Intelligent acupoint positioning method and system based on multi-modal imaging

Abstract

The invention relates to the technical field of medical positioning, and discloses an intelligent acupoint positioning method and system based on multi-modal imaging, and the method comprises the steps: carrying out the dual-modal feature decoupling of infrared thermal imaging data and ultrasonic elastography data of a target patient, and obtaining a feature waveform set; matching the characteristic waveform set with a corresponding acupuncture point initial position mark point set in a preset acupuncture point waveform template library; projecting the acupuncture point initial position mark point set to a body surface area corresponding to the target patient, and performing distortion correction on acupuncture point projection to obtain a high-precision acupuncture point coordinate set; mapping the high-precision acupoint coordinate set to a three-dimensional space to obtain a three-dimensional acupoint model; performing real-time parameter registration on the three-dimensional acupuncture point model based on the real-time physiological state of the target patient to obtain a standard space parameter set of the three-dimensional acupuncture point model; and superposing the standard space parameter set to the three-dimensional acupoint model to obtain a visual acupoint view. According to the invention, the accuracy of intelligent acupoint positioning of multi-modal imaging can be improved.

Description

Technical Field

[0001] This invention relates to the field of medical positioning technology, and in particular to an intelligent acupoint positioning method and system based on multimodal imaging. Background Technology

[0002] Existing technologies for intelligent acupoint positioning lack the ability to deeply integrate multimodal imaging data processing, failing to effectively decouple and fuse infrared thermal imaging and ultrasonic elastography data. This results in incomplete extraction of frequency domain energy distribution and spatial gradient features related to acupoints, making it impossible to accurately construct feature waveforms that reflect the essential attributes of acupoints, thus leading to insufficient reliability of the initial positioning data.

[0003] Meanwhile, existing technologies have significant deficiencies in the projection conversion and dynamic adaptation of acupoint locations. They have not established an effective mechanism for correcting surface distortion, nor do they have a three-dimensional modeling and parameter registration method that combines human anatomical rules with real-time physiological states. As a result, the positioning results are easily affected by individual differences, surface morphology, and physiological fluctuations. Not only is the three-dimensional model poorly aligned with the actual anatomical structure, but the spatial accuracy of the visualization view is also difficult to guarantee, which fails to meet the requirements for high-precision and dynamic acupoint positioning. Summary of the Invention

[0004] This invention provides a method and system for intelligent acupoint positioning based on multimodal imaging to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an intelligent acupoint localization method based on multimodal imaging, comprising: S1. Perform dual-modal feature decoupling on the infrared thermal imaging data and ultrasound elastography data of the target patient to obtain the characteristic waveform set of the target patient; S2. Match the feature waveform set with the corresponding acupoint initial position marker point set in the preset acupoint waveform template library; S3. Project the initial position marker set of the acupoints onto the body surface area corresponding to the target patient, and perform distortion correction on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient; S4. Map the high-precision acupoint coordinate set to three-dimensional space to obtain the three-dimensional acupoint model of the target patient; S5. Based on the real-time physiological state of the target patient, perform real-time parameter registration on the three-dimensional acupoint model to obtain the standard spatial parameter set of the three-dimensional acupoint model; S6. The standard spatial parameter set is superimposed onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient.

[0006] In a preferred embodiment, the dual-modal feature decoupling of the infrared thermal imaging data and ultrasound elastography data of the target patient to obtain the characteristic waveform set of the target patient includes: Wavelet transform decomposition is performed on the infrared thermal imaging data to obtain the frequency domain energy distribution feature set of the infrared thermal imaging data; Time-frequency joint analysis of ultrasonic elastography data is performed to obtain the spatial gradient tensor set of ultrasonic elastography data; The frequency domain energy distribution feature set is coupled with the spatial gradient tensor set in the tensor domain to obtain the feature waveform set of the target patient. The tensor domain coupling calculation formula is as follows: ; In the formula, For the characteristic waveform set, It is a set of frequency domain energy distribution features. For the spatial gradient tensor set, It is the transpose of the spatial gradient tensor. is the coupling coefficient.

