to input the eyeball movement features into a preset picture processing model to obtain picture movement parameters, and to input two picture movement parameters into a preset second calculation formula to obtain comprehensive picture movement parameters.

[0019] The generation subunit is respectively connected with the first processing subunit and the second processing subunit, and is configured to generate the first adjustment instruction according to the comprehensive head rotation data, and to generate the second adjustment instruction according to the comprehensive picture movement parameters.

[0020] Further, the first calculation formula is configured as:

[0021] ;

[0022] wherein, is used to represent the head comprehensive rotation data;

[0023] is used to represent one of the head rotation data;

[0024] is used to represent another of the head rotation data;

[0025] is used to represent the distance between the two angle detection units;

[0026] is used to represent the included angle between the two angle detection units.

[0027] Further, the second calculation formula is configured as:

[0028] ;

[0029] wherein, is used to represent the picture comprehensive movement parameter;

[0030] is used to represent a preset first coefficient, the first coefficient being a constant;

[0031] is used to represent a preset second coefficient, the second coefficient being a constant;

[0032] is used to represent one of the picture movement parameters;

[0033] is used to represent another of the picture movement parameters;

[0034] is used to represent the distance between the two shooting units;

[0035] is used to represent the included angle between the two shooting units.

[0036] Further, the second processing unit comprises:

[0037] an introducing subunit, used to introduce a preset glove three-dimensional initial model;

[0038] The superposition subunit is connected to the introduction subunit and is configured to input each of the direction data, each of the motion data, and each of the gravity data into a preset motion target tracking algorithm to obtain a real-time motion model of the hand corresponding to the three-dimensional initial model of the glove.

[0039] Further, the comparison unit comprises:

[0040] The correction subunit is configured to generate a correction stimulus degree according to a spatial position distance of the real-time motion model of the hand and the standard motion model of the hand at the same knuckle, the correction stimulus degree being linearly positively correlated with the spatial position distance.

[0041] The comparison subunit is connected to the correction subunit and is configured to generate the deviation result according to the correction stimulus degree, the deviation result being linearly positively correlated with the correction stimulus degree.

[0042] Further, the training glove, the data processing module, and the smart glasses are each provided with a wireless connection device, and the training glove and the smart glasses are wirelessly connected to the data processing module.

[0043] Further, the smart glasses are provided with a foldable double-layer lens, and the foldable double-layer lens comprises a foldable double-convex lens and a flat lens.

[0044] Further, the stimulation circuit comprises:

[0045] a pulse output chip;

[0046] a MOS tube, a gate of which is connected to an output pin of the pulse output chip, and a source of which is connected to an input pin of the pulse output chip;

[0047] a transformer, one end of a secondary winding of which is grounded, the other end of the secondary winding of which is connected to an emitter of the MOS tube, and both ends of a primary resistance of which are connected to stimulation metal electrodes;

[0048] a protection diode, which is connected in parallel between the MOS tube and the transformer, a positive electrode of which is connected to one end of the secondary winding of the transformer, and a negative electrode of which is connected to the source of the MOS tube.

[0049] A medical operation training method based on virtual reality, comprising:

[0050] In step S1, a sensing module detects direction data, motion data, and gravity data at positions of knuckles of a training glove in real time;

[0051] In step S2, an angular motion detection unit detects angular motion data of head motion in real time, and a shooting unit shoots eye movement pictures of eye motion in real time.

[0052] Step S3, the first processing unit respectively processes the angular motion data and the eyeball movement picture to obtain a first adjustment instruction and a second adjustment instruction;

[0053] Step S4, the second processing unit processes each of the direction data, each of the motion data and each of the gravity data to obtain a hand real-time motion model;

[0054] Step S5, the comparison unit compares the hand real-time motion model with a hand standard motion model after the operator selects the surgical virtual scene to obtain a deviation result at each of the knuckles;

[0055] Step S6, the adjustment unit processes a reminding instruction according to the deviation result and adjusts the visual angle of the operator in the selected surgical virtual scene in real time according to the first adjustment instruction and the second adjustment instruction;

[0056] Step S7, the stimulation circuit stimulates the corresponding knuckle position according to the reminding instruction.

