Intelligent auxiliary method and system for nursing midwifery teaching training

Abstract

The invention relates to an intelligent auxiliary method and system for nursing midwifery teaching training. The method comprises the steps of obtaining hand posture original data of each calibration point of a student in a simulated birth canal virtual scene and a forefinger fingertip space coordinate true value, performing optimization calculation processing, and generating a personalized hand biomechanical parameter set of the student; calculating the current position of the real fingertip of the student and the virtual force borne by the real fingertip based on the personalized hand biomechanical parameter set in combination with the hand posture data of the student in the midwifery operation training; mapping the virtual force to generate a force feedback equipment control instruction; after the control instruction of the force feedback equipment is executed, obtaining operation data of the student; and based on the operation data, an operation evaluation result of nursing midwifery teaching training is generated. The method can adapt to the individual difference of the hands of the students, solves the problem that the force sense and the visual focus of a traditional system are not matched, and helps the students to form correct muscle memory.

Description

Technical Field

[0001] This invention belongs to the field of nursing and midwifery teaching and training, and in particular relates to intelligent auxiliary methods and systems. Background Technology

[0002] With the development of virtual reality and force feedback technologies, highly realistic simulation training systems have been widely used in nursing and midwifery education and training. These systems construct virtual delivery room environments and physiological models, allowing trainees to repeatedly practice key procedures such as cervical palpation, fetal position assessment, and delivery assistance in a risk-free setting, effectively compensating for the limited opportunities of traditional clinical clerkships. Currently, these systems generally utilize force feedback devices to simulate the feel of the procedures, aiming to enhance the immersion and practicality of the training.

[0003] In traditional techniques, such systems typically employ standardized force feedback devices, such as force feedback gloves or styluses. During training, the system calculates the corresponding reaction forces based on collisions and interactions occurring in a virtual environment, and then drives the device to generate physical force feedback. Trainees learn and master the essentials of various midwifery procedures by observing the virtual scene on the screen and combining this with the force sensations felt in their hands. This method relies heavily on the universal force feedback provided by the device itself.

[0004] However, current training methods have a significant drawback: due to the natural individual differences in the size of trainees' hands, while the physical structure and force application point of force feedback devices are fixed and uniform, there is often a spatial deviation between the actual point of application of the device's output force and the contact point of the trainee's virtual fingertips in the scene. For trainees with smaller hands, the delicate force that should be felt at the fingertips, simulating the palpation of the cervix, may actually be applied to the knuckles or palm. This severe mismatch between the "force focus" and the "visual focus" creates confusing sensory conflict, making it difficult to form correct muscle memory and potentially misleading trainees into developing incorrect operating habits, thus limiting the accuracy and reliability of the teaching effect. Summary of the Invention

[0005] Therefore, it is necessary to provide an intelligent auxiliary method for nursing and midwifery teaching and training that can adapt to the individual hand characteristics of trainees and ensure the precise unity of force perception and visual spatial position, in order to address the above-mentioned technical problems.

[0006] Firstly, this application provides an intelligent auxiliary method for nursing and midwifery teaching and training, including:

[0007] Acquire calibration data of trainees in a simulated birth canal virtual scenario; and summarize all calibration data to generate a calibration dataset; the calibration data includes raw hand posture data and true values ​​of index fingertip spatial coordinates;

[0008] Based on the calibrated dataset, personalized sets of hand biomechanical parameters for trainees are generated through optimized computational processing.

[0009] Acquire trainees' current hand posture data during midwifery operation training; and based on the current hand posture data and personalized hand biomechanical parameter set, calculate the current position of the trainees' real fingertips and the virtual force on the real fingertips;

[0010] Based on a personalized set of hand biomechanical parameters and current hand posture data, virtual forces are mapped and processed to generate force feedback device control commands; these commands instruct the force feedback device to output a realistic force sensation corresponding to the virtual scene.

[0011] After executing the control commands from the force feedback device, the system acquires the trainee's operational data and generates operational assessment results for nursing and midwifery teaching and training based on the operational data.

[0012] Furthermore, based on the calibration dataset, and through optimized computational processing, a personalized set of hand biomechanical parameters for each student is generated, including:

[0013] The raw hand posture data in the calibration dataset is calibrated to obtain standard hand posture data;

[0014] Based on the standard hand posture data and the true value of the index fingertip spatial coordinates corresponding to all calibration points, an optimization objective function is constructed; the objective function uses the set of hand biomechanical parameters as variables;

[0015] The minimum value of the objective function is calculated using a nonlinear least squares algorithm; and the set of hand biomechanical parameters corresponding to the minimum value is determined as the student's personalized hand biomechanical parameter set.

[0016] Furthermore, based on the standard hand posture data corresponding to all calibration points and the true value of the index fingertip spatial coordinates, an optimization objective function is constructed, including:

[0017] Based on the standard hand posture data and the true spatial coordinates of the index fingertip corresponding to all calibration points, an optimization objective function is constructed:

[0018]

[0019] in, To optimize the objective function, Φ is the iterative variable of the parameter set. For positive kinematic functions, This represents the standard hand posture data corresponding to the i-th calibration point. Let N be the true spatial coordinates of the index fingertip corresponding to the i-th calibration point, and N be the total number of calibration points. Let be the square of the Euclidean norm. This is a minimum value operation.

[0020] Furthermore, based on the current hand posture data and a personalized set of hand biomechanical parameters, the current position of the student's actual fingertips and the virtual force acting on those fingertips are calculated, including:

[0021] The current hand posture data is calibrated to obtain the current standard hand posture data;

[0022] Based on a personalized set of hand biomechanical parameters and current standard hand posture data, the current position of the student's actual fingertips is obtained through forward kinematics calculations.

[0023] Virtual contact points are determined based on collision detection; and virtual forces acting on virtual contact points are calculated based on the biomechanical model of virtual organs; the virtual forces are the virtual forces experienced by the real fingertips.

[0024] Furthermore, based on a personalized set of hand biomechanical parameters and current hand posture data, virtual forces are mapped to generate force feedback device control commands, including:

[0025] Based on a personalized set of hand biomechanical parameters and current hand posture data, the Jacobian matrix from the fixed reference point of the force feedback device to the current position of the actual fingertip is calculated.