[0007] In a preferred embodiment, matching the feature waveform set with the corresponding initial position marker set of acupoints in a preset acupoint waveform template library includes: The phase space of the feature waveform set is reconstructed to obtain the phase space waveform set of the feature waveform set; The Euclidean distance between each point in the phase space waveform set and the corresponding waveform marker point in the preset acupoint waveform template library is used as the dimensionality matching result of the preset acupoint waveform template library. The Euclidean distance calculation formula is as follows: ; In the formula, For the dimension matching results, For the characteristic values ​​of the target patient, Indexed by feature dimensions, For phase space waveform point indexing, These are the marker points for the phase space waveform set. For the marker points of the preset acupoint template library, For acupoint template index, The total number of feature dimensions. The feature values ​​of the template library; The dimensional matching results are mapped to bioelectric conduction path markers to obtain the initial location marker set of acupoints for the target patient.

[0008] In a preferred embodiment, phase space reconstruction is performed on the feature waveform set to obtain a phase space waveform set of the feature waveform set, including: The feature waveform set is separated into a dynamic component subset and a static component subset; Map the dynamic component subset and the static component subset to the spatial extension waveform corresponding to the physiological dimension of the target patient; The phase space waveform set of the feature waveform set is obtained by fusing the topological rules of human bioelectric conduction with the spatially extended waveform.

[0009] In a preferred embodiment, the initial location marker set of the acupoints is projected onto the body surface area corresponding to the target patient, and distortion correction is performed on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient, including: The initial location markers of the acupoints are mapped to the body surface space coordinate system to obtain the original projected coordinate set of the target patient; The original projection coordinate set is spatially calibrated based on the preset acupoint distribution topology. Based on the physical structure parameters of the preset acupoint distribution topology, optical distortion compensation is performed on the calibrated projection coordinate set to obtain the high-precision acupoint coordinate set of the target patient.

[0010] In a preferred embodiment, mapping the high-precision acupoint coordinate set to three-dimensional space to obtain a three-dimensional acupoint model of the target patient includes: The two-dimensional surface coordinates of the high-precision acupoint coordinate set are extended in the depth direction to obtain the initial three-dimensional acupoint cloud of the target patient; The initial three-dimensional acupoint cloud was spatially corrected based on the distribution pattern of human acupoints. The corrected 3D acupoint cloud is topologically connected to the anatomical landmarks of the target patient to obtain the 3D acupoint model of the target patient.

[0011] In a preferred embodiment, the three-dimensional acupoint model is registered in real time based on the real-time physiological state of the target patient to obtain a standard spatial parameter set for the three-dimensional acupoint model, including: Spatial coordinate acquisition is performed on the high-precision acupoint coordinate set to obtain the initial three-dimensional coordinate set of the target patient; The initial three-dimensional coordinate set is physically integrated with the structured coordinate topology to obtain the standard spatial parameter set of the three-dimensional acupoint model.

[0012] In a preferred embodiment, the standard spatial parameter set is superimposed onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient, including: The standard spatial parameter set is projected onto the patient coordinate system of the three-dimensional acupoint model to obtain the initial projection mark point set of the target patient; The initial set of projected marker points is compared with the spatial location of the target patient to obtain the deviation dataset of the spatial location; The projection angle is adjusted based on the spatial position deviation dataset to obtain the calibrated projection parameters for the target patient; The calibration projection parameters are projected onto the target patient, resulting in a visualized acupoint view of the target patient.

[0013] In a preferred embodiment, adjusting the projection angle based on the spatial position deviation dataset to obtain the calibrated projection parameters of the target patient includes: Analyze the deviation direction and magnitude data of the projection angle in the spatial position deviation data set; The spatial pointing parameters of the projection are dynamically corrected based on the deviation direction and the deviation magnitude data. The corrected spatial pointing parameters are registered and verified with the surface anatomical coordinate system of the target patient to obtain the calibration projection parameters of the target patient.

