Traditional Chinese medicine pediatric acupuncture training system combined with virtual reality guidance

Abstract

The invention relates to the technical field of virtual reality, in particular to a traditional Chinese medicine pediatric acupuncture training system combined with virtual reality guidance, which comprises a virtual behavior generation module, a feedback module, an evaluation module and a guidance module. By simulating the physiological behavior of a pediatric child patient, providing sensory feedback, quantitatively evaluating the operation technique in real time and guiding the dialectical diagnosis and treatment process, high-simulation pediatric acupuncture skill training is achieved, and the clinical pertinence of training is effectively improved.

Description

Technical Field

[0001] This invention relates to the field of virtual reality technology, and in particular to a training system for pediatric acupuncture in traditional Chinese medicine that incorporates virtual reality guidance. Background Technology

[0002] Traditional Chinese medicine acupuncture training mainly relies on methods such as human model practice and clinical observation. While these methods have a certain teaching effect, they have problems when simulating the reactions of real pediatric patients.

[0003] Existing virtual reality training systems focus on adult acupuncture simulation, failing to consider the unique characteristics of pediatric acupuncture and thus unable to simulate the operational challenges of real-world scenarios such as crying and movement of children. Furthermore, existing virtual reality training systems lack precision in key skill indicators such as force control and needle insertion angle calibration. Summary of the Invention

[0004] The purpose of this invention is to provide a traditional Chinese medicine pediatric acupuncture training system that combines virtual reality guidance to solve the technical problem that existing acupuncture training systems cannot simulate the behavioral responses of real children.

[0005] This invention provides a training system for pediatric acupuncture in traditional Chinese medicine that incorporates virtual reality guidance, comprising:

[0006] The virtual behavior generation module is used to dynamically generate physiological and behavioral state data of virtual children based on preset pediatric patient characteristics.

[0007] The feedback module is used to receive status data and generate corresponding visual, auditory, and tactile feedback;

[0008] The assessment module is used to collect the operational data of trainees and compare and analyze it with the standard operation database to generate assessment results.

[0009] The guidance module generates diagnostic and treatment guidance information based on the assessment results and the theory of syndrome differentiation in traditional Chinese medicine pediatrics.

[0010] The virtual behavior generation module includes a physiological state simulation unit and a behavioral response simulation unit. The physiological state simulation unit constructs digital models of the heart, liver, spleen, lungs, and kidneys based on the theory of viscera in traditional Chinese medicine pediatrics. The behavioral response simulation unit constructs a typical behavioral pattern library for children aged 0 to 12 years based on pediatric developmental psychology. The behavioral response simulation unit receives the heart, liver, spleen, lung, and kidney status data output by the physiological state simulation unit, and calculates the virtual child's behavior at the current moment by combining age grouping parameters and using a decision tree algorithm. The behavior includes at least the range of limb movement, the intensity of crying, and the degree of attention.

[0011] In some embodiments, the evaluation module includes at least a data acquisition unit, a feature extraction unit, and an evaluation and analysis unit; the data acquisition unit acquires spatial pose data of acupuncture needles; the feature extraction unit performs sliding window processing on the spatial pose data and extracts the average needle insertion speed, standard deviation of insertion and withdrawal amplitude, and peak twisting angle from each window; the evaluation and analysis unit performs dynamic time warping calculations on the average needle insertion speed, standard deviation of insertion and withdrawal amplitude, and peak twisting angle against a standard database, outputs a quantitative score for the proficiency of the technique, and identifies operational errors such as angle deviation and excessive force.

[0012] In some embodiments, the guidance module includes a diagnostic information integration unit, a syndrome judgment unit, and a prescription generation unit; the diagnostic information integration unit receives simulated symptom data from the virtual behavior generation module and student operation data from the evaluation module; the syndrome judgment unit constructs a syndrome reasoning model based on a Bayesian network; the prescription generation unit stores a syndrome-acupoint prescription mapping rule library established based on classic Chinese pediatric medical texts, and generates recommended acupoint prescriptions and acupuncture techniques based on the data from the syndrome judgment unit, and displays them in real time overlaid in the virtual scene.

[0013] In some embodiments, the network nodes of the syndrome reasoning model include eight common pediatric syndrome types and 35 corresponding key symptoms and signs.