[0057] The beneficial effects of the present application are as follows:

[0058] The present application realizes real-time collection of direction data, motion data and gravity data at the knuckle positions of the training gloves and sends them to the data processing module, so that the data processing module can generate a hand real-time motion model according to each of the direction data, each of the motion data and each of the gravity data, compare it with a hand standard motion model to obtain a deviation result, and then generate a reminding instruction according to the deviation result to stimulate the knuckles of the operator by the stimulation circuit at the knuckle positions of the training gloves, so that the operator can continuously correct the non-standard actions during the surgical training process using the training gloves and finally form a muscle memory that can rely on standard actions for surgery, thereby effectively improving the medical operation precision of the operator. Meanwhile, the present application also detects the eyeball movement picture by the built-in shooting detection unit of the smart glasses, detects the angular motion data of the head of the operator by the angular motion detection unit, and then processes the first adjustment instruction and the second adjustment instruction by the first processing unit, so that the data processing module adjusts the visual angle of the operator in the selected surgical virtual scene according to the first adjustment instruction and the second adjustment instruction, realizes the adjustment of the visual angle according to the rotation of the head and the eyeball of the operator, and improves the authenticity and immersion of the medical surgical training of the operator. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 is a structural schematic diagram of the medical surgical training system in the present application;

[0060] Figure 2 is a step flowchart of the medical surgical training method in the present application.

[0061] Fig. 1, stimulation circuit; 2, sensing module; 3, angular motion detection unit; 4, shooting unit; 5, first processing unit; 51, first processing subunit; 52, second processing subunit; 53, generation subunit; 6, data processing module; 61, second processing unit; 611, introduction subunit; 612, superposition subunit; 62, storage unit; 63, comparison unit; 631, correction subunit; 632, comparison subunit; 64, adjustment unit. DETAILED DESCRIPTION

[0062] The application will be further described in detail below in combination with the drawings and examples. Identical parts are denoted by identical reference numerals. It should be noted that the words "front", "back", "left", "right", "up" and "down" used in the following description refer to the directions in the drawings, and the words "bottom surface" and "top surface", "inner" and "outer" refer to the directions towards or away from the geometric center of a particular part.

[0063] As shown in the drawings, the medical operation training system based on virtual reality of the embodiment comprises: Figure 1

[0064] A training glove, each finger joint of the training glove is provided with a stimulation circuit 1, and each outer wall of the finger joint of the training glove is provided with a sensing module 2, each sensing module 2 corresponds to the stimulation circuit 1 at the same finger joint position, and the sensing module 2 is used for detecting the direction data, motion data and gravity data at each finger joint position in real time;

[0065] An intelligent glasses, which is internally provided with:

[0066] At least one angular motion detection unit 3, which is used for detecting the angular motion data of head motion in real time;

[0067] At least one shooting unit 4, which is used for shooting the eyeball movement pictures of eye motion in real time;

[0068] A first processing unit 5, which is connected with the angular motion detection unit 3 and the shooting unit 4 respectively, and is used for processing the first adjustment instruction and the second adjustment instruction according to the angular motion data and the eyeball movement pictures respectively;

[0069] A data processing module 6, which is connected with the training glove and the intelligent glasses, and comprises:

[0070] A second processing unit 61, which is used for processing a hand real-time motion model according to each direction data, each motion data and each gravity data;

[0071] A storage unit 62, which stores a plurality of surgical virtual scenes, and each surgical virtual scene is configured with a corresponding hand standard motion model;

[0072] ​The comparison unit 63 is connected to the second processing unit 61 and the storage unit 62 respectively, and is configured to compare the real-time hand movement model with the standard hand movement model after the operator selects the virtual surgery scene, and obtain the deviation results of each finger joint.

[0073] The adjustment unit 64 is connected to the comparison unit 63, and is configured to obtain a reminding instruction according to the deviation results, and adjust the visual angle of the operator in the selected virtual surgery scene in real time according to the first adjustment instruction and the second adjustment instruction.

[0074] The stimulation circuit 1 stimulates the corresponding finger joint position according to the reminding instruction.