[0014] To address the aforementioned problems, the present invention also provides an intelligent acupoint positioning system based on multimodal imaging, the system comprising: The feature decoupling module is used to perform dual-modal feature decoupling on the infrared thermal imaging data and ultrasound elastography data of the target patient to obtain the feature waveform set of the target patient; The waveform matching module is used to match the feature waveform set with the corresponding acupoint initial position marker point set in the preset acupoint waveform template library; The projection correction module is used to project the initial position marker set of the acupoints onto the body surface area corresponding to the target patient, and to perform distortion correction on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient. A three-dimensional acupoint model construction module is used to map the high-precision acupoint coordinate set to three-dimensional space to obtain a three-dimensional acupoint model of the target patient. The parameter registration module is used to perform real-time parameter registration of the three-dimensional acupoint model based on the real-time physiological state of the target patient, so as to obtain the standard spatial parameter set of the three-dimensional acupoint model. The visualization overlay module is used to overlay the standard spatial parameter set onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention decouples infrared thermal imaging data and ultrasonic elastography data in two modes, generates a precise feature waveform set by tensor domain coupling, and then reconstructs the phase space and matches it with a preset acupoint waveform template library. This can effectively capture the physical and physiological characteristics of acupoints, improve the accuracy of the initial acupoint location, and lay a reliable foundation for subsequent processing.

[0016] 2. This invention obtains a high-precision coordinate set by correcting the distortion of acupoint projection, maps it to three-dimensional space to construct a three-dimensional acupoint model, and combines it with real-time physiological status for parameter registration and visualization overlay. This can accurately present the spatial location and distribution characteristics of acupoints, improve the accuracy and intuitiveness of acupoint positioning, and provide precise acupoint positioning support for related medical applications. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an intelligent acupoint positioning method based on multimodal imaging, provided in an embodiment of the present invention. Figure 2 A functional block diagram of an acupoint intelligent positioning system based on multimodal imaging provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides a method for intelligent acupoint location based on multimodal imaging. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for intelligent acupoint location based on multimodal imaging can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating an intelligent acupoint localization method based on multimodal imaging according to an embodiment of the present invention. In this embodiment, the intelligent acupoint localization method based on multimodal imaging includes: S1. Perform dual-modal feature decoupling on the infrared thermal imaging data and ultrasound elastography data of the target patient to obtain the characteristic waveform set of the target patient; In this embodiment of the invention, the step of decoupling the infrared thermal imaging data and ultrasound elastography data of the target patient into a dual-modal feature set to obtain the target patient's feature waveform set includes: Wavelet transform decomposition is performed on the infrared thermal imaging data to obtain the frequency domain energy distribution feature set of the infrared thermal imaging data; Time-frequency joint analysis of ultrasonic elastography data is performed to obtain the spatial gradient tensor set of ultrasonic elastography data; The frequency domain energy distribution feature set is coupled with the spatial gradient tensor set in the tensor domain to obtain the feature waveform set of the target patient. The tensor domain coupling calculation formula is as follows: ; In the formula, For the characteristic waveform set, It is a set of frequency domain energy distribution features. For the spatial gradient tensor set, It is the transpose of the spatial gradient tensor. is the coupling coefficient.

[0021] Specifically, wavelet transform decomposition is performed on the infrared thermal imaging data to obtain the frequency domain energy distribution feature set of the infrared thermal imaging data. The infrared thermal imaging data is a continuous image sequence of the surface temperature distribution of the target patient. Each image contains temperature values ​​at different locations. During wavelet transform decomposition, a preset wavelet basis function is selected.

[0022] Furthermore, the temperature signal of each image is decomposed into multiple sub-signals of different frequencies, where low-frequency sub-signals correspond to regions with gentle temperature changes and high-frequency sub-signals correspond to regions with rapid temperature changes. The energy value of each sub-signal is calculated, and the energy value is the sum of the temperature change intensities contained in that sub-signal.