[0014] In some embodiments, when the syndrome reasoning model is updated, historical data of the operations performed by the trainee in the current training session are incorporated as a new basis for adjusting the prior probability of subsequent syndrome judgments.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] 1. The virtual behavior generation module is based on the theory of viscera in traditional Chinese medicine and developmental psychology. It can dynamically generate realistic challenge scenarios, including crying and limb movement of different intensities. This allows trainees to adapt to and learn to cope with the actual difficulties in pediatric clinical practice in a highly simulated environment, effectively improving the clinical relevance of the training.

[0017] 2. The guidance module guides trainees through the entire process from collecting diagnostic information, through syndrome analysis, to prescription formulation, ensuring that virtual skills training is closely integrated with TCM clinical diagnosis and treatment practice, and solving the problem of disconnect between skills operation and theoretical decision-making in previous training. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the overall technical structure of the Traditional Chinese Medicine Pediatric Acupuncture Training System of the present invention;

[0020] Figure 2 This is the core principle diagram of the guiding module in this invention. Detailed Implementation

[0021] The following will be based on embodiments of the present invention. Figures 1-2 The technical solutions in the embodiments of the present invention will be clearly and completely described together. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0022] Partial interpretation:

[0023] 1. The nonlinear differential equation system of the physiological state simulation unit can be generally represented as a series of coupled equations;

[0024] 2. The core of the dynamic time warping algorithm used by the evaluation and analysis unit lies in calculating the minimum cumulative distance between the real-time operational feature sequence and the standard sequence. This distance calculation takes into account the weights of different feature dimensions.

[0025] Example

[0026] This embodiment provides a TCM pediatric acupuncture training system combined with virtual reality guidance, including a virtual behavior generation module, a feedback module, an assessment module, and a guidance module. The virtual behavior generation module dynamically generates physiological and behavioral state data of a virtual child based on preset pediatric patient characteristics. The feedback module receives the state data and generates corresponding visual, auditory, and tactile feedback. The assessment module collects the trainee's operational data and compares it with a standard operational database to generate assessment results. Based on the assessment results and according to TCM pediatric diagnostic theories, the guidance module generates diagnostic and treatment plan guidance information.

[0027] The virtual behavior generation module includes a physiological state simulation unit and a behavioral response simulation unit. The physiological state simulation unit constructs digital models of the five core organ systems—heart, liver, spleen, lungs, and kidneys—based on the visceral theory of traditional Chinese medicine pediatrics. Under normal circumstances, each digital model describes the dynamic evolution of its state variables through a set of mutually coupled nonlinear differential equations. For example, the heart model might describe parameters related to consciousness and blood vessels, the liver model to parameters related to the free flow of qi and tendons, the spleen model to parameters related to the transportation and transformation of qi and muscles, the lung model to parameters related to qi and skin, and the kidney model to parameters related to growth and development and bones. The remaining key parameters in the non-differential equations can be set by the instructor through a pre-configured interface to simulate different specific medical conditions.

[0028] Furthermore, the physiological state simulation unit performs numerical integration calculations at a frequency of 100Hz and outputs vector data of the current overall physiological state of the virtual child in real time. This vector contains at least 20 physiological parameter variables from the five major organ systems: heart, liver, spleen, lungs, and kidneys.

[0029] Furthermore, the behavioral response simulation unit is based on the theory of pediatric developmental psychology and constructs a behavioral pattern library covering ages 0 to 12. This behavioral pattern library divides the entire age range into 6 age groups, and each age group has corresponding typical behavioral pattern characteristics, including the range of basic physical activity ability, typical response patterns to stimuli, attention span, and emotional expression.

[0030] Understandably, the behavioral response simulation unit achieves the aforementioned results using a behavioral decision tree algorithm. This algorithm takes the real-time physiological state vector output by the physiological state simulation unit and preset age group parameters as input. The decision tree first normalizes and extracts features from the input vector, identifying the currently dominant organ dysfunction type and its severity level. Then, the decision tree, combined with the age group parameters, retrieves the corresponding behavioral baseline from the behavioral pattern library. Finally, through preset logical judgment nodes, it calculates the quantitative indicators of the virtual child's specific behavioral performance at the current moment. Generally, the quantitative indicators need to include three aspects: limb movement amplitude, crying intensity level, and attention concentration index. Limb movement amplitude is a normalized value between 0 and 1, where 0 represents complete stillness and 1 represents violent struggle; crying intensity level is usually divided into six discrete levels from 0 to 5, where 0 represents no crying and 5 represents extreme crying; the attention concentration index is a percentage value, representing the virtual child's ability to maintain attention to current environmental stimuli.

