Anesthesia anatomy display and interaction method based on virtuality and reality combination

Abstract

The invention discloses an anesthesia anatomy display and interaction method based on virtuality and reality combination, and relates to the technical field of anesthesia monitoring, and the method comprises the steps: carrying out the registration of a three-dimensional anatomy model and a real-time space coordinate system, marking a high-risk region through a nerve excitement index, and generating a dynamic risk marking graph; according to the dynamic risk marker graph, generating a real-time tactile feedback gain vector through spinal dorsal root gating algorithm and three-dimensional gain field synthesis calculation; and gradient tactile signal output is driven based on the real-time tactile feedback gain vector, impedance data is acquired in real time for characteristic analysis, and when it is detected that impedance characteristics abnormally change, a safe path correction instruction is generated and a nerve contact identifier is triggered. The problem that in the prior art, individual nerve excitation states and dynamic changes cannot be reflected by depending on static and universal anatomical models is effectively solved, and the safety and accuracy of anesthesia operation are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of anesthesia monitoring, and in particular to an anesthesia anatomy display and interaction method based on a combination of virtual and real. Background Art

[0002] With the convergence of medical imaging and virtual reality (VR) / augmented reality (AR) technologies, anesthesia anatomy teaching and clinical assistance are gradually moving towards digitalization, visualization, and intelligentization. In recent years, virtual-reality anatomy methods based on technologies such as 3D reconstruction, depth sensing, and tactile feedback have achieved initial application in education, surgical navigation, and pain management. In particular, during regional anesthesia procedures, precise identification of neural structures, avoidance of high-risk areas, and improved puncture path safety have become research hotspots.

[0003] However, existing technologies still have many limitations. First, most solutions rely on static anatomical models, making it difficult to reflect individualized neural excitation states and dynamic changes in real time, resulting in delayed risk assessment. Second, traditional tactile feedback mechanisms lack a sensitive response to changes in physiological impedance, making it impossible to achieve active risk avoidance and nerve contact identification during the puncture process, affecting operational safety. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an anesthesia anatomy display and interaction method based on the combination of virtual and real to solve the problems of delayed risk assessment and insufficient operational safety in existing anesthesia anatomy display and interaction methods.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides an anesthesia anatomy display and interaction method based on the combination of virtual and real, which includes collecting original charge fluctuation data, combining it with medical image segmentation results, and generating a three-dimensional anatomical model; aligning the three-dimensional anatomical model with the real-time spatial coordinate system, marking high-risk areas through the neural excitability index and generating a dynamic risk marker map; based on the dynamic risk marker map, generating a real-time tactile feedback gain vector through the spinal dorsal root gating algorithm and three-dimensional gain field synthesis calculation; driving the gradient tactile signal output based on the real-time tactile feedback gain vector, and acquiring impedance data in real time for feature analysis, generating a safe path correction instruction and triggering a neural contact marker when an abnormal mutation of the impedance characteristic is detected; activating dopamine concentration estimation based on the neural contact marker, and calculating the synaptic cleft dopamine concentration value in real time through the piezoelectric patch charge signal; generating an anesthesia anatomy safety efficacy report based on the dynamic risk marker map, the safe path correction instruction and the synaptic cleft dopamine concentration value.

[0007] As a preferred solution of the anesthesia anatomy display and interaction method based on the combination of virtual and real described in the present invention, the original charge fluctuation data includes time domain current series, frequency domain impedance spectrum, spatial charge density distribution and time differential signal.

[0008] As a preferred solution of the anesthesia anatomy display and interaction method based on the combination of virtual and real in the present invention, the specific steps of generating a three-dimensional anatomy model are as follows: Capture the original charge fluctuation data, connect it to the signal processor for time-frequency wavelet transform, and generate the neural action potential characteristic spectrum; Based on the characteristic spectrum of neural action potentials, a spatial distribution map of neural excitability is constructed, and decoded through a graph neural network to generate an excitability scalar field matrix; The structured grid data of medical image segmentation and the excitability scalar field matrix are aligned through affine transformation and fused to generate a three-dimensional neural function thermal model; The neural function thermal values ​​are rendered at the mesh vertices of the three-dimensional neural function thermal model. When the excitability exceeds the neural excitation warning threshold, it is marked as a red warning area to generate a three-dimensional anatomical model.

[0009] As a preferred solution of the anesthesia anatomy display and interaction method based on the combination of virtual and real in the present invention, the specific steps of generating a dynamic risk marker map are as follows: Capture the patient's body surface point cloud data and establish a real-time spatial coordinate system, perform registration transformation on the 3D anatomical model based on the real-time spatial coordinate system, and generate a registered 3D anatomical model; The spatial sensitivity adjustment coefficient is calculated based on the registered 3D anatomical model, and the neural excitability index is loaded to generate a binary risk marker matrix. The binary risk marker matrix is ​​subjected to pulsation enhanced fusion rendering to generate a dynamic risk marker map.

[0010] As a preferred solution of the anesthesia anatomy display and interaction method based on the combination of virtual and real according to the present invention, the specific steps of generating the real-time tactile feedback gain vector are as follows: Based on the dynamic risk marker map, the entropy fusion calculation of the dorsal root gating algorithm is performed to generate a multi-scale risk entropy tensor; According to the multi-scale risk entropy tensor, a neural gating signal is generated through the neuron type selection matrix and spatiotemporal integration operations; Three-dimensional gain field synthesis is calculated based on neural gating signals to generate real-time tactile feedback gain vectors.