[0075] Specifically, in the embodiment, the sensing module 2 includes a direction sensor, a motion sensor, and a gravity sensor. The direction sensor detects the direction data of each finger joint position on the training glove in real time. The motion sensor detects the motion data of each finger joint position on the training glove in real time. The gravity sensor detects the gravity data of each finger joint position on the training glove in real time. The data processing module 6 can be a computer. Before the operator performs the medical surgery training, the operator needs to wear the training glove. The inner wall of each finger joint of the training glove is provided with the stimulation circuit 1. The training glove corresponds to 10 fingers, each finger has three finger joints, and the corresponding sensing module 2 and stimulation circuit 1 are provided with 30, respectively. The device address of each sensing module 2 and stimulation circuit 1 corresponds to the corresponding finger joint. The operator selects the virtual surgery scene by using the data processing module 6, and then displays the virtual surgery scene on the lenticular lens of the smart glasses. Before the medical surgery training starts, the operator first needs to move each finger joint to position. After positioning, the medical surgery training starts in the virtual surgery scene. Each virtual surgery scene is associated with a corresponding standard hand movement model. When the medical surgery training starts, the standard hand movement model starts to move. The operator only needs to operate according to the text prompt and the movement of the standard hand movement model to complete the surgery simulation training.

[0076] In the surgical simulation training process, the data processing module 6 processes the direction data, motion data and gravity data uploaded by the sensing module 2 in real time to obtain a hand real-time motion model, compares the hand real-time motion model with the position of each finger joint of the hand standard motion model to obtain a deviation result at the same finger joint position, and then obtains a corresponding stimulation instruction according to the deviation result, wherein the deviation result is linearly positively correlated with the spatial position distance of the hand real-time motion model and the hand standard motion model at the same finger joint position, the smaller the spatial position distance, the smaller the deviation result, and the stimulation current value in the generated stimulation instruction, the stimulation of the operator's finger joint by the stimulation circuit 1 is weaker, and vice versa. By such a setting, the operator can continuously correct non-standard actions during the surgical training process using the training gloves to ultimately form muscle memory and perform surgery by standard actions, thereby effectively improving the operator's medical operation precision.

[0077] In the surgical simulation training process, the shooting detection unit built in the smart glasses detects the eyeball movement pictures in real time, the angular motion detection unit 3 detects the angular motion data of the operator's head in real time, and then the first processing unit 5 processes the first adjustment instruction and the second adjustment instruction, and the data processing module 6 adjusts the view angle of the operator in the selected virtual surgical scene according to the first adjustment instruction and the second adjustment instruction, which realizes the adjustment of the view angle according to the rotation of the operator's head and eyeballs, and improves the authenticity and immersion of the operator's medical surgical training.

[0078] Preferably, the smart glasses are configured with a folding double-layer lens, which includes a foldable double-convex lens and a flat lens.

[0079] Specifically, in the present embodiment, by configuring the folding double-layer lens, the double-convex lens can be used when the virtual scene image needs to be viewed, and the double-convex lens can be folded and stored when the virtual scene image does not need to be viewed, and only the flat lens is used, which expands the use scene of the smart glasses and improves the practicality and convenience of the smart glasses.

[0080] In this embodiment, the smart glasses combine AR, VR and MR technologies. After the operator selects a virtual surgery scene in the data processing module 6, the operator can display and perform virtual surgery operation on the lenticular lens of the smart glasses, which embodies the application of VR technology. After the operator folds up the lenticular lens, the real world scene is displayed on the flat lens. The smart glasses are provided with a camera, which can capture the picture of the medical equipment in the real scene in real time and send it to the data processing module 6. The data processing module 6 can obtain the specific model name of the corresponding equipment according to the comparison and analysis of the equipment picture and the cloud database, and display and mark the corresponding equipment on the flat lens, which embodies the application of AR technology. When the operator uses the flat lens, the operator can also select the mixed reality operation function, that is, the training gloves in the real world scene and the virtual patient in the virtual surgery scene are displayed on the flat lens at the same time. The judgment of the precision of the medical operation is consistent with that in the virtual surgery scene, which embodies the application of MR technology.

[0081] Preferably, the angle motion detection unit 3 and the shooting unit 4 are each provided with two, so that two angle motion data and eye movement pictures can be detected at the same time. The first processing unit 5 includes:

[0082] The first processing subunit 51 is configured to convert the two angle motion data into head rotation data respectively according to a preset head posture algorithm, and input the two head rotation data into a preset first calculation formula to obtain comprehensive head rotation data.