[0023] Furthermore, by associating the energy values ​​of each sub-signal with their corresponding frequency ranges in ascending order of frequency, the resulting ordered set is the frequency domain energy distribution feature set of the infrared thermal imaging data.

[0024] Furthermore, time-frequency joint analysis was performed on the ultrasound elastography data to obtain the spatial gradient tensor set of the ultrasound elastography data. The ultrasound elastography data is a sequence of elastic deformation images of the target patient tissue under different pressures, and each image contains the elastic coefficient value of each location of the tissue.

[0025] Furthermore, in the joint time-frequency analysis, the elastic coefficient change signal of each spatial location at different time points is first extracted, and then the frequency components of the signal in different time periods are analyzed to determine the elastic change frequency characteristics of each location. At the same time, the difference in elastic coefficient between each location and its adjacent locations at the same time point is calculated to obtain the spatial gradient value.

[0026] Furthermore, the frequency characteristics and spatial gradient values ​​of each location are integrated into a multidimensional array according to three-dimensional spatial coordinates and time order. Each array element contains the time-frequency characteristics and spatial gradient information of the corresponding location, and the resulting set is the spatial gradient tensor set of the ultrasound elastography data.

[0027] Furthermore, the frequency domain energy distribution feature set and the spatial gradient tensor set are coupled in the tensor domain to obtain the feature waveform set of the target patient. First, the spatial correspondence between the features in the two sets is determined, that is, the temperature signal position in the frequency domain energy distribution feature set is matched one-to-one with the tissue elastic position in the spatial gradient tensor set.

[0028] Furthermore, the frequency domain energy value at the corresponding position is combined with the elements in the spatial gradient tensor according to a preset rule. For example, the energy value at the same position is used as the weight of a certain dimension of the tensor. The tensor values ​​are adjusted to form a fused multidimensional tensor. Then, each fused tensor is expanded along the time axis or frequency axis to obtain a curve that changes with time or frequency. Each curve is a feature waveform, and the set of all feature waveforms is the feature waveform set of the target patient.

[0029] Specifically, It is a frequency domain energy distribution feature set, which is obtained by wavelet transform decomposition of infrared thermal imaging data of the target patient, and contains the energy distribution features of infrared thermal imaging data at different frequencies. It is a set of spatial gradient tensors, which is derived from the time-frequency joint analysis of the ultrasound elastography data of the target patient. It contains the spatial gradient and temporal frequency characteristics of the ultrasound elastography data. It is the transpose of the spatial gradient tensor, derived from the set of spatial gradient tensors. The transpose operation is performed to obtain the result by swapping the elements. The row and column positions of elements are adjusted without changing the numerical values ​​of the elements themselves, only adjusting the dimensional structure of the tensor to meet the dimensional matching requirements of coupled operations.

[0030] Furthermore, The coupling coefficient is a value pre-set based on the modal characteristics of infrared thermal imaging data and ultrasonic elastography data. The setting criteria include the signal-to-noise ratio of the two modal data and their importance in feature description. It is used to balance the contribution ratio of the frequency domain energy distribution feature set and the spatial gradient tensor set in the coupling process.

[0031] Furthermore, the significance of this formula lies in fusing the frequency domain energy distribution feature set with the spatial gradient tensor set through tensor multiplication. Specifically, the spatial gradient tensor set is first multiplied... Transpose to obtain ,make With frequency domain energy distribution feature set Dimensional matching is performed for multiplication, and the product of the two is then combined with the coupling coefficient. Multiplication, through Adjusting the weights of the two feature sets during the fusion process yields the final result. This is a feature waveform set that integrates infrared thermal characteristics and ultrasonic mechanical characteristics, achieving effective integration of dual-modal features.