[0011] As a preferred solution of the anesthesia anatomy display and interaction method based on the combination of virtual and real described in the present invention, wherein: the real-time tactile feedback gain vector drives the gradient tactile signal output, and the impedance data is obtained in real time for feature analysis. When an abnormal mutation of the impedance feature is detected, a safe path correction instruction is generated and a nerve contact mark is triggered. The specific steps are as follows: Drive the gradient tactile signal output based on the real-time tactile feedback gain vector to generate an impedance-weighted gradient tactile field; Impedance data is extracted in real time from the impedance-weighted gradient tactile field, and the critical catastrophe index is calculated through multi-scale fractional differentiation. When the critical mutation index exceeds the safety threshold, neural constrained optimal control is executed, a safe path correction instruction is generated, and a neural contact identifier is obtained.

[0012] As a preferred embodiment of the anesthesia anatomy display and interaction method based on the combination of virtual and real in the present invention, the following specific steps are used to calculate the dopamine concentration value in the synaptic cleft in real time based on the activation of the nerve contact marker and the charge signal of the piezoelectric patch. Based on the neural contact marker, the charge signal of the piezoelectric patch is collected and passed through the quantum gate multiplier to generate a gated charge signal; Perform fractional differential operation on the gated charge signal to obtain a feature enhanced signal; Based on the feature enhancement signal, an adaptive kernel function is constructed and the Mel-frequency cepstral coefficients of the gated charge signal are simultaneously extracted as frequency domain suppression factors. The feature enhancement signal is integrated with the adaptive kernel function in the time domain and fused with the frequency domain inhibition factor to generate the dopamine concentration value in the synaptic cleft.

[0013] As a preferred solution of the anesthesia anatomy display and interaction method based on the combination of virtual and real in the present invention, the specific steps of generating the anesthesia anatomy safety and efficacy report are as follows: A unified spatiotemporal coordinate system for surgical navigation is established based on a dynamic risk marker map, safe path correction instructions, and synaptic cleft dopamine concentration values. The dopamine concentration value in the synaptic cleft is converted into the spatiotemporal coordinate system of surgical navigation, and a quantum coupling field is generated by combining the dynamic risk marker map and the safety path correction instruction to generate a dynamic safety effectiveness index. Based on the dynamic safety effectiveness index, high-risk areas are marked on the augmented reality interface. When the dynamic safety effectiveness index exceeds the neurostability threshold, blockchain evidence storage is activated to generate an anesthesia anatomy safety effectiveness report.

[0014] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the anesthesia anatomy display and interaction method based on the combination of virtual and real as described in the first aspect of the present invention is implemented.

[0015] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the anesthesia anatomy display and interaction method based on the combination of virtual and real as described in the first aspect of the present invention.

[0016] The beneficial effects of the present invention are as follows: by generating a three-dimensional anatomical model and combining it with the results of medical image segmentation, the problem in the existing technology that relies on static and universal anatomical models and cannot reflect the individual nerve excitation state and dynamic changes is effectively solved, thereby improving the safety and accuracy of anesthesia operations; further, by combining the three-dimensional gain field synthesis calculation to output a real-time tactile feedback gain vector with spatial gradient characteristics, closed-loop control from nerve contact perception to tactile feedback output is realized, thereby enhancing the user's perception accuracy of puncture depth, direction and tissue type, and improving the realism and controllability of the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 This is a flowchart of the anesthesia anatomy display and interaction method based on the combination of virtual and real.

[0019] Figure 2 Flowchart for generating 3D anatomical models.

[0020] Figure 3 Flowchart for generating a dynamic risk marker map.

[0021] Figure 4 Flowchart for generating real-time haptic feedback gain vectors. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides an anesthesia anatomy display and interaction method based on a combination of virtual and real, including the following steps: S1: Collect raw charge fluctuation data and combine it with medical image segmentation results to generate a three-dimensional anatomical model.

[0026] S1.1: The raw charge fluctuation data include time domain current series, frequency domain impedance spectrum, spatial charge density distribution and time differential signal.

[0027] It should be noted that the time domain current series is collected by a microelectrode array, reflecting the instantaneous motion characteristics of the charge carriers; the frequency domain impedance spectrum is measured by a phase-locked amplifier, characterizing the changing law of dielectric properties at different frequencies; the spatial charge density distribution is obtained by a scanning probe microscope, showing the aggregation state of charge in the three-dimensional structure; the time differential signal is recorded by a high-speed data acquisition card to capture the instantaneous rate of charge change.

[0028] S1.2: Capture the raw charge fluctuation data, connect it to the signal processor for time-frequency wavelet transform, and generate the neural action potential characteristic spectrum.

[0029] The specific process includes: the zinc oxide nanowire piezoelectric patch captures the charge fluctuation data and transmits it to the signal processor. The signal processor uses the Morlet wavelet basis function to perform continuous wavelet transform on the time domain signal to decompose and obtain wavelet coefficients of different scales; at the same time, the signal is converted to the frequency domain through fast Fourier transform, and the time-frequency joint distribution matrix is ​​constructed based on the time-frequency analysis results; the energy distribution characteristics and instantaneous frequency characteristics of the signal are extracted based on the wavelet coefficients and the time-frequency joint distribution matrix, and after feature fusion, a neural action potential characteristic spectrum containing three-dimensional information of time-frequency-energy is generated. Each data point in the neural action potential characteristic spectrum corresponds to the signal energy intensity at a specific frequency at a specific time, which fully characterizes the time-frequency characteristics of neural electrical activity.