[0026] Based on the Jacobian matrix and virtual force, the equivalent force and equivalent torque that the force feedback device needs to apply at a fixed reference point are obtained through mapping processing using the principle of virtual work.

[0027] The equivalent force and equivalent torque are combined to generate control commands for the force feedback device.

[0028] Furthermore, after executing the control commands from the force feedback device, the trainee's operational data is acquired; and based on the operational data, operational assessment results for nursing and midwifery teaching and training are generated, including:

[0029] After executing the control commands of the force feedback device, the training process acquires the trainee's operational data; the operational data includes force time-series data of virtual force changing over time and trajectory time-series data of hand movement trajectory changing over time;

[0030] The operational data is compared with the preset expert standard operational data to obtain the comparison results;

[0031] Based on the comparison results, operational assessment results for nursing and midwifery teaching and training are generated.

[0032] Furthermore, calibration data of trainees in a simulated birth canal virtual scenario is obtained; and all calibration data is aggregated to generate a calibration dataset, including:

[0033] In a simulated birth canal virtual scene, calibration points are set according to a preset clinical palpation path;

[0034] Obtain raw data of the student's hand posture and true spatial coordinates of the index fingertip when the student touches each calibration point;

[0035] The calibration dataset is obtained by summarizing all the raw hand posture data and the true values ​​of the spatial coordinates of the index fingertip.

[0036] Secondly, this application also provides an intelligent auxiliary system for nursing and midwifery teaching and training, including:

[0037] The data acquisition module is used to acquire calibration data of trainees in a simulated birth canal virtual scene; and to summarize all calibration data to generate a calibration dataset; the calibration data includes raw hand posture data and true values ​​of the spatial coordinates of the index fingertip;

[0038] The parameter set generation module is used to generate a personalized set of hand biomechanical parameters for trainees based on the calibration dataset through optimized computational processing.

[0039] The virtual force calculation module is used to acquire the trainee's current hand posture data during midwifery operation training; and based on the current hand posture data and personalized hand biomechanical parameter set, to calculate the current position of the trainee's real fingertips and the virtual force acting on the real fingertips.

[0040] The instruction generation module is used to map virtual forces based on a personalized set of hand biomechanical parameters and current hand posture data, and generate force feedback device control instructions; the force feedback device control instructions are used to instruct the force feedback device to output a realistic force sensation corresponding to the virtual scene;

[0041] The assessment result generation module is used to acquire the trainee's operation data after the control command of the force feedback device; and to generate operation assessment results for nursing and midwifery teaching and training based on the operation data.

[0042] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by the processor to implement any of the intelligent auxiliary methods for nursing and midwifery teaching and training described in the embodiments of this application.

[0043] Fourthly, this application also provides a computer-readable storage medium storing at least one piece of program code, which is loaded and executed by a processor to implement the intelligent auxiliary method for nursing and midwifery teaching and training as described in any of the embodiments of this application.

[0044] The aforementioned intelligent auxiliary method and system for nursing and midwifery teaching and training acquires raw hand posture data and true spatial coordinates of the index fingertip at each calibration point in a simulated birth canal virtual scenario. This data is then optimized and processed to generate a personalized set of hand biomechanical parameters for each trainee. Based on this personalized set of parameters, combined with the trainee's current hand posture data during midwifery operation training, the system calculates the trainee's actual fingertip position and the virtual force acting on it. The virtual force is mapped to generate control commands for a force feedback device, instructing the device to output a realistic force sensation corresponding to the virtual scenario. After executing these commands, the system acquires the trainee's operational data and generates an operational evaluation result for the nursing and midwifery teaching and training based on this data. This system adapts to individual differences in trainees' hands, solves the problem of mismatch between force perception and visual focus in traditional systems, helps trainees develop correct muscle memory, effectively avoids the formation of incorrect operating habits, and improves the accuracy, reliability, and practicality of nursing and midwifery teaching and training. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a flowchart illustrating an intelligent auxiliary method for nursing and midwifery teaching and training in one embodiment.

[0047] Figure 2 This is a flowchart illustrating the steps involved in generating a personalized set of hand biomechanical parameters for trainees based on a calibration dataset through optimized computational processing, as shown in one embodiment.

[0048] Figure 3 This is a schematic diagram of the structure of an intelligent auxiliary system for nursing and midwifery teaching and training in one embodiment. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0050] In one embodiment, an intelligent auxiliary method for nursing and midwifery teaching and training is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and can also be applied to a system including a terminal and a server, and implemented through the interaction between the terminal and the server. Figure 1As shown, in this embodiment, the method includes the following steps:

[0051] Step S101: Obtain the calibration data of the trainee in the simulated birth canal virtual scene; and summarize all the calibration data to generate a calibration dataset; the calibration data includes the original hand posture data and the true value of the spatial coordinates of the index fingertip.

[0052] For example, a virtual birth canal simulation scenario closely resembling a clinical midwifery setting is constructed. This scenario includes a series of calibration points covering key operational areas such as cervical palpation and fetal position assessment. Raw hand posture data of trainees during palpation at each calibration point is collected. This data includes the angles, angular velocities, and spatial positions of each hand joint. Simultaneously, the true spatial coordinates of the index fingertip are obtained, representing the precise position of the fingertip within the virtual scenario's coordinate system. All collected raw hand posture data and the true spatial coordinates of the index fingertip are categorized and organized to generate a calibration dataset.

[0053] Step S102: Based on the calibration dataset, a personalized set of hand biomechanical parameters for the trainee is generated through optimized computational processing.

[0054] Among them, optimization computation refers to the process of using specific algorithms or methods to perform calculations on the objective function, that is, the mathematical expression of the quantification requirement, adjust relevant parameters, and gradually approach and find the parameter combination that makes the objective function reach the optimal value.

[0055] For example, based on the hand operation posture data and the true values ​​of the index fingertip spatial coordinates corresponding to all calibration points in the calibration dataset, an optimization objective function is constructed with the hand biomechanical parameter set as the iterative variable. This function is then minimized to obtain a personalized hand biomechanical parameter set adapted to the trainee's hand characteristics. This hand biomechanical parameter set includes parameters reflecting the physiological characteristics of the hand, such as joint length, joint stiffness, and muscle attachment point location.

[0056] Step S103: Obtain the trainee's current hand posture data during midwifery operation training; and based on the current hand posture data and personalized hand biomechanical parameter set, calculate the current position of the trainee's real fingertips and the virtual force on the real fingertips.