[0031] The feedback module receives behavior state data from the virtual behavior generation module. In some optional implementations, after receiving the behavior state data, the feedback module will realize visual, auditory and tactile feedback through visual rendering, auditory synthesis and tactile simulation.

[0032] Furthermore, during the visual rendering process, different anatomical models of children's bodies corresponding to the six age groups are constructed. Each anatomical model includes skeletal structure, muscle distribution, skin texture, and the location and depth of key acupoints. During simulation training, the visual rendering not only presents the overall appearance and dynamic behavior of the virtual child but also dynamically renders the acupoint areas with colors. The visual rendering continuously receives feedback data from the evaluation module, including but not limited to the three-dimensional spatial deviation between the needle tip and the center of the target acupoint, and the difference between the needle insertion force and the standard value.

[0033] In some optional embodiments, the system of the present invention also has a preset continuous color mapping table from green through yellow to red. When the operation accuracy is high and the deviation value is lower than the set threshold, the acupoint area presents a reassuring green; as the deviation increases, the color gradually transitions to yellow; when the deviation exceeds the allowable range, the acupoint area will turn significantly red.

[0034] Auditory synthesis is based on a rigorously collected and categorized audio sample library. All original audio materials in the library are derived from recordings of pediatric acupuncture clinical environments, and the library is indexed according to sound type and intensity. Auditory synthesis uses real-time data on the crying intensity level transmitted by the behavioral response simulation unit to synthesize and play corresponding environmental sounds. For example, a crying intensity level of 0 results in only mild background noise; a crying intensity level of 1 results in low-intensity sobbing; a crying intensity level of 3 or 4 results in loud crying and struggling / friction sounds; and a crying intensity level of 5 results in screaming and intense movement noise. The volume and reverberation of all audio elements are adjusted in real-time according to the crying intensity level to ensure complete synchronization between auditory feedback and visually observed child behavior, enhancing immersion.

[0035] Tactile simulation is integrated into the grip area of ​​a specially designed simulated needle. Inside the simulated needle, a set of miniature linear resonant actuator arrays is embedded. This array consists of independent actuator elements arranged along the needle axis. Each actuator can independently receive control signals and generate vibrations of different frequencies and amplitudes.

[0036] The tactile simulation uses data from a virtual pediatric anatomical model, along with real-time needle position and motion data from the evaluation module, to perform calculations. For example, as the needle penetrates the epidermis, dermis, subcutaneous tissue, and approaches fascia or periosteum, the resistance changes significantly. The calculated resistance feedback signal is converted into specific driving commands and sent to a miniature linear resonant actuator array. The actuator array generates a stepped vibration waveform that simulates the sensation of tissue layers according to the commands. Understandably, on the skin surface, a low-frequency, low-amplitude slight vibration may be generated to simulate the sensation of dense connective tissue in the epidermis; after penetrating the epidermis, the vibration may briefly disappear to simulate a breakthrough sensation; entering the fat layer may produce a low-frequency, soft vibration; and when the needle tip approaches deep sensitive structures or encounters simulated muscle knots, a strong, noticeable resistance vibration may be generated.

[0037] The evaluation module uses sensor data acquisition, signal processing, and pattern recognition algorithms to evaluate key indicators of acupuncture techniques. Furthermore, the evaluation module includes at least a data acquisition unit, a feature extraction unit, and an evaluation and analysis unit.

[0038] The data acquisition unit is a highly integrated sensor system, mainly consisting of two parts: a 9-axis inertial measurement unit inside the simulated needle and an optical positioning system set around the training space. The 9-axis inertial measurement unit integrates a three-axis gyroscope, a three-axis accelerometer, and a three-axis magnetometer, enabling it to continuously acquire data on the needle's three-axis angular velocity, three-axis linear acceleration, and orientation relative to the Earth's magnetic field in space at a high sampling frequency of 1000Hz.