[0083] The second processing subunit 52 is configured to extract eye movement features in the eye movement pictures according to a preset feature extraction algorithm, input the eye movement features into a preset picture processing model to obtain picture movement parameters, and input the two picture movement parameters into a preset second calculation formula to obtain comprehensive picture movement parameters.

[0084] The generating subunit 53 is connected to the first processing subunit 51 and the second processing subunit 52 respectively, and is configured to generate a first adjustment instruction according to the comprehensive head rotation data, and generate a second adjustment instruction according to the comprehensive picture movement parameters.

[0085] Specifically, in this embodiment, the angle detection unit can be a gyroscope, and the shooting unit 4 can be a camera module. By providing two angle detection units and two shooting units 4, two sets of head rotation data and eye movement pictures can be measured at the same time, and then two sets of head rotation data and picture movement parameters can be obtained. Finally, the comprehensive head rotation data can be calculated according to the two sets of head rotation data, and the comprehensive picture movement parameters can be calculated according to the two sets of picture movement parameters, which effectively improves the precision of the obtained comprehensive head rotation data and comprehensive picture movement parameters.

[0086] Preferably, the first calculation formula is configured as:

[0087] ;

[0088] wherein, for representing the head comprehensive rotation data;

[0089] for representing one of the head rotation data;

[0090] for representing the other head rotation data;

[0091] for representing the distance between the two angle detection units;

[0092] for representing the included angle between the two angle detection units.

[0093] Further, the second calculation formula is configured as:

[0094] ;

[0095] wherein, for representing the picture comprehensive movement parameter;

[0096] for representing the preset first coefficient, the first coefficient being a constant;

[0097] for representing the preset second coefficient, the second coefficient being a constant;

[0098] for representing one of the picture movement parameters;

[0099] for representing the other picture movement parameter;

[0100] for representing the distance between the two shooting units;

[0101] for representing the included angle between the two shooting units.

[0102] Preferably, the second processing unit 61 comprises:

[0103] an introducing sub-unit 611 for introducing a preset glove three-dimensional initial model;

[0104] a superimposing sub-unit 612 connected to the introducing sub-unit 611, for inputting the direction data, the motion data and the gravity data into a preset motion target tracking algorithm to obtain a hand real-time motion model corresponding to the glove three-dimensional initial model.

[0105] Specifically, in this embodiment, the glove three-dimensional initial model is a three-dimensional model pre-constructed according to the three-dimensional size of the training glove. The superposition subunit 612 inputs the real-time detected directional data, motion data, and gravity data into the camshift algorithm, and the camshift algorithm adjusts the glove three-dimensional initial model in real time according to the directional data, motion data, and gravity data to obtain a hand real-time motion model.

[0106] Preferably, the comparison unit 63 comprises:

[0107] The correction subunit 631 is configured to generate a correction stimulation degree according to the spatial position distance of the hand real-time motion model and the hand standard motion model at the same knuckle, and the correction stimulation degree is linearly positively correlated with the spatial position distance;

[0108] The comparison subunit 632 is connected to the correction subunit 631 and is configured to generate a deviation result according to the correction stimulation degree, and the deviation result is linearly positively correlated with the correction stimulation degree.

[0109] Specifically, in this embodiment, the greater the spatial position distance, the greater the correction stimulation degree, and the greater the deviation result. Conversely, the smaller the spatial position distance, the smaller the correction stimulation degree, and the smaller the deviation result. By such a setting, the spatial position distance of the hand real-time motion model and the hand standard motion model at the same knuckle is represented by the deviation result, which is clear and convenient for subsequent processing.

[0110] Preferably, the training glove, the data processing module 6, and the smart glasses are each provided with a wireless connection device, and the training glove and the smart glasses are wirelessly connected to the data processing module 6.

[0111] Specifically, in this embodiment, the wireless connection device can be a Bluetooth device. By providing the wireless connection device, wireless connection between the training glove, the data processing module 6, and the smart glasses is achieved, which eliminates the complexity of wiring and improves installation convenience.

[0112] Preferably, a power supply battery is placed in the temple of the smart glasses for supplying power to the smart glasses. The back of the palm of the training glove is provided with a battery pack for supplying power to the training glove.