[0057] For example, during midwifery training, the trainee's hand posture data is collected. A personalized set of hand biomechanical parameters and the current hand posture data are then input into a forward kinematic model to obtain the current position of the trainee's actual fingertips in the virtual scene, ensuring that the calculated fingertip position matches the trainee's actual hand characteristics. By judging the contact state between the actual fingertips and virtual organs such as the virtual birth canal and fetus in real time, virtual contact points are determined. Based on a preset virtual organ biomechanical model, combined with contact information such as contact depth and contact area, the virtual force acting on the virtual contact point is calculated. This virtual force is the mechanical feedback experienced by the actual fingertips during training, simulating the real tactile sensation in clinical operations. Among them, the forward kinematics model is a core foundational model in fields such as robotics and computer animation. It can derive the pose of the end effector from joint motion parameters. That is, given the motion states such as the angles and displacements of each joint, the final position and posture of the end working parts such as robot grippers and animated character hands in space can be calculated through geometric and kinematic relationships. The joint kinematic equation is an expression that describes the mathematical relationship between the kinematic parameters such as the position, velocity, and acceleration of each component (such as bones) and variables such as time and joint angles during the movement of the joint. The preset virtual organ biomechanical model is a digital tool that integrates medical anatomy, biomechanics and computer modeling technology. It constructs a virtual copy of a human organ with pre-set basic parameters and structure in a computer environment and focuses on simulating the mechanical behavior of the organ in physiological or pathological states.

[0058] Step S104: Based on the personalized hand biomechanical parameter set and the current hand posture data, the virtual force is mapped and processed to generate force feedback device control instructions; the force feedback device control instructions are used to instruct the force feedback device to output a realistic force sensation corresponding to the virtual scene.

[0059] Among them, mapping processing is a data transformation logic that takes three core types of information as input: virtual force, personalized hand biomechanical parameters, and current hand posture. Through algorithms, it establishes the correspondence between the mechanical information of the virtual scene and the physical output parameters of the device. For example, it combines the "5N virtual resistance" with the characteristics of the student's joints and transforms it into the "3N·m torque" that the device motor needs to output.

[0060] For example, based on a personalized set of hand biomechanical parameters and current standard hand posture data, the virtual force experienced by the real fingertips is mapped to obtain the equivalent force and equivalent torque required to be applied to the fixed reference point of the force feedback device. According to the communication protocol and control specifications of the force feedback device, the equivalent force and equivalent torque are combined into a standardized control command. This command contains key parameters such as the magnitude, direction, and duration of the force, which are used to accurately instruct the force feedback device to output a realistic force sensation corresponding to the mechanical characteristics of the virtual scene.

[0061] Step S105: After executing the control command of the force feedback device, acquire the trainee's operation data; and based on the operation data, generate the operation assessment results of nursing and midwifery teaching and training.

[0062] For example, after executing the control command of the force feedback device, two types of operational data are simultaneously collected during the training process: one is force time-series data showing the change of virtual force over time, recording the dynamic change of virtual force throughout the training; the other is trajectory time-series data showing the change of hand movement trajectory over time, including the hand's movement path, velocity, and acceleration information. By comparing these two types of operational data with preset expert standard operational data, comparison results are obtained. These results are then analyzed to generate operational evaluation results for nursing and midwifery teaching and training. The preset expert standard operational data is a benchmark dataset formed by collecting force time-series data and trajectory time-series data from experienced midwifery experts performing standard operations in the same virtual scenario, and then standardizing them.

[0063] In this embodiment, calibration data is collected from a simulated birth canal virtual scene; the calibration data is optimized and calculated to generate a personalized set of hand biomechanical parameters for the trainee; based on the trainee's hand posture data during midwifery operation training and the personalized set of hand biomechanical parameters, the current position of the trainee's actual fingertips and the virtual force acting on the actual fingertips are calculated; the virtual force is mapped to generate force feedback device control commands; and an evaluation report is generated based on the trainee's operation data under the action of real force and preset expert standard operation data. This approach can adapt to individual differences in trainees' hands, effectively solve the problem of mismatch between force perception and visual focus in traditional systems, avoid the formation of incorrect operating habits, and improve the accuracy, reliability, and practicality of nursing and midwifery teaching and training.

[0064] In one embodiment, such as Figure 2 As shown, based on the calibration dataset, a personalized set of hand biomechanical parameters for trainees is generated through optimized computational processing, including:

[0065] Step S201: The original hand posture data in the calibration dataset is calibrated to obtain standard hand posture data.

[0066] For example, the characteristics of the raw hand posture data in the calibration dataset are analyzed. This raw data contains the angles, angular velocities, and spatial position information of each joint of the hand. However, due to factors such as the accuracy of the sensing device and environmental interference, there may be random noise and errors. A digital filtering algorithm can be used to suppress noise in the raw data and retain effective posture information. At the same time, since the raw data may be collected based on the coordinate system of the sensing device itself, which differs from the coordinate system of the simulated birth canal virtual scene, the raw data can be uniformly transformed to the global coordinate system of the virtual scene to ensure the consistency of the data in the spatial dimension. After dual calibration processing of noise filtering and coordinate unification, standard hand posture data that can accurately reflect the trainee's real hand posture is obtained. Among them, the digital filtering algorithm is a technique that uses digital calculations to process discrete signals such as voltage values ​​and audio sampling data collected by sensors, filters out noise, and extracts useful information, replacing analog circuits with mathematical models such as difference equations and convolution operations.

[0067] Step S202: Based on the standard hand posture data and the true value of the index fingertip spatial coordinates corresponding to all calibration points, construct an optimization objective function; the objective function uses the set of hand biomechanical parameters as variables.

[0068] The optimization objective function determines the optimal set of hand biomechanical parameters by quantifying the difference between the fingertip position calculated based on the parameters and the true value of the actual measured fingertip position.

[0069] For example, based on the standard hand posture data corresponding to all calibration points and the true value of the spatial coordinates of the index fingertip, an optimization objective function is constructed with the set of hand biomechanical parameters as the only iterative variable. The value of this function directly reflects the degree of matching between the parameter set and the trainee's real hand features.