[0039] Optical positioning systems typically employ high-speed infrared cameras to capture reflective markers fixed on simulated needles, providing absolute position and orientation data of the needle tip and specific parts in three-dimensional space at a frequency of 500Hz.

[0040] The feature extraction unit employs a sliding window processing technique to segment and analyze the input time-series data. The window length is fixed at 128 sampling points, with a sampling frequency of 1000Hz. Adjacent windows have a 50% overlap, ensuring feature continuity and capturing rapidly changing operational details.

[0041] For each data window, the feature extraction unit performs digital signal processing and statistical calculations, ultimately outputting a 12-dimensional operational feature vector. The operational feature vector includes at least: the average speed during needle insertion, peak acceleration, average insertion / lifting amplitude, standard deviation of insertion / lifting amplitude, insertion / lifting frequency, peak twisting angle, average twisting angle, twisting frequency, and overall operational stability indices such as acceleration variance and attitude angle change rate.

[0042] The evaluation and analysis unit contains a standard feature vector database, which includes ideal feature vectors for various standard techniques in different contexts.

[0043] The evaluation and analysis unit is equipped with a dynamic time warping algorithm, which can effectively compare the similarity of two sequences that may have differences in scaling or speed along the time axis. When the feature extraction unit sends a real-time 12-dimensional operational feature vector sequence, the evaluation and analysis unit performs dynamic time warping calculations against the corresponding standard sequences pre-stored in the standard database. The algorithm assigns a differentiated weight coefficient to each dimension. These weight coefficients are obtained through rigorous discriminant analysis of a large amount of expert operation data and novice operation data. The weight coefficients reflect the importance of different feature dimensions in distinguishing operational proficiency. The dynamic time warping calculation ultimately outputs a comprehensive similarity score, which is mapped to a quantitative score for skill proficiency, typically expressed as a value between 0 and 100.

[0044] The guidance module guides trainees through the entire process from information collection to diagnosis to treatment. The guidance module consists of at least a diagnosis and information integration unit, a syndrome judgment unit, and a prescription generation unit.

[0045] Furthermore, the diagnostic information integration unit receives two types of data: First, simulated symptom data from the virtual child behavior generation module. This data represents the external manifestations of the virtual child's physiological and behavioral state, corresponding to information from traditional Chinese medicine's observation and auscultation methods, including but not limited to the child's complexion, mental state, limb movements, and crying characteristics. Second, information actively input by the trainee through the system's interactive interface. This simulates a consultation process, where the trainee can ask the virtual child or their simulated guardian about medical history, symptom details, etc. After receiving these two types of data, the diagnostic information integration unit processes the aforementioned multi-source heterogeneous information to form a unified diagnostic information dataset suitable for subsequent analysis.

[0046] The syndrome judgment unit is based on a syndrome reasoning model constructed using a Bayesian network. It should be noted that the Bayesian network model was constructed under the guidance of experts in the field of pediatric traditional Chinese medicine. It includes nodes of various clinically common pediatric syndromes, such as: wind-cold binding the lungs syndrome, wind-heat invading the lungs syndrome, food stagnation syndrome, spleen and stomach weakness syndrome, heart fire excess syndrome, liver qi stagnation syndrome, and kidney qi deficiency syndrome.

[0047] Each syndrome node is connected to related symptom and sign nodes. The entire network covers 35 key symptom and sign indicators, including fever, chills, cough, cough sound, sputum quality, appetite, sleep, bowel movements, urination, tongue appearance, and pulse. When the diagnostic information integration unit inputs data into the syndrome judgment unit, the syndrome judgment unit uses the input data as a basis to feed into the Bayesian network, triggering probability propagation and updates within the network. The syndrome judgment unit typically outputs the two most likely syndromes and their corresponding calculated probability values, such as a probability of 0.75 for spleen and stomach deficiency syndrome and 0.20 for food stagnation syndrome. Based on these results, trainees can more quickly determine the likelihood of a diagnosis.

[0048] It should be noted that the Bayesian network model is a dynamic learning model. In a single training session, as trainees continuously perform operations and input information, these historical operation and decision data are incorporated into the model as new evidence to adjust the prior probabilities of subsequent evidence judgments, so that the model's reasoning can adapt to the progress of the training process.