[0032] Furthermore, when the frequency domain energy distribution feature set When the eigenvalues ​​in the matrix increase, and Under the condition that remains unchanged, the characteristic waveform set The corresponding eigenvalue will increase accordingly, meaning that the energy characteristics in infrared thermal imaging data are affected by... The effect is amplified when the spatial gradient tensor set... As the eigenvalues ​​in the tensor increase, its transpose tensor... The corresponding eigenvalues ​​will also increase, in and If it remains unchanged, The corresponding eigenvalues ​​will increase accordingly, that is, the spatial gradient characteristics of ultrasound elastography data will affect... The effect is amplified when the coupling coefficient is increased. When it increases, at and If it remains unchanged, All feature values ​​will increase proportionally, and the two modal features will increase proportionally. The overall contribution has increased; when When decreasing, All eigenvalues ​​will decrease proportionally, and the overall contribution of both modal features will weaken.

[0033] In summary, wavelet transform decomposition of infrared thermal imaging data yields a frequency domain energy distribution feature set, which can accurately extract temperature energy features at different frequencies, reflecting the differences in thermal metabolism in acupoint areas and providing thermal characteristic basis for subsequent matching.

[0034] In summary, the spatial gradient tensor set obtained by performing time-frequency joint analysis on ultrasound elastography data can capture the changes in tissue elasticity in the spatiotemporal dimensions, reflect the differences in mechanical properties of acupoint areas, and supplement structural feature information in addition to thermal features.

[0035] In summary, the characteristic waveform set obtained by tensor-domain coupling of the frequency domain energy distribution feature set and the spatial gradient tensor set can integrate thermal and mechanical dual-modal features, comprehensively reflect the multi-dimensional attributes of acupoints, improve the completeness and accuracy of feature description, and provide a more reliable feature basis for acupoint positioning.

[0036] S2. Match the feature waveform set with the corresponding acupoint initial position marker point set in the preset acupoint waveform template library; In this embodiment of the invention, matching the feature waveform set with the corresponding initial position marker point set of acupoints in a preset acupoint waveform template library includes: The phase space of the feature waveform set is reconstructed to obtain the phase space waveform set of the feature waveform set; The Euclidean distance between each point in the phase space waveform set and the corresponding waveform marker point in the preset acupoint waveform template library is used as the dimensionality matching result of the preset acupoint waveform template library. The Euclidean distance calculation formula is as follows: ; In the formula, For the dimension matching results, For the characteristic values ​​of the target patient, Indexed by feature dimensions, For phase space waveform point indexing, These are the marker points for the phase space waveform set. For the marker points of the preset acupoint template library, For acupoint template index, The total number of feature dimensions. The feature values ​​of the template library; The dimensional matching results are mapped to bioelectric conduction path markers to obtain the initial location marker set of acupoints for the target patient.

[0037] Phase space reconstruction is performed on the feature waveform set to obtain the phase space waveform set of the feature waveform set, including: The feature waveform set is separated into a dynamic component subset and a static component subset; Map the dynamic component subset and the static component subset to the spatial extension waveform corresponding to the physiological dimension of the target patient; The phase space waveform set of the feature waveform set is obtained by fusing the topological rules of human bioelectric conduction with the spatially extended waveform.

[0038] Specifically, the phase space is reconstructed from the feature waveform set to obtain the phase space waveform set of the feature waveform set. The feature waveform set is a collection of bioelectric signal waveforms related to acupoints collected by the detection device. Each waveform changes with time as the axis. The phase space reconstruction is to convert these one-dimensional time waveforms into a set of points in a multi-dimensional space.

[0039] Furthermore, by selecting a fixed time interval, signal values ​​at different times are extracted from each waveform sequentially. These signal values ​​are then used as coordinates in different dimensions according to the extraction order. Each waveform is thus transformed into a point in a multi-dimensional space. The set of points corresponding to all waveforms is the phase space waveform set of the feature waveform set.

[0040] Furthermore, the Euclidean distance between each point in the phase space waveform set and the corresponding waveform marker point in the preset acupoint waveform template library is used as the dimensional matching result of the preset acupoint waveform template library.

[0041] Furthermore, the preset acupoint waveform template library stores the marker points of the standard bioelectric signal waveforms corresponding to known acupoints in the phase space. Each marker point also contains coordinates in multiple dimensions. Euclidean distance refers to the square root of the sum of the squares of the differences between the coordinates of two points in each dimension in multidimensional space.