[0070] Step S203: Calculate the minimum value of the objective function using a nonlinear least squares algorithm; and determine the set of hand biomechanical parameters corresponding to the minimum value as the student's personalized set of hand biomechanical parameters.

[0071] Among them, the nonlinear least squares algorithm finds a set of model parameters that minimizes the sum of squared residuals between the model's predicted values ​​and the actual observed values. Here, nonlinearity means that the relationship between the model parameters and the observed values ​​cannot be described by a linear expression, such as when the model contains exponential, logarithmic, or trigonometric function terms in the parameters.

[0072] For example, a set of hand biomechanical parameters is initialized as the initial values ​​for iteration, and substituted into the objective function to calculate the initial error value. Then, a nonlinear least squares algorithm is used for iteration, calculating the partial derivatives of the error function with respect to each parameter to determine the direction and step size of parameter adjustment. The parameter set is updated step by step, and the objective function value is recalculated. This iterative process is repeated until the objective function value converges to a preset minimum threshold, at which point the impact of parameter set adjustments on the error is negligible. The hand biomechanical parameter set corresponding to this minimum value is determined as the student's personalized parameter set, which can accurately match individual differences such as hand size and joint characteristics. The preset minimum threshold is a pre-set criterion value used when solving for the minimum value of the objective function using the algorithm.

[0073] In this embodiment, standard hand posture data is obtained by calibrating the original hand posture data; an optimization objective function with personalized parameters as variables is constructed based on the standard data and the true values; the optimal parameters are solved using the least squares algorithm for adapting nonlinear problems, generating a personalized set of hand biomechanical parameters for each student. This effectively solves the problem of poor individual adaptability caused by the use of uniform hand parameters in traditional methods.

[0074] In one embodiment, based on the standard hand posture data corresponding to all calibration points and the true value of the spatial coordinates of the index fingertip, an optimization objective function is constructed, including:

[0075] Step S301: Based on the standard hand posture data and the true value of the index fingertip spatial coordinates corresponding to all calibration points, construct the optimization objective function:

[0076]

[0077] in, To optimize the objective function, Φ is the iterative variable of the parameter set. For positive kinematic functions, This represents the standard hand posture data corresponding to the i-th calibration point. Let N be the true spatial coordinates of the index fingertip corresponding to the i-th calibration point, and N be the total number of calibration points. Let be the square of the Euclidean norm. This is a minimum value operation.

[0078] For example, the standard hand posture data corresponding to each calibration point and the set of hand biomechanical parameters to be optimized are substituted into the forward kinematics function. By solving the kinematic equations of the hand joints, the theoretical spatial position of the index fingertip under this parameter set is calculated. Subsequently, the square of the Euclidean norm is used as an error metric to calculate the squared error between the theoretical position of the fingertip and the true value of the spatial coordinates of the index fingertip at each calibration point. The squared error values ​​of all calibration points are summed to construct an optimization objective function with the set of hand biomechanical parameters as the sole iterative variable. The value of this function directly reflects the degree of matching between the parameter set and the trainee's real hand features. The forward kinematics function, based on the principles of hand biomechanics, can calculate the theoretical spatial position of the index fingertip under this parameter set by solving the multi-joint cascade kinematic equations based on the input set of hand biomechanical parameters and standard hand posture data.

[0079] In this embodiment, a mapping relationship between parameters and fingertip position is established through a positive kinematic function; the error is quantified by the square of the Euclidean norm and summed to construct an optimization objective function that directly reflects the matching degree between the parameter set and the student's actual hand features. This provides a scientific and quantitative goal orientation for solving personalized hand biomechanical parameter sets using nonlinear least squares algorithms, effectively avoiding subsequent force feedback deviations caused by mismatch between parameters and individual hand features.

[0080] In one embodiment, based on current hand posture data and a personalized set of hand biomechanical parameters, the current position of the student's actual fingertips and the virtual force acting on those fingertips are calculated, including:

[0081] Step S401: The current hand posture data is calibrated to obtain the current standard hand posture data.

[0082] For example, during midwifery training, raw hand posture data is collected in real time. This data includes the real-time angles, angular velocities, and preliminary spatial position information of each hand joint. However, due to factors such as natural hand tremors and electromagnetic interference from sensor signals during training, the raw data is prone to random noise and local bias. A Kalman filter algorithm can be used to suppress noise in the raw data. By establishing state and observation equations, data bias is dynamically corrected, preserving effective posture information. The raw data is then transformed from its original coordinate system to a global coordinate system. After calibration, standard hand posture data that accurately reflects the trainee's real-time hand state is obtained. The Kalman filter algorithm is a linear, unbiased, minimum variance recursive state estimation algorithm used to address the problem that the state cannot be directly and accurately measured, and that the measurement process is noisy. Through a prediction-update loop, it fuses theoretical models (prior information) and actual observation data (posterior information) to obtain the optimal estimate of the true state.

[0083] Step S402: Based on the personalized hand biomechanical parameter set and the current standard hand posture data, the current position of the student's actual fingertips is obtained through forward kinematics calculation.

[0084] For example, a personalized set of hand biomechanical parameters and current standard hand posture data are input into a forward kinematics model. By solving the joint coordinate transfer equations—that is, starting from the wrist joint (the root node of hand movement) and deriving the spatial coordinates of the metacarpophalangeal joints, interphalangeal joints, and finally the fingertip of the index finger based on the angles of each joint and the relative position parameters between joints—the current position of the student's actual fingertip in the simulated birth canal virtual scene is calculated. The forward kinematics model is based on the anatomical structure of the human hand and follows the kinematic principles of multi-joint serial mechanisms, establishing a quantitative mapping relationship between hand biomechanical parameters and fingertip spatial position. The personalized set of hand biomechanical parameters includes parameters specific to the student, such as joint length, joint range of motion, and three-dimensional coordinates of muscle attachment points, serving as the core basis for adapting to the individual characteristics of the student's hand. The current standard hand posture data provides the real-time movement status of each joint in the hand.

[0085] Step S403: Determine the virtual contact point based on collision detection; and calculate the virtual force acting on the virtual contact point according to the biomechanical model of the virtual organ; the virtual force is the virtual force experienced by the real fingertip.

[0086] Collision detection is a core technology in fields such as computer graphics, game development, and physics simulation. It is used to determine whether two or more objects are in contact or overlapping in virtual space, and to determine the specific state of the contact, such as position, time, and force.