[0049] The prescription generation unit is connected to the syndrome judgment unit. It stores a rule library of syndrome-acupoint prescription mapping established by classic Chinese pediatric texts and modern clinical consensus. Generally speaking, the syndrome-acupoint prescription mapping rule library is essentially a lookup table or a set of production rules. It clearly specifies the recommended main acupoints and acupoint combinations when a certain syndrome is diagnosed, as well as the suggested acupuncture techniques for each acupoint, such as tonification, sedation, and even tonification and sedation, and the specific stimulation intensity and needle retention time.

[0050] Once the syndrome differentiation unit outputs the two most likely syndrome types, the prescription generation unit immediately retrieves the corresponding treatment plan from the syndrome-acupoint prescription mapping rule base: for the diagnosed spleen and stomach deficiency syndrome, the rule base may recommend Zusanli, Pishu, and Weishu as the main acupoints, using tonifying acupuncture techniques; for concurrent food stagnation, Zhongwan and Tianshu acupoints may be added, using reducing techniques. The generated recommended acupoint prescriptions and detailed acupuncture technique suggestions are overlaid and displayed in real time in the virtual reality scene, usually appearing in the form of a semi-transparent text box near the side of the trainee's field of vision, providing direct theoretical guidance for the trainee's final operational decision.

[0051] In summary, this embodiment constructs a closed-loop training environment for pediatric acupuncture in traditional Chinese medicine for trainees, encompassing physiological and pathological simulation, behavioral interaction, operation, and clinical thinking training.

[0052] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0053] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A training system for pediatric acupuncture in Traditional Chinese Medicine that combines virtual reality guidance, characterized in that: include: The virtual behavior generation module is used to dynamically generate physiological and behavioral state data of virtual children based on preset pediatric patient characteristics. The feedback module is used to receive status data and generate corresponding visual, auditory, and tactile feedback; The assessment module is used to collect the operational data of trainees and compare and analyze it with the standard operation database to generate assessment results. The guidance module generates diagnostic and treatment guidance information based on the assessment results and the theory of syndrome differentiation in traditional Chinese medicine pediatrics. The virtual behavior generation module includes a physiological state simulation unit and a behavioral response simulation unit. The physiological state simulation unit constructs digital models of the heart, liver, spleen, lungs, and kidneys based on the theory of viscera in traditional Chinese medicine pediatrics. The behavioral response simulation unit constructs a typical behavioral pattern library for children aged 0 to 12 years based on pediatric developmental psychology. The behavioral response simulation unit receives the heart, liver, spleen, lung, and kidney status data output by the physiological state simulation unit, and calculates the virtual child's behavior at the current moment using a decision tree algorithm, taking into account age grouping parameters. The behavior includes at least the range of limb movement, the intensity of crying, and the level of attention.

2. The system according to claim 1, characterized in that, The evaluation module includes at least a data acquisition unit, a feature extraction unit, and an evaluation and analysis unit. The data acquisition unit acquires spatial pose data of acupuncture needles. The feature extraction unit performs sliding window processing on the spatial pose data and extracts the average needle insertion speed, standard deviation of insertion and withdrawal amplitude, and peak twisting angle from each window. The evaluation and analysis unit performs dynamic time warping calculations on the average needle insertion speed, standard deviation of insertion and withdrawal amplitude, and peak twisting angle against a standard database, outputting a quantitative score for the proficiency of the technique, as well as operational error indicators for angle deviation and excessive force.

3. The system according to claim 1, characterized in that, The guidance module includes a diagnostic information integration unit, a syndrome judgment unit, and a prescription generation unit. The diagnostic information integration unit receives simulated symptom data from the virtual behavior generation module and student operation data from the evaluation module. The syndrome judgment unit constructs a syndrome reasoning model based on a Bayesian network. The prescription generation unit stores a syndrome-acupoint prescription mapping rule library established based on classic TCM pediatric texts. Based on the data from the syndrome judgment unit, the prescription generation unit generates recommended acupoint prescriptions and acupuncture techniques, and displays them in real time overlaid in the virtual scene.

4. The system according to claim 3, characterized in that, The network nodes of the syndrome reasoning model include 8 common pediatric syndromes and 35 corresponding key symptoms and signs.

5. The system according to claim 3, characterized in that, When the evidence reasoning model is updated, it incorporates historical data of the operations performed by the trainee in the current training session as a new basis for adjusting the prior probability of subsequent evidence type judgments.

Images