[0113] Preferably, the stimulation circuit 1 comprises:

[0114] a pulse output chip;

[0115] a MOS tube, the gate electrode of which is connected to the output pin of the pulse output chip, and the source electrode of which is connected to the input pin of the pulse output chip;

[0116] a transformer, one end of the secondary winding of which is grounded, the other end of the secondary winding of which is connected to the emitter of the MOS tube, and the primary resistance of which is connected to the stimulation metal electrode.

[0117] A protection diode is connected in parallel between the MOS tube and the transformer, with the positive terminal connected to one end of the secondary winding of the transformer and the negative terminal connected to the source of the MOS tube.

[0118] Specifically, in the embodiment, the pulse output chip generates a pulse signal, the MOS tube controls the on-off of the pulse signal, and the transformer outputs the pulse signal at a reduced voltage. Finally, by stimulating the metal electrode in contact with the human finger joint, the maximum current, maximum voltage and maximum power of the signal output by the transformer are within the safety indicators of the human body, avoiding harm to human health, and the protection diode is used to avoid reverse current loss of circuit components.

[0119] A medical operation training method based on virtual reality, as shown in Figure 2 , comprising:

[0120] Step S1, the sensing module 2 detects the direction data, motion data and gravity data at each finger joint position of the training glove in real time;

[0121] Step S2, the angular motion detection unit 3 detects the angular motion data of the head movement in real time, and the shooting unit 4 shoots the eye movement picture of the eye movement in real time;

[0122] Step S3, the first processing unit 5 processes the angular motion data and the eye movement picture to obtain a first adjustment instruction and a second adjustment instruction, respectively;

[0123] Step S4, the second processing unit processes each direction data, each motion data and each gravity data to obtain a hand real-time motion model;

[0124] Step S5, the comparison unit 63 compares the hand real-time motion model with the hand standard motion model after the operator selects the operation virtual scene to obtain the deviation result at each finger joint;

[0125] Step S6, the adjustment unit 64 processes a prompt instruction according to the deviation result, and adjusts the visual angle of the operator in the selected operation virtual scene in real time according to the first adjustment instruction and the second adjustment instruction;

[0126] Step S7, the stimulation circuit 1 stimulates the corresponding finger joint position according to the prompt instruction.

[0127] The above is only the preferred embodiment of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solution falling within the scope of the present application is within the protection scope of the present application. It should be noted that, for ordinary skilled persons in the art, some improvements and refinements without departing from the principles of the present application are also considered as the protection scope of the present application.

Claims

1. A medical surgical training system based on virtual reality, characterized in that, include: The training glove has stimulation circuits (1) distributed on the inner arm of each finger joint, and a sensing module (2) is provided on the outer wall of each finger joint. Each sensing module (2) corresponds to the stimulation circuit (1) at the same finger joint position. The sensing module (2) is used to detect the direction data, motion data and gravity data at each finger joint position in real time. Smart glasses, with the following features inside: At least one angular motion detection unit (3) is used to detect the angular motion data of head movement in real time; At least one camera unit (4) is used to capture images of eye movements in real time; The first processing unit (5) is connected to the angular motion detection unit (3) and the shooting unit (4) respectively, and is used to process the angular motion data and the eye movement image to obtain the first adjustment instruction and the second adjustment instruction respectively; Data processing module (6), connecting the training glove and the smart glasses, includes: The second processing unit (61) is used to process the direction data, motion data and gravity data to obtain a real-time hand motion model. The storage unit (62) stores several surgical virtual scenes, and each of the surgical virtual scenes is configured with a corresponding standard hand movement model; The comparison unit (63) is connected to the second processing unit (61) and the storage unit (62) respectively, and is used to compare the real-time hand motion model with the standard hand motion model after the operator selects the surgical virtual scene, so as to obtain the deviation results at each of the finger joints; The adjustment unit (64) is connected to the comparison unit (63) and is used to process the deviation result to obtain a reminder instruction, and to adjust the operator's viewing angle in the selected surgical virtual scene in real time according to the first adjustment instruction and the second adjustment instruction; The stimulation circuit (1) provides stimulation reminders to the corresponding finger joint positions according to the reminder instructions; The first calculation formula is configured as follows: ； in, Used to represent the comprehensive head rotation data; Used to represent one of the head rotation data; Used to represent another head rotation data; Used to indicate the distance between the two angular motion detection units (3); Used to represent the included angle between the two angular motion detection units (3); The second calculation formula is configured as follows: ； in, Used to represent overall screen movement parameters; Used to represent a preset first coefficient, wherein the first coefficient is a constant; This is used to represent a preset second coefficient, which is a constant; Used to represent one of the screen movement parameters; Used to represent another screen movement parameter; Used to indicate the distance between the two shooting units (4); Used to indicate the included angle between the two shooting units (4); Both the angular motion detection unit (3) and the imaging unit (4) are provided in twos, so that both the angular motion data and the eye movement image can be detected at the same time. The first processing unit (5) includes: The first processing subunit (51) is used to convert the two angular motion data into head rotation data according to the preset head posture algorithm, and input the two head rotation data into the preset first calculation formula to obtain the comprehensive head rotation data. The second processing subunit (52) is used to extract eye movement features in the eye movement image according to a preset feature extraction algorithm, input the eye movement features into a preset image processing model to obtain image movement parameters, and put the two image movement parameters into a preset second calculation formula to obtain comprehensive image movement parameters. A generation subunit (53) is connected to the first processing subunit (51) and the second processing subunit (52) respectively, and is used to generate the first adjustment instruction based on the comprehensive head rotation data and generate the second adjustment instruction based on the comprehensive screen movement parameters.