[0087] For example, collision detection can be used to determine the spatial intersection between the current position of the real fingertip and virtual organs such as the virtual fetal head and virtual birth canal wall in the virtual scene. When an overlap is detected between the fingertip position and the spatial range of a virtual organ, the overlapping area is defined as a virtual contact point, and key contact characteristic parameters such as contact depth and contact area are recorded simultaneously. The contact characteristic parameters are substituted into a preset virtual organ biomechanical model, and the contact depth and elastic reaction force are correlated through Hooke's law. The force value change during the dynamic contact process is corrected by the damping effect, and the virtual force acting on the virtual contact point is calculated. This virtual force completely corresponds to the mechanical feedback that the trainee's real fingertip should perceive during training, and can accurately simulate the real force feeling of touching human organs in clinical midwifery operations. The pre-defined virtual organ biomechanical model is constructed based on clinical human organ biomechanical experimental data, including core mechanical parameters such as the organ's elastic coefficient and damping coefficient, and can simulate the mechanical response characteristics of different organs under external contact. Damping effect describes the phenomenon where a kinetic system gradually decreases in amplitude and slows down until it stabilizes due to external resistance or internal energy dissipation; its core principle is to consume energy to suppress excessive movement. Hooke's law is a fundamental law in materials mechanics and elasticity, describing the relationship between the deformation of an elastic body and the applied external force within its elastic limit. The core expression of Hooke's law is:

[0088]

[0089] F is the elastic force acting on the elastic body; the negative sign indicates that the direction of the elastic force is opposite to the direction of deformation, which hinders deformation and conforms to the elastic characteristic of restoring the original length; k is the stiffness coefficient, also known as the elastic coefficient, which reflects the softness or hardness of the elastic body. The larger the value of k, the more difficult it is for the object to deform; x is the deformation of the elastic body, that is, the difference in length between the object after deformation and its original length (the distance of stretching or compression).

[0090] In this embodiment, by acquiring hand posture data during midwifery training and combining it with a personalized biomechanical parameter set for forward kinematic calculation, the current position of the trainee's actual fingertips is obtained; virtual contact points are determined based on collision detection; and virtual forces acting on the virtual contact points are calculated based on the biomechanical model of virtual organs. This effectively solves the problems of inaccurate fingertip position calculation and the disconnect between virtual force perception and clinical reality in traditional nursing and midwifery training.

[0091] In one embodiment, based on a personalized set of hand biomechanical parameters and current hand posture data, virtual forces are mapped to generate force feedback device control commands, including:

[0092] Step S501: Based on the personalized hand biomechanical parameter set and the current hand posture data, calculate the Jacobian matrix from the fixed reference point of the force feedback device to the current position of the actual fingertip.

[0093] Among them, the fixed reference point of the force feedback device is the preset mechanical transmission benchmark point of the device, which is usually located in the core area where the device contacts the hand, such as the grip of the handle, and is used to unify the calculation benchmark of force and torque; the Jacobian matrix is ​​a linearized expression of the kinematic mapping, with each row corresponding to the movement of the fingertip in a certain spatial coordinate axis direction, and each column corresponding to the movement of a certain joint of the device, which can quantitatively describe the transmission relationship between the joint force / torque at the fixed reference point and the force / displacement at the fingertip.

[0094] For example, a personalized set of hand biomechanical parameters (including student-specific joint lengths, muscle attachment point coordinates, etc.) and current standard hand posture data (calibrated joint angles and spatial positions) are substituted into the hand kinematic model to construct a kinematic link from a fixed reference point to the current position of the actual fingertip. Based on this link, the Jacobian matrix is ​​calculated by solving the kinematic partial differential equations. The hand kinematic model is a mathematical / physical model describing the skeletal structure and joint movement of the hand. Through quantification, it accurately represents the mapping relationship from skeletal posture to spatial movement of the hand. Unlike the hand dynamics model, it does not involve dynamic factors such as muscle force. The kinematic partial differential equations characterize the probability distribution functions of particle states such as position, velocity, and energy over time and space, focusing on macroscopic / statistical motion characteristics rather than directly tracking individual particles.

[0095] Step S502: Based on the Jacobian matrix and virtual force, the equivalent force and equivalent torque that the force feedback device needs to apply at the fixed reference point are obtained through virtual work principle mapping processing.

[0096] Among them, the virtual work principle mapping process combines the virtual work principle in classical mechanics with the mathematical / physical tool of mapping to simplify complex systems, such as multi-degree-of-freedom, nonlinear or continuous media systems. By mapping, an equivalent relationship is established between the real complex system and the simplified model, so that the virtual work principle can be applied to complex scenarios more efficiently. Complex systems refer to systems composed of a large number of interrelated and interacting components. Their overall behavior cannot be predicted by simply analyzing each part. They usually have characteristics such as nonlinearity, self-organization, emergence and adaptability.

[0097] Specifically, the virtual work principle can be used to map and process the force at the fingertip into a signal executable by the force feedback device. The virtual work done by the virtual force at the fingertip is equal to the virtual work done by the force / torque applied by the device at the fixed reference point, ensuring the conservation of mechanical energy transfer. For example, the virtual force vector is transposed with the Jacobian matrix. Matrix transpose is used to match the directions of force and torque. The equivalent force (linear force along the spatial coordinate axis) and equivalent torque (rotational torque about the spatial coordinate axis) at the fixed reference point can be calculated using vector-matrix multiplication. Vector-matrix multiplication involves linearly transforming a vector using a matrix, outputting a new vector.

[0098] Step S503: Combine the equivalent force and equivalent torque to generate control commands for the force feedback device.

[0099] The control commands of force feedback devices typically include parameter fields such as the magnitude and direction of the force, the axial direction of the torque, the duration, and the execution priority.

[0100] For example, the obtained equivalent force is decomposed into components along the X, Y, and Z spatial axes, and the equivalent torque is decomposed into components around the X, Y, and Z axes. Both are then parameterized and encapsulated, for instance, by converting the force / torque values ​​into digital signals readable by the device through binary encoding. A check field is added to ensure the accuracy of command transmission. After encapsulation, the combined control command is sent to the force feedback device. This command directly instructs the servo motor, piezoelectric sensor, and other drive units of the force feedback device to output a realistic force sensation according to preset force / torque parameters. The binary encoding, through the combination of 0s and 1s, bridges the gap between human-understandable information and machine-processable signals.