[0030] S1.3: Construct a spatial distribution map of neural excitability based on the characteristic spectrum of neural action potentials, and generate an excitability scalar field matrix through decoding through a graph neural network.

[0031] The specific process includes: the characteristic spectrum of neural action potentials is converted into three-dimensional voxel data after spatial interpolation processing, and each three-dimensional voxel node stores the time-frequency energy characteristics of the corresponding spatial position; the graph neural network uses three-dimensional voxel nodes as vertices and the spatial distance between adjacent nodes as edge weights to construct a spatial distribution map of neural excitability; through the graph convolution layer, neighborhood features are aggregated layer by layer to learn spatial dependencies; the final output layer maps the multi-dimensional features of each node to a single-value excitation, forming an excitation scalar field matrix covering the target area; the row and column indices of the excitation scalar field matrix correspond to spatial coordinates, and the matrix element value represents the neural excitation intensity at that position, realizing the quantitative conversion from spectral features to spatial distribution.

[0032] S1.4: Align the structured grid data of medical image segmentation with the excitability scalar field matrix through affine transformation and fuse them to generate a three-dimensional neural function thermal model.

[0033] The specific process includes: aligning the structured grid data of medical image segmentation with the excitability scalar field matrix through the spatial coordinate transformation matrix, analyzing the mapping relationship between the structured grid vertex coordinates and the excitability scalar field matrix index; using the bilinear interpolation method to distribute the excitability value to each facet of the structured grid; the aligned structured grid vertices retain both anatomical structure information and neural excitability values; converting the excitability values ​​into thermal color scales through the color mapping function, and rendering on the structured grid surface to generate a three-dimensional neural function thermal model; the geometric shape of the three-dimensional neural function thermal model is determined by the structured grid, and the color distribution reflects the degree of neural excitability in different areas, realizing the visual fusion of anatomical structure and functional activity.

[0034] The structured grid data for medical image segmentation is generated by three-dimensional segmentation and gridding of medical images such as CT and MRI, and is derived from the original image data in DICOM format in the hospital's picture archiving and communication system (PACS).

[0035] S1.4: Render the neural function thermal values ​​at the mesh vertices of the three-dimensional neural function thermal model. When the excitability exceeds the neural excitation warning threshold, mark it as a red warning area and generate a three-dimensional anatomical model.

[0036] The specific process includes: storing neural function thermal values ​​in the mesh vertices of the three-dimensional neural function thermal model, which reflect the discharge intensity of the neuronal cluster; traversing all mesh vertices and comparing the size relationship between the neural function thermal values ​​and the neural excitation warning threshold; when the neural function thermal value exceeds the neural excitation warning threshold, setting the vertex color attribute to red; for vertices below the neural excitation warning threshold, assigning colors within the blue-green spectrum according to the linear interpolation algorithm; after completing the color assignment of all vertices, reconstructing the surface topology based on the structured grid of medical image segmentation to generate a three-dimensional anatomical model that integrates functional information; the red area in the three-dimensional anatomical model represents the warning area of ​​excessive neuronal excitation, and the blue-green gradient area reflects the normal neural activity gradient.

[0037] The neural excitation alert threshold was determined by statistically analyzing the differences in EEG activity characteristics between healthy subjects and epilepsy patients, and was specifically set to the 95th percentile of the maximum neural function thermal value of the healthy population.

[0038] S2: The 3D anatomical model is registered with the real-time spatial coordinate system, high-risk areas are marked using the neural excitability index, and a dynamic risk marking map is generated.

[0039] S2.1: Capture the patient's body surface point cloud data and establish a real-time spatial coordinate system. Perform registration transformation on the three-dimensional anatomical model based on the real-time spatial coordinate system to generate a registered three-dimensional anatomical model.

[0040] The specific process includes: after the depth sensor collects the patient's body surface point cloud data, the spatial correspondence between the point cloud and the surface feature points of the three-dimensional anatomical model is obtained through the iterative nearest point algorithm; the real-time space coordinate system transformation parameters including the translation vector and the rotation matrix are established; the vertex coordinates of the three-dimensional anatomical model are multiplied by the transformation matrix to complete the spatial mapping; the vertex coordinates of the three-dimensional anatomical model after alignment are kept synchronously updated with the real-time space coordinate system; the alignment error is evaluated through the root mean square of the feature point distance to ensure that the three-dimensional anatomical model is consistent with the patient's actual anatomical position; the final output of the aligned three-dimensional anatomical model accurately reflects the patient's current anatomical structure in the real-time space coordinate system.

[0041] S2.2: Calculate the spatial sensitivity adjustment coefficient based on the registered 3D anatomical model and load the neural excitability index to generate a binary risk marker matrix, which is expressed as: ; in, represents the spatial sensitivity adjustment coefficient ( ∈[0,+∞]), represents the natural exponential function, Represents the Euclidean distance from the current position to the nearest nerve root, It represents the effective conduction radius of the neural electrical signal. represents the linear rectification function, represents the neural excitability index ( ∈[0, 1]), represents the time variable, represents the standard deviation, Indicates high frequency band The standard deviation of the wavelet energy is Indicates the basic frequency band The standard deviation of the wavelet energy.