[0042] Furthermore, the distance between each point in the phase space waveform set and the standard marker point of the corresponding acupoint in the template library is calculated. All these distance values ​​are organized according to the corresponding acupoints, and the resulting set is the dimensional matching result of the preset acupoint waveform template library.

[0043] Furthermore, the dimensional matching results are mapped to bioelectric conduction path markers to obtain the initial position marker set of acupoints for the target patient. The bioelectric conduction path markers refer to the pre-determined bioelectric signal conduction paths between different acupoints and their corresponding position identifiers. Each path marker is associated with a specific initial position of the acupoint.

[0044] Furthermore, the degree of matching is determined based on the distance value in the dimensional matching results. The smaller the distance, the higher the degree of matching. The path marker with the highest degree of matching is selected, and the acupoint location information corresponding to the marker is extracted. The set of all these location information is the initial acupoint location marker set of the target patient.

[0045] Specifically, the feature waveform set is separated into a dynamic component subset and a static component subset. Each waveform in the feature waveform set contains a feature component that changes over time. By calculating the change amplitude of each component within a preset time window, the component with a change amplitude greater than a preset threshold is classified as a dynamic component, and the component with a change amplitude less than or equal to the preset threshold is classified as a static component.

[0046] Furthermore, all dynamic components are arranged in their corresponding waveform order to form a dynamic component subset, and all static components are arranged in the same order to form a static component subset.

[0047] Furthermore, the dynamic component subset and the static component subset are mapped to the spatial extension waveforms corresponding to the physiological dimensions of the target patient. The physiological dimensions of the target patient include body surface coordinates, tissue depth, and functional dimensions related to physiological activities, while the dynamic component subset corresponds to the physiological activity dimensions.

[0048] Furthermore, the waveforms are expanded along this dimension according to the time series to form a spatially distributed waveform that changes with physiological activities; the static component subset corresponds to a fixed anatomical dimension, and its waveforms are arranged along this dimension according to spatial position to form a stable spatially distributed waveform. The two waveforms together constitute the spatially extended waveform corresponding to the physiological dimension of the target patient.

[0049] Furthermore, the topological rules of human bioelectric conduction are fused with spatially extended waveforms to obtain the phase space waveform set of the characteristic waveform set. The topological rules of human bioelectric conduction are the paths and distribution patterns of bioelectricity conduction along meridians, nerves and tissue gaps in the body, including conduction direction, node connection relationship and signal attenuation characteristics.

[0050] Furthermore, the connection structure of the spatially extended waveform is adjusted according to this rule so that the waveform of the dynamic component is distributed along the bioelectrical conduction path, and the waveform of the static component is matched with the position of the conduction node.

[0051] Furthermore, the amplitude of the waveform is simultaneously modified according to rules to reflect the attenuation characteristics of the bioelectric signal. The resulting multidimensional waveform set, which includes spatial location, temporal variation, and bioelectric conduction characteristics, is the phase space waveform set of the characteristic waveform set.

[0052] Specifically, It is a dimensional matching result, which is obtained by calculating the Euclidean distance between the marker points of the phase space waveform set and the marker points of the preset acupoint template library. These are the feature values ​​of the target patient, derived from the phase space waveform set markers obtained after reconstructing the phase space of the target patient's feature waveform set. The feature values ​​of each dimension reflect the bioelectric signal characteristics of the acupoints of the target patient. These are feature values ​​from the template library, sourced from marker points in the preset acupoint template library. The standard feature values ​​for each dimension are pre-set based on the bioelectrical signal characteristics of known acupoints. It is a feature dimension index, which is derived from the dimension number corresponding to the feature value. Each number corresponds to a specific feature dimension, such as the amplitude and frequency of a bioelectric signal.

[0053] Furthermore, It is a phase space waveform point index, which is derived from the number of each marked point in the phase space waveform set of the target patient. Each number corresponds to a unique waveform point. It is an acupoint template index, which is derived from the number of each standard marker point in the