[0101] In this embodiment, the Jacobian matrix adapted to the student's hand features is calculated; based on the principle of virtual work, the equivalent force and equivalent torque that the force feedback device needs to apply at a fixed reference point are obtained, and control commands for the force feedback device are generated. This effectively solves the problem of force focus offset caused by general parameters in traditional force feedback devices, ensuring that the real force output by the device perfectly matches the virtual force that the student's fingertips should perceive.

[0102] In one embodiment, after executing the control command from the force feedback device, the trainee's operational data is acquired; and based on the operational data, an operational assessment result for nursing and midwifery teaching and training is generated, including:

[0103] Step S601: After executing the control command of the force feedback device, acquire the trainee's operation data during the training process; the operation data includes force time-series data of virtual force changing over time and trajectory time-series data of hand movement trajectory changing over time.

[0104] For example, control commands are received in real time, and after executing the control commands from the force feedback device, the trainee's operational data during training is acquired. The acquisition of force timing data requires associating it with equivalent force parameters contained in the control commands: recording the virtual force values ​​corresponding to the control commands at preset time intervals (e.g., milliseconds), and adding timestamps to form a "time-virtual force" sequence data. This data reflects the dynamic change trend of the virtual force throughout the training process, such as the rise and fall of force during uterine contraction simulation. Trajectory timing data records the spatial coordinates of key hand feature points such as the trainee's wrist joint and index fingertip in real time, combining them with timestamps to generate a "time-space coordinate" trajectory sequence, fully presenting the path and rhythm of hand movements. Both types of data are aligned by timestamps and stored as operational data.

[0105] Step S602: Compare the operation data with the preset expert standard operation data to obtain the comparison results.

[0106] Among them, the pre-stored expert standard operation data is a benchmark dataset generated by collecting standard operations from multiple senior midwives in the same simulated birth canal virtual scenario. It includes force time series standard sequences and trajectory time series standard sequences, which cover clinically recognized average operation pressure range, pressure fluctuation threshold, motion trajectory smoothness standards, etc.

[0107] For example, a multi-dimensional feature extraction and comparison algorithm can be used: for force time series data, features such as average force value, force fluctuation variance reflecting pressure stability, and force change rate can be extracted; for trajectory time series data, features such as trajectory overlap, motion speed standard deviation reflecting smoothness, and arrival time of key operation points can be extracted. The feature difference between student operation data and expert standard data is calculated using a dynamic time warping algorithm. This algorithm can adapt to differences in sequence length, accurately quantify the similarity of the two types of data in the time dimension, and output the difference values ​​for each feature dimension, forming a comparison result. The multi-dimensional feature extraction and comparison algorithm refers to accurately extracting key information from complex data and scientifically comparing data differences / similarity, solving the problem that single-dimensional analysis cannot depict the full picture of the data; the dynamic time warping algorithm is a classic algorithm used to align and calculate the similarity between two time series of different lengths and mismatched time axes, solving the problem that time scaling / offset makes it difficult to directly compare similar sequences.

[0108] Step S603: Based on the comparison results, generate operational assessment results for nursing and midwifery teaching and training.

[0109] For example, weights are assigned to each difference feature based on the importance of the clinical operation; for instance, pressure stability is weighted higher than trajectory overlap. The differences in each dimension are converted into quantitative scores, with smaller differences resulting in higher scores. A total score is then calculated based on these weights to form an intuitive evaluation of the operational compliance level. Improvement suggestions are generated for the differences in each feature dimension. For example, if the pressure stability difference is high (force fluctuations exceed the standard range), it is recommended to focus on the evenness of hand force application during training, avoiding sudden increases or decreases in force. If the trajectory smoothness difference is high, it is recommended to refer to expert trajectories to optimize hand movement paths, reducing unnecessary pauses and reversals. The quantitative scores (including the total score and sub-scores for each dimension), difference analysis charts, and targeted improvement suggestions are integrated into a structured evaluation report, presented in visual formats such as tables and line graphs.

[0110] In this embodiment, the operational data of trainees during training is acquired; the operational data and expert standards are quantitatively compared in multiple dimensions to obtain comparison results; based on the comparison results, the differences are transformed into intuitive scores and targeted suggestions, generating operational assessment results for nursing and midwifery teaching and training. This effectively solves the problems of lack of data support and vague feedback in traditional nursing and midwifery training assessments, helping trainees to improve operational details in a targeted manner, quickly align with clinical standards, and improve training efficiency and practical accuracy.

[0111] In one embodiment, calibration data of trainees in a simulated birth canal virtual scenario is acquired; and all calibration data are aggregated to generate a calibration dataset, including:

[0112] Step S701: Set calibration points in the simulated birth canal virtual scene according to the preset clinical palpation path.

[0113] For example, based on human birth canal anatomy data and clinical midwifery operation standards, a highly realistic virtual birth canal scenario is constructed. The scenario includes key anatomical structures such as a virtual cervix, a virtual fetus, and a virtual birth canal wall, with structural morphology and mechanical properties closely matching clinical reality. Referring to the standard palpation procedures of experienced midwives, a pre-set clinical palpation path is established. This path covers key operational steps such as cervical position exploration, fetal presenting part assessment, and birth canal elasticity evaluation, conforming to the operational logic and spatial trajectory of clinical midwifery palpation. Calibration points are evenly deployed along the pre-set path according to operational stages, covering the path's starting point, operational turning points, and key judgment nodes. Each calibration point is precisely located using the three-dimensional coordinates of the simulated scenario's global coordinate system, ensuring that the calibration points reflect the hand posture requirements at different operational stages and cover the core movement range of clinical palpation.

[0114] Step S702: Obtain the raw data of the student's hand posture and the true value of the spatial coordinates of the index fingertip when the student touches each calibration point.