[0042] The specific process involves adjusting the mapping weight of the neural excitability index using the spatially sensitive adjustment coefficient on the registered 3D anatomical model. The spatially sensitive adjustment coefficient is determined by both anatomical structural features and functional zoning. The neural excitability index and the spatially sensitive adjustment coefficient are multiplied point by point, and then threshold segmentation is performed to generate a binary risk marker matrix. In this binary risk marker matrix, high-risk areas are marked as 1, and low-risk areas are marked as 0. The generation process of the binary risk marker matrix ensures that high-risk areas precisely correspond to the locations of key neural structures in the anatomical model, while low-risk areas maintain a background value of zero. The final output binary risk marker matrix is ​​used to generate the subsequent dynamic risk marker map, which is enhanced using a temporal convolution kernel and synchronized with physiological rhythm signals.

[0043] S2.3: Perform pulsation enhancement fusion rendering on the binary risk marker matrix to generate a dynamic risk marker map.

[0044] The specific process includes: the binary risk marker matrix is ​​subjected to pulsation enhancement processing through the time convolution kernel, and the convolution kernel weight decays along the time axis according to the Gaussian distribution; the continuous numerical value generated after time convolution and physiological rhythm signal modulation is point multiplied with the physiological rhythm signal collected in real time to enhance the temporal correlation of the risk marker; the discrete matrix value is mapped to a continuous three-dimensional anatomical surface using a bilinear interpolation algorithm; the high-risk area is rendered in a red semi-transparent volume in three-dimensional space, and the rendering transparency changes dynamically with the pulsation enhancement intensity; the final dynamic risk marker map is synchronously displayed in the augmented reality field of view, and the high-risk area shows periodic changes in light and dark, and the frequency of change is consistent with the patient's physiological rhythm.

[0045] S3: Based on the dynamic risk marker map, a real-time tactile feedback gain vector is generated through the spinal dorsal root gating algorithm and three-dimensional gain field synthesis calculation.

[0046] S3.1: Perform entropy fusion calculation of the dorsal root gating algorithm based on the dynamic risk labeling map to generate a multi-scale risk entropy tensor, which is expressed as: ; in, Indicates the spatial coordinates ( ),depth ,time The multi-scale risk entropy tensor of represents the wavelet scale index, Indicates the The adaptive weight coefficients of wavelet scales ( ∈[0.12, 0.28]), represents the Morlet wavelet entropy function, Indicates the The wavelet transform operator of wavelet scale, represents the dynamic risk labeling map matrix, represents the curvature term weight factor, represents the surface Laplacian operator, represents the surface parameterization function in the Frenet-Serret framework, represents the curvature sensitivity adjustment coefficient ( ∈[0.1, 0.3]), represents the local maximum curvature value, represents the domain range for local integration on the 3D anatomical model, represents the tangential coordinate axis along the principal curvature of the anatomical surface in the curvilinear coordinate system, Represents the vertical coordinate axis along the normal direction of the anatomical surface in the curvilinear coordinate system, represents the risk gradient direction angle.

[0047] The specific process includes establishing a curvilinear coordinate system under the Frenet-Serret framework on the surface of the three-dimensional anatomical model; performing a multi-scale Morlet wavelet transform on the dynamic risk marker matrix and calculating the wavelet transform operator output of each wavelet scale; inputting the wavelet transform operator output into the Morlet wavelet entropy function to calculate the uncertainty measure in the time-frequency domain; simultaneously calculating the surface Laplace operator response of the three-dimensional anatomical model surface, and constructing the curvature term weight factor by combining the curvature sensitivity adjustment coefficient and the local maximum curvature value; integrating the risk gradient direction angle along the tangential coordinate axis and the perpendicular coordinate axis in the curvilinear coordinate system to determine the local integral neighborhood range; weightedly summing the adaptive weight coefficient of each wavelet scale with the wavelet entropy value of the corresponding scale, and then superimposing the curvature constraint term to finally generate a multi-scale risk entropy value tensor containing spatial coordinates, depth, and time dimensions; each element value of the multi-scale risk entropy value tensor reflects the degree of risk uncertainty of the corresponding anatomical position at a specific spatiotemporal scale.

[0048] S3.2: Generate neural gating signals based on the multi-scale risk entropy tensor through neuron type selection matrix and spatiotemporal integration operations.

[0049] The specific process includes: after the multi-scale risk entropy value tensor is input into the dual-channel pulse control processor, the exponentially weighted moving average algorithm is used to integrate along the time dimension in the excitatory pathway, and the Gaussian difference filter is used to integrate along the spatial dimension in the inhibitory pathway; the neuron type selection matrix linearly combines the dual-channel outputs through preset excitatory neuron weights and inhibitory neuron weights; the spatiotemporal integration operation convolves the combined signal in the four-dimensional spatiotemporal domain, and the convolution kernel is determined by the risk gradient direction angle and the local maximum curvature value; the final generated neural gating signal contains a time pulse sequence and a spatial modulation coefficient, the pulse frequency encodes risk intensity information, and the spatial coefficient reflects the anatomical position specificity.

[0050] The preset excitatory neuron weights are determined by analyzing the differences in the neural electrical activity characteristics of healthy subjects and patients with pathological conditions, and are specifically set to the statistical mean of the energy proportion of the high-frequency components of the neural electrical signals of healthy people.