2. The virtual reality-based medical surgical training system according to claim 1, characterized in that: The second processing unit (61) includes: Sub-unit (611) is introduced to introduce a preset three-dimensional initial model of the glove; The superposition subunit (612) is connected to the introduction subunit (611) and is used to input the direction data, motion data and gravity data of each of the above into a preset motion target tracking algorithm to obtain the real-time motion model of the hand corresponding to the three-dimensional initial model of the glove.

3. The virtual reality-based medical surgical training system according to claim 1, characterized in that: The comparison unit (63) includes: The correction subunit (631) is used to generate a correction stimulus based on the spatial distance between the real-time hand motion model and the standard hand motion model at the same finger joint, wherein the correction stimulus is linearly positively correlated with the spatial distance. The comparison subunit (632) is connected to the correction subunit (631) and is used to generate the deviation result based on the correction stimulus intensity, the deviation result being linearly positively correlated with the correction stimulus intensity.

4. The virtual reality-based medical surgical training system according to claim 1, characterized in that: The training glove, the data processing module (6), and the smart glasses are all equipped with wireless connection devices, and the training glove and the smart glasses are wirelessly connected to the data processing module (6).

5. The virtual reality-based medical surgical training system according to claim 1, characterized in that: The smart glasses are equipped with foldable double-layer lenses, which include a foldable biconvex lens and a flat lens.

6. The virtual reality-based medical surgical training system according to claim 1, characterized in that: The stimulation circuit (1) includes: Pulse output chip; The MOSFET has its gate connected to the output pin of the pulse output chip and its source connected to the input pin of the pulse output chip. The transformer has one end of its secondary winding grounded and the other end connected to the emitter of the MOS transistor. Both ends of the primary resistor are connected to a stimulating metal electrode. A protection diode is connected in parallel between the MOSFET and the transformer, with its positive terminal connected to one end of the secondary winding of the transformer and its negative terminal connected to the source of the MOSFET.

7. A virtual reality-based medical surgical training method, applied to the virtual reality-based medical surgical training system according to any one of claims 1-6, characterized in that, include: Step S1, the sensing module (2) detects the direction data, motion data and gravity data of each finger joint of the training glove in real time; Step S2, the angular motion detection unit (3) detects the angular motion data of the head movement in real time, and the shooting unit (4) captures eye movement images of the eye movement in real time; Step S3, the first processing unit (5) processes the angular motion data and the eye movement image to obtain the first adjustment instruction and the second adjustment instruction respectively; Step S4: The second processing unit processes the direction data, motion data, and gravity data to obtain a real-time hand motion model. Step S5, the comparison unit (63) compares the real-time hand motion model with the standard hand motion model after the operator selects the surgical virtual scene, and obtains the deviation results at each of the finger joints; Step S6, the adjustment unit (64) processes the deviation result to obtain a reminder instruction, and adjusts the operator's viewing angle in the selected surgical virtual scene in real time according to the first adjustment instruction and the second adjustment instruction; Step S7, the stimulation circuit (1) stimulates the corresponding knuckle position according to the reminder instruction.

Images