[0115] For example, when a student's fingertip touches a calibration point in a virtual scene, the system automatically collects the student's raw hand posture data and the true spatial coordinates of the index fingertip at each calibration point. The raw hand posture data is acquired through an inertial measurement unit and hand joint sensors, recording in real time the angles, angular velocities, and preliminary spatial position information of each joint of the hand when the student touches the calibration point. This data directly reflects the dynamic posture characteristics of the hand. The true spatial coordinates of the index fingertip are collected through an optical tracking system. This system is based on the principle of multi-camera visual positioning, which calculates and outputs the high-precision three-dimensional coordinates of the fingertip in the global coordinate system of the simulated scene by capturing the marker points worn on the fingertip. Among them, an inertial measurement unit (IMU) is a sensor component based on the principle of inertia, which can measure the motion state of an object in real time and provide key data on the device's position and attitude, and can work without relying on external signals; a hand joint sensor is a small electronic device used to collect hand joint motion data, physiological signals or environmental interaction information, which can convert physical changes such as the angle, displacement, and pressure of the hand joint into processable electrical signals; an optical tracking system is a technical system that captures the spatial position, motion trajectory or attitude information of a target through optical principles (such as visible light and infrared light), and can realize real-time positioning and dynamic tracking of specific targets such as objects, human bodies, and equipment parts; the multi-camera visual positioning principle is to collect environmental information from different perspectives by multiple cameras, combine the data with computer vision algorithms, and finally calculate the position and attitude of the target (or the camera itself) in three-dimensional space.

[0116] Step S703: Summarize all raw hand posture data and the true values ​​of the index fingertip spatial coordinates to obtain the calibration dataset.

[0117] For example, a data index is established according to the calibration point number. The original hand posture data and the true spatial coordinates of the index fingertip corresponding to each calibration point are paired to form an associated data unit of "calibration point (IDIdentifier) ​​- original hand posture data - true spatial coordinates of the index fingertip". All associated data units are combined to generate a calibration dataset. This dataset contains fields such as global coordinates of calibration points, data acquisition timestamps, original parameters of each joint of the hand, and true coordinates of the fingertip.

[0118] In this embodiment, calibration points are set according to a preset clinical palpation path in a simulated birth canal virtual scene; raw data of the trainee's hand posture and true values ​​of the index fingertip spatial coordinates are obtained when the trainee performs palpation at each calibration point; all raw data of hand posture and true values ​​of the index fingertip spatial coordinates are summarized to obtain a calibration dataset. This effectively avoids the problems of poor data quality and insufficient representativeness in traditional calibration processes, which lead to low adaptability of personalized parameters.

[0119] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0120] Based on the same inventive concept, this application also provides an intelligent auxiliary system for nursing and midwifery teaching and training to implement the intelligent auxiliary method for nursing and midwifery teaching and training described above. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the intelligent auxiliary system for nursing and midwifery teaching and training provided below can be found in the limitations of the intelligent auxiliary method for nursing and midwifery teaching and training described above, and will not be repeated here.

[0121] In one exemplary embodiment, such as Figure 3 As shown, a nursing and midwifery teaching and training intelligent auxiliary system 300 is provided, including:

[0122] The data acquisition module 301 is used to acquire the calibration data of trainees in the simulated birth canal virtual scene; and to summarize all the calibration data to generate a calibration dataset; the calibration data includes the original hand posture data and the true value of the spatial coordinates of the index fingertip;

[0123] The parameter set generation module 302 is used to generate a personalized set of hand biomechanical parameters for trainees based on the calibration dataset through optimized computational processing.

[0124] The virtual force calculation module 303 is used to acquire the trainee's current hand posture data during midwifery operation training; and based on the current hand posture data and personalized hand biomechanical parameter set, to calculate the current position of the trainee's real fingertips and the virtual force on the real fingertips.

[0125] The instruction generation module 304 is used to map virtual forces based on a personalized set of hand biomechanical parameters and current hand posture data, and generate force feedback device control instructions; the force feedback device control instructions are used to instruct the force feedback device to output a realistic force sensation corresponding to the virtual scene;

[0126] The assessment result generation module 305 is used to acquire the trainee's operation data after the control command of the force feedback device; and to generate the operation assessment results of nursing and midwifery teaching and training based on the operation data.

[0127] In one embodiment, the parameter set generation module 302 is further configured to:

[0128] The raw hand posture data in the calibration dataset is calibrated to obtain standard hand posture data;

[0129] Based on the standard hand posture data and the true value of the index fingertip spatial coordinates corresponding to all calibration points, an optimization objective function is constructed; the objective function uses the set of hand biomechanical parameters as variables;

[0130] The minimum value of the objective function is calculated using a nonlinear least squares algorithm; and the set of hand biomechanical parameters corresponding to the minimum value is determined as the student's personalized hand biomechanical parameter set.

[0131] In one embodiment, the parameter set generation module 302 is further configured to:

[0132] Based on the standard hand posture data and the true spatial coordinates of the index fingertip corresponding to all calibration points, an optimization objective function is constructed:

[0133]

[0134] in, To optimize the objective function, Φ is the iterative variable of the parameter set. For positive kinematic functions, This represents the standard hand posture data corresponding to the i-th calibration point. Let N be the true spatial coordinates of the index fingertip corresponding to the i-th calibration point, and N be the total number of calibration points. Let be the square of the Euclidean norm. This is a minimum value operation.

[0135] In one embodiment, the virtual force calculation module 303 is further configured to:

[0136] The current hand posture data is calibrated to obtain the current standard hand posture data;

[0137] Based on a personalized set of hand biomechanical parameters and current standard hand posture data, the current position of the student's actual fingertips is obtained through forward kinematics calculations.

[0138] Virtual contact points are determined based on collision detection; and virtual forces acting on virtual contact points are calculated based on the biomechanical model of virtual organs; the virtual forces are the virtual forces experienced by the real fingertips.

[0139] In one embodiment, the instruction generation module 304 is further configured to:

[0140] Based on a personalized set of hand biomechanical parameters and current hand posture data, the Jacobian matrix from the fixed reference point of the force feedback device to the current position of the actual fingertip is calculated.

[0141] Based on the Jacobian matrix and virtual force, the equivalent force and equivalent torque that the force feedback device needs to apply at a fixed reference point are obtained through mapping processing using the principle of virtual work.

[0142] The equivalent force and equivalent torque are combined to generate control commands for the force feedback device.

[0143] In one embodiment, the evaluation result generation module 305 is further configured to:

[0144] After executing the control commands of the force feedback device, the training process acquires the trainee's operational data; the operational data includes force time-series data of virtual force changing over time and trajectory time-series data of hand movement trajectory changing over time;

[0145] The operational data is compared with the preset expert standard operational data to obtain the comparison results;

[0146] Based on the comparison results, operational assessment results for nursing and midwifery teaching and training are generated.