[0051] S3.3: Perform three-dimensional gain field synthesis calculation based on neural gating signals to generate real-time tactile feedback gain vectors, expressed as: ; in, represents the real-time tactile feedback gain vector, represents the amplitude gain, represents the frequency gain, represents the phase gain, Represents neural gating signal ( ∈[0, 1]), Indicates neural gating signal Perform a hyperbolic sine transform, Represents a dynamic risk labeling matrix exist The spatial gradient of the direction, Represents a dynamic risk labeling matrix exist The spatial gradient of the direction, Represents the vertical coordinate axis of the three-dimensional anatomical model, Represents the horizontal coordinate axis of the 3D anatomical model.

[0052] The specific process includes: after the neural gating signal is processed by hyperbolic sine transform, a two-dimensional projection coordinate system is established in the plane formed by the vertical and horizontal coordinate axes of the three-dimensional anatomical model; the spatial gradient of the dynamic risk marker in the direction and the spatial gradient in the direction are obtained by calculating the Sobel operator, which respectively characterize the rate of change of the risk distribution in the vertical and horizontal directions; the amplitude gain is obtained by normalizing the dot product operation result of the transformed neural gating signal and the vertical spatial gradient; the frequency gain is output by the convolution operation of the horizontal spatial gradient and the neural gating signal; the phase gain takes the vector cross product modulus of the vertical spatial gradient and the horizontal spatial gradient; the final generated real-time tactile feedback gain vector contains three orthogonal components, and the component values ​​correspond to the calculation results of the amplitude gain, frequency gain and phase gain, respectively, and the vector direction is determined by the vertical and horizontal coordinate axes of the three-dimensional anatomical model.

[0053] S4: Based on the real-time tactile feedback gain vector, the gradient tactile signal output is driven, and the impedance data is acquired in real time for feature analysis. When an abnormal mutation in the impedance feature is detected, a safe path correction instruction is generated and the nerve contact mark is triggered.

[0054] S4.1: Drive the gradient tactile signal output based on the real-time tactile feedback gain vector to generate an impedance-weighted gradient tactile field.

[0055] The specific process includes: the real-time tactile feedback gain vector adjusts the amplitude and direction of the gradient tactile signal so that the gradient tactile field can change dynamically according to the impedance distribution. The gradient tactile signal output is jointly determined by the tactile feedback gain vector and the impedance weighting coefficient. The impedance weighting coefficient comes from the quantitative result of the critical mutation index on the degree of impedance abnormality. The gradient tactile field presents a continuously changing tactile stimulation intensity in space. The tactile stimulation intensity is proportional to the local impedance change rate, ensuring that areas with high impedance changes receive stronger tactile feedback. The impedance-weighted gradient tactile field ultimately forms a tactile feedback pattern with spatial resolution. The tactile feedback pattern matches the spatial distribution of the risk of tissue state mutation, and the tactile feedback intensity is higher in high-risk areas.

[0056] S4.2: Extract impedance data from the impedance-weighted gradient tactile field in real time and calculate the critical catastrophe index through multi-scale fractional differentiation. The expression is: ; in, represents the critical mutation index, represents the Caputo-type fractional differential operator, represents the order of the Caputo fractional differential operator, Represents real-time impedance time series data, represents the integration variable, represents the frequency attenuation factor ( ∈(0, 1]), represents a frequency variable, represents the frequency response function, represents the spatiotemporal coupling coefficient ( ∈(0, 1]), Indicates the interval between two adjacent times. represents the impedance space gradient operator, Indicates impedance.

[0057] The specific process includes: the impedance data collected in real time in the gradient tactile field is obtained by measuring with a four-electrode method to form real-time impedance time series data that changes with time; a Caputo-type fractional-order differential operator is applied to the real-time impedance time series data, and the differential order is dynamically adjusted by the frequency attenuation factor; the frequency response function is extracted from the real-time impedance time series data through fast Fourier transform, reflecting the frequency domain characteristics of the impedance signal; the spatiotemporal coupling coefficient controls the weighted ratio of the time differential result and the spatial gradient operator. The spatial gradient operator calculates the difference in impedance values ​​between adjacent measurement points and divides it by the Euclidean distance between the two points to obtain the normalized impedance change rate; the fractional-order differential result is convolved with the frequency response function and then multiplied by the spatiotemporal coupling coefficient and the output of the impedance spatial gradient operator; the critical mutation index finally generated quantifies the degree of abnormal change in impedance in the spatiotemporal dimension. A larger critical mutation index indicates a higher risk of tissue state mutation.

[0058] S4.3: When the critical mutation index exceeds the safety threshold, neural constrained optimal control is executed, a safe path correction instruction is generated, and a neural contact identifier is obtained.

[0059] The specific process includes: when the critical mutation index exceeds the safety threshold, the neural constraint optimal control algorithm is activated to analyze the minimum distance between the current surgical path and the boundary of the safe area; the optimal path correction amount is obtained based on the Hamilton-Jacobi-Bellman equation, and a safe path correction instruction containing a three-dimensional displacement vector is generated; at the same time, a quantum random number generator is used to generate a pulse sequence that obeys the Poisson distribution, and the pulse frequency is proportional to the critical mutation index; the pulse sequence is input into a bistable trigger to generate a quantized neural contact identifier, and the piezoelectric signal acquisition channel is activated when the identifier state is 1; the safe path correction instruction is mapped to the surgical navigation coordinate system through the spatial coordinate transformation matrix, and the instrument motion trajectory is updated in real time; after the quantized neural contact identifier is triggered, the dopamine concentration detection process of the corresponding anatomical position is automatically associated.

[0060] The safety threshold was determined by statistically analyzing the difference in the critical mutation index distribution between healthy tissue and diseased tissue, and was specifically set to 1.5 times the 99% quantile of the critical mutation index of the healthy group.

[0061] S5: The dopamine concentration in the synaptic cleft is calculated in real time based on the activation dopamine concentration of the nerve contact marker and the charge signal of the piezoelectric patch.

[0062] S5.1: Based on the neural contact marker, the piezoelectric patch charge signal is collected and passed through the quantum gate multiplier to generate a gated charge signal.

[0063] The specific process includes: when the neural contact marker is activated, the zinc oxide nanowire piezoelectric patch captures the charge fluctuation signal; the quantum gate multiplier receives the neural contact marker as a gate enable signal, and allows the piezoelectric charge signal to pass when the marker is logic 1; the gated charge signal is generated by performing a dot product operation on the piezoelectric charge signal and the quantum state probability amplitude of the neural contact marker; during the operation, the quantum tunneling effect is used to adjust the signal gain to ensure the effective extraction of weak charge signals; the final output gated charge signal only retains the piezoelectric response data during the activation period of the neural contact marker, and the signal during the non-activation period is forced to zero.

[0064] S5.2: Perform a fractional differentiation operation on the gated charge signal to obtain a feature enhanced signal.

[0065] The specific process includes: the gated charge signal is input into the fractional-order differential operation unit, and the Caputo definition is used to analyze the 0.25-order differential; the differential kernel function uses the power-law attenuation form of the normalized Gamma function to ensure the short-term memory effect; the preprocessed signal is obtained by convolving the gated charge signal with the differential kernel function to retain the non-local characteristics of the gated charge signal; high-frequency noise is automatically suppressed during the operation, and the signal mutation characteristics are enhanced; the time domain resolution of the final output preprocessed signal is consistent with the original gated charge signal, and the amplitude envelope reflects the fractional-order derivative characteristics of the charge change.

[0066] S5.3: Based on the feature enhancement signal, an adaptive kernel function is constructed and the Mel-frequency cepstral coefficients of the gated charge signal are simultaneously extracted as frequency domain suppression factors.

[0067] The specific process includes: extracting the instantaneous amplitude envelope of the preprocessed signal through Hilbert transform, which is used as the time-varying parameter of the kernel function; the adaptive kernel function adopts the modified Mittag-Leffler function form, and the exponential parameter is dynamically adjusted by the fractional differential result of the preprocessed signal; at the same time, the gated charge signal is converted to the bionic auditory frequency domain through a Mel-scale filter bank, and a discrete cosine transform is performed after analyzing the logarithm of the energy of each subband; the first 12 Mel-frequency cepstral coefficients are taken as the frequency domain suppression factor, and the factor weight is inversely proportional to the attenuation rate of the kernel function; the adaptive kernel function finally constructed has the characteristics of time-frequency dual variable adjustment, and the peak position of the kernel function is automatically aligned with the energy distribution of the frequency domain suppression factor.

[0068] S5.4: The feature enhancement signal is integrated with the adaptive kernel function in the time domain and fused with the frequency domain inhibition factor to generate the synaptic cleft dopamine concentration value, which is expressed as: ; in, Indicates time The dopamine concentration in the synaptic cleft is represents the dopamine conversion gain coefficient, represents a very small constant, Indicates time The dynamic suppression factor ( ∈(0, 1]), represents the decay exponent of the kernel function ( ∈(0,1)), Represents the memory decay index ( ∈(0,1)), represents the two-parameter Mittag-Leffler function, Indicates the integral variable of The order of the differential operator, Indicates nerve contact Modulated gated charge signal, Indicates the integral variable Dynamic quantification of dopamine concentration in the synaptic cleft.

[0069] The specific process includes performing convolution integral operations on the preprocessed signal and the kernel function in the time domain, and processing the integral variable using a fractional-order differential operator; calculating the dynamic inhibition factor through the energy distribution of the Mel-frequency cepstral coefficients to adjust the decay rate of the kernel function; using the two-parameter Mittag-Leffler function as the integral kernel function, and determining the two exponential parameters of the two-parameter Mittag-Leffler function by the memory decay exponent and the kernel function decay exponent; performing a Hadamard product operation on the gated charge signal modulated by the neural contact representation after fractional differentiation and the dynamic inhibition factor; inputting the operation result into the integral expression to generate the synaptic cleft dopamine concentration value under the control of the dopamine conversion gain coefficient; using a very small constant to ensure that the denominator is not zero, thus ensuring the numerical stability of the calculation process; the final output synaptic cleft dopamine concentration value reflects the intensity of neurotransmitter release in real time, and the numerical value is nonlinearly positively correlated with the intensity of the gated charge signal.

[0070] S6: Generate an anesthetic anatomy safety efficacy report based on the dynamic risk marker map, safety path correction instructions and synaptic cleft dopamine concentration value.

[0071] S6.1: Establish a unified spatiotemporal coordinate system for surgical navigation based on the dynamic risk marker map, safe path correction instructions, and synaptic cleft dopamine concentration values.