[0147] In one embodiment, the data acquisition module 301 is further configured to:

[0148] In a simulated birth canal virtual scene, calibration points are set according to a preset clinical palpation path;

[0149] Obtain raw data of the student's hand posture and true spatial coordinates of the index fingertip when the student touches each calibration point;

[0150] The calibration dataset is obtained by summarizing all the raw hand posture data and the true values ​​of the spatial coordinates of the index fingertip.

[0151] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the intelligent auxiliary method for nursing and midwifery teaching and training as described above.

[0152] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0153] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0154] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. An intelligent auxiliary method for nursing and midwifery teaching and training, characterized in that: The method includes: Acquire calibration data of trainees in a simulated birth canal virtual scenario; and summarize all the calibration data to generate a calibration dataset; the calibration data includes raw hand posture data and true values ​​of the spatial coordinates of the index fingertip; Based on the calibration dataset, a personalized set of hand biomechanical parameters for the trainee is generated through optimized computational processing. Acquire the trainee's current hand posture data during midwifery operation training; and based on the current hand posture data and the personalized hand biomechanical parameter set, calculate the current position of the trainee's actual fingertips and the virtual force acting on the actual fingertips; Based on the personalized hand biomechanical parameter set and the current hand posture data, the virtual force is mapped to generate force feedback device control instructions; the force feedback device control instructions are used to instruct the force feedback device to output a realistic force sensation corresponding to the virtual scene. After executing the control command of the force feedback device, the operation data of the trainee is acquired; and based on the operation data, an operation evaluation result for nursing and midwifery teaching and training is generated.

2. The method according to claim 1, characterized in that, Based on the calibration dataset, the personalized hand biomechanical parameter set of the trainee is generated through optimized computational processing, including: The original hand posture data in the calibration dataset is calibrated to obtain standard hand posture data; Based on the standard hand posture data corresponding to all calibration points and the true value of the spatial coordinates of the index fingertip, an optimization objective function is constructed; the objective function uses the set of hand biomechanical parameters as variables; The minimum value of the optimization objective function is calculated using a nonlinear least squares algorithm; and the set of hand biomechanical parameters corresponding to the minimum value is determined as the student's personalized set of hand biomechanical parameters.

3. The method according to claim 2, characterized in that, The optimization objective function is constructed based on the standard hand posture data corresponding to all calibration points and the true value of the index fingertip spatial coordinates, including: Based on the standard hand posture data corresponding to all calibration points and the true value of the spatial coordinates of the index fingertip, an optimization objective function is constructed: in, To optimize the objective function, Φ is the iterative variable of the parameter set. For positive kinematic functions, This represents the standard hand posture data corresponding to the i-th calibration point. Let N be the true spatial coordinates of the index fingertip corresponding to the i-th calibration point, and N be the total number of calibration points. Let be the square of the Euclidean norm. This is a minimum value operation.

4. The method according to claim 1, characterized in that, The calculation of the current position of the student's actual fingertips and the virtual force acting on those fingertips, based on the current hand posture data and the personalized hand biomechanical parameter set, includes: The current hand posture data is calibrated to obtain the current standard hand posture data; Based on the personalized hand biomechanical parameter set and the current standard hand posture data, the current position of the student's actual fingertips is obtained through forward kinematics calculation; The virtual contact point is determined based on collision detection; and the virtual force acting on the virtual contact point is calculated based on the biomechanical model of the virtual organ; the virtual force is the virtual force experienced by the real fingertip.

5. The method according to claim 1, characterized in that, The process of mapping the virtual force based on the personalized hand biomechanical parameter set and the current hand posture data to generate force feedback device control commands includes: Based on the personalized hand biomechanical parameter set and the current hand posture data, calculate the Jacobian matrix from the fixed reference point of the force feedback device to the current position of the actual fingertip; Based on the Jacobian matrix and the virtual force, the equivalent force and equivalent torque that the force feedback device needs to apply at the fixed reference point are obtained through virtual work principle mapping. The equivalent force and equivalent torque are combined to generate the control command for the force feedback device.

6. The method according to claim 1, characterized in that, After executing the control command of the force feedback device, the student's operation data is acquired; Based on the operational data, operational assessment results for nursing and midwifery teaching and training are generated, including: After executing the control command of the force feedback device, the operation data of the trainee during the training process is acquired; the operation data includes force time-series data of virtual force changing over time and trajectory time-series data of hand movement trajectory changing over time; The operation data is compared with the preset expert standard operation data to obtain the comparison result; Based on the comparison results, operational assessment results for nursing and midwifery teaching and training are generated.

7. The method according to claim 1, characterized in that, The process involves acquiring calibration data from trainees in a simulated birth canal virtual scenario; And by summarizing all the calibration data, a calibration dataset is generated, including: In a simulated birth canal virtual scene, calibration points are set according to a preset clinical palpation path; Obtain the raw hand posture data and true spatial coordinates of the index fingertip when the student touches each of the calibration points; The calibration dataset is obtained by summarizing all the original hand posture data and the true values ​​of the index fingertip spatial coordinates.

8. An intelligent auxiliary system for nursing and midwifery teaching and training, characterized in that: The system includes: The data acquisition module is used to acquire calibration data of trainees in a simulated birth canal virtual scene; and to summarize all the calibration data to generate a calibration dataset; the calibration data includes raw hand posture data and true values ​​of the spatial coordinates of the index fingertip; The parameter set generation module is used to generate a personalized set of hand biomechanical parameters for the trainee based on the calibration dataset through optimized computational processing. The virtual force calculation module is used to acquire the trainee's current hand posture data during midwifery operation training; and based on the current hand posture data and the personalized hand biomechanical parameter set, to calculate the current position of the trainee's real fingertips and the virtual force acting on the real fingertips. The instruction generation module is used to map the virtual force based on the personalized hand biomechanical parameter set and the current hand posture data, and generate force feedback device control instructions; the force feedback device control instructions are used to instruct the force feedback device to output a realistic force sensation corresponding to the virtual scene; The assessment result generation module is used to acquire the trainee's operation data after executing the control command of the force feedback device; and to generate an operation assessment result for nursing and midwifery teaching and training based on the operation data.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images