[0072] The specific process includes: mapping the spatial coordinates of the dynamic risk marker map to the surgical navigation coordinate system through the rigid body transformation matrix; converting the three-dimensional displacement vector contained in the safe path correction instruction into homogeneous coordinate form and superimposing it with the spatial transformation result of the dynamic risk marker map; aligning the dopamine concentration value in the synaptic cleft to the corresponding spatial coordinate point through the timestamp to establish the association between neurochemical signals and anatomical positions; using the quaternion interpolation algorithm to unify the time sampling intervals of different data sources to eliminate timing jitter; the final generated surgical navigation spatiotemporal coordinate system uses the grid vertices of the dynamic risk marker map as the spatial reference and the dopamine concentration sampling time as the time reference, and the safe path correction instruction is used as the coordinate transformation parameter to update the spatial mapping relationship in real time to achieve spatiotemporal synchronization of the three types of data.

[0073] S6.2: The dopamine concentration value in the synaptic cleft is converted to the spatiotemporal coordinate system of surgical navigation, and a quantum coupling field is generated by combining the dynamic risk marker map and the safety path correction instructions to generate a dynamic safety effectiveness index.

[0074] The specific process includes: mapping the dopamine concentration value in the synaptic cleft to the grid vertices of the dynamic risk marker map through a spatial interpolation algorithm to form a neurotransmitter concentration field; performing a tensor product operation on the binary matrix of the dynamic risk marker map and the displacement vector of the safety path correction instruction to generate a risk gradient field; constructing a quantum coupling field by coupling the neurotransmitter concentration field and the risk gradient field through the Schrödinger equation, and the Hamiltonian operator contains the concentration field potential function and the risk gradient field kinetic energy term; obtaining the dynamic safety effectiveness index by solving the eigenvalue problem of the quantum coupling field, and the eigenstate probability amplitude reflects the safety level of the spatial position; the final output dynamic safety effectiveness index is updated in real time in the space-time coordinate system, and the numerical range is normalized to 0 to 1, where 1 represents absolute safety and 0 represents high risk.

[0075] S6.3: High-risk areas are marked on the augmented reality interface based on the dynamic safety effectiveness index. When the dynamic safety effectiveness index exceeds the neurostability threshold, blockchain evidence storage is initiated to generate an anesthesia anatomy safety effectiveness report.

[0076] The specific process includes converting the dynamic safety efficacy index into a heat map through a color mapping algorithm, superimposing it on the surface of the three-dimensional anatomical model in the augmented reality interface, and marking high-risk areas with pulsating red contour lines; the neural stability threshold is determined by the gamma-band energy distribution of the resting-state EEG of healthy subjects, and the smart contract is triggered when the dynamic safety efficacy index is lower than the neural stability threshold; the blockchain evidence storage uses the SHA-256 algorithm to hash the quantum coupling field parameters, safety efficacy index distribution map and anesthesia depth monitoring data at the current moment, and writes them into the distributed ledger; the anesthesia anatomy safety efficacy report automatically integrates the spatial coordinates of the dynamic risk marker map, the synaptic cleft dopamine concentration time series curve and the historical record of the safety path correction instructions to generate a PDF document containing a three-dimensional risk topology map and a neural stability assessment curve; the key data fields in the anesthesia anatomy safety efficacy dynamic assessment report are verified for authenticity through zero-knowledge proof technology, and then digital signatures are attached and uploaded to the medical blockchain network.

[0077] The neural stability threshold was set by statistically analyzing the upper limit of the 95% confidence interval of the energy distribution of the resting-state EEG band (30-80 Hz) of healthy subjects, and 1.8 times the mean energy of the band of the healthy group was taken as the critical value.

[0078] The synaptic cleft dopamine concentration time series curve is a dynamic concentration change curve generated by real-time monitoring of the chemical signals in the synaptic cleft through microdialysis sampling or voltammetry, and after time series filtering and calibration.

[0079] This embodiment also provides a computer device, which is suitable for the anesthesia anatomy display and interaction method based on the combination of virtual and real, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the anesthesia anatomy display and interaction method based on the combination of virtual and real proposed in the above embodiment.

[0080] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0081] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the anesthesia anatomy display and interaction method based on the combination of virtual and real as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0082] In summary, the present invention effectively solves the problem in the existing technology that static and universal anatomical models cannot reflect the individual nerve excitation state and dynamic changes by generating a three-dimensional anatomical model and combining it with the results of medical image segmentation, thereby improving the safety and accuracy of anesthesia operations; further, by combining the three-dimensional gain field synthesis calculation to output a real-time tactile feedback gain vector with spatial gradient characteristics, closed-loop control from nerve contact perception to tactile feedback output is achieved, thereby enhancing the user's perception accuracy of puncture depth, direction and tissue type, and improving the realism and controllability of the operation.

[0083] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for displaying and interacting with anesthesia anatomy based on a combination of virtual and real, characterized by: include, Collect raw charge fluctuation data and combine it with medical image segmentation results to generate a three-dimensional anatomical model; The 3D anatomical model is registered with the real-time spatial coordinate system, high-risk areas are marked using the neural excitability index, and a dynamic risk marker map is generated; According to the dynamic risk marker map, a real-time tactile feedback gain vector is generated through the spinal cord dorsal root gating algorithm and three-dimensional gain field synthesis calculation; Based on the real-time tactile feedback gain vector, the gradient tactile signal output is driven, and the impedance data is acquired in real time for feature analysis. When an abnormal mutation in the impedance feature is detected, a safe path correction instruction is generated and a nerve contact mark is triggered. The dopamine concentration in the synaptic cleft is calculated in real time based on the activation dopamine concentration of the nerve contact marker and the charge signal of the piezoelectric patch; Generate an anesthetic anatomy safety efficacy report based on the dynamic risk marker map, safety path correction instructions and synaptic cleft dopamine concentration value.

2. The anesthesia anatomy display and interaction method based on virtual-real integration according to claim 1, characterized in that: The raw charge fluctuation data includes time domain current series, frequency domain impedance spectrum, spatial charge density distribution and time differential signal.

3. The anesthesia anatomy display and interaction method based on virtual-real integration according to claim 1, characterized in that: The specific steps of generating a three-dimensional anatomical model are as follows: Capture the original charge fluctuation data, connect it to the signal processor for time-frequency wavelet transform, and generate the neural action potential characteristic spectrum; Based on the characteristic spectrum of neural action potentials, a spatial distribution map of neural excitability is constructed, and decoded through a graph neural network to generate an excitability scalar field matrix; The structured grid data of medical image segmentation and the excitability scalar field matrix are aligned through affine transformation and fused to generate a three-dimensional neural function thermal model; The neural function thermal values ​​are rendered at the mesh vertices of the three-dimensional neural function thermal model. When the excitability exceeds the neural excitation warning threshold, it is marked as a red warning area to generate a three-dimensional anatomical model.

4. The anesthesia anatomy display and interaction method based on virtual-real integration according to claim 1, characterized in that: The specific steps of generating a dynamic risk marker map are as follows: Capture the patient's body surface point cloud data and establish a real-time spatial coordinate system, perform registration transformation on the 3D anatomical model based on the real-time spatial coordinate system, and generate a registered 3D anatomical model; The spatial sensitivity adjustment coefficient is calculated based on the registered 3D anatomical model, and the neural excitability index is loaded to generate a binary risk marker matrix. The binary risk marker matrix is ​​subjected to pulsation enhanced fusion rendering to generate a dynamic risk marker map.

5. The anesthesia anatomy display and interaction method based on the combination of virtual and real according to claim 1, characterized in that: The specific steps of generating the real-time tactile feedback gain vector are as follows: Based on the dynamic risk marker map, the entropy fusion calculation of the dorsal root gating algorithm is performed to generate a multi-scale risk entropy tensor; According to the multi-scale risk entropy tensor, a neural gating signal is generated through the neuron type selection matrix and spatiotemporal integration operations; Three-dimensional gain field synthesis is calculated based on neural gating signals to generate real-time tactile feedback gain vectors.

6. The anesthesia anatomy display and interaction method based on the combination of virtual and real according to claim 5, characterized in that: The method drives the gradient tactile signal output based on the real-time tactile feedback gain vector, obtains the impedance data in real time for feature analysis, generates a safe path correction instruction and triggers the nerve contact mark when an abnormal mutation of the impedance feature is detected, and the specific steps are as follows: Drive the gradient tactile signal output based on the real-time tactile feedback gain vector to generate an impedance-weighted gradient tactile field; Impedance data is extracted in real time from the impedance-weighted gradient tactile field, and the critical catastrophe index is calculated through multi-scale fractional differentiation. When the critical mutation index exceeds the safety threshold, neural constrained optimal control is executed, a safe path correction instruction is generated, and a neural contact identifier is obtained.

7. The anesthesia anatomy display and interaction method based on virtual-real integration according to claim 6, characterized in that: The method of calculating the dopamine concentration based on the activation of the nerve contact marker and calculating the synaptic cleft dopamine concentration value in real time through the piezoelectric patch charge signal is as follows: Based on the neural contact marker, the charge signal of the piezoelectric patch is collected and passed through the quantum gate multiplier to generate a gated charge signal; Perform fractional differential operation on the gated charge signal to obtain a feature enhanced signal; Based on the feature enhancement signal, an adaptive kernel function is constructed and the Mel-frequency cepstral coefficients of the gated charge signal are simultaneously extracted as frequency domain suppression factors; The feature enhancement signal is integrated with the adaptive kernel function in the time domain and fused with the frequency domain inhibition factor to generate the dopamine concentration value in the synaptic cleft.

8. The anesthesia anatomy display and interaction method based on virtual-real integration according to claim 1, characterized in that: The specific steps for generating the anesthesia anatomy safety efficacy report are as follows: A unified spatiotemporal coordinate system for surgical navigation is established based on a dynamic risk marker map, safe path correction instructions, and synaptic cleft dopamine concentration values. The dopamine concentration value in the synaptic cleft is converted into the spatiotemporal coordinate system of surgical navigation, and a quantum coupling field is generated by combining the dynamic risk marker map and the safety path correction instruction to generate a dynamic safety effectiveness index. Based on the dynamic safety effectiveness index, high-risk areas are marked on the augmented reality interface. When the dynamic safety effectiveness index exceeds the neurostability threshold, blockchain evidence storage is activated to generate an anesthesia anatomy safety effectiveness report.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the anesthesia anatomy display and interaction method based on the combination of virtual and real are implemented as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the anesthesia anatomy display and interaction method based on the combination of virtual and real according to any one of claims 1 to 8 are implemented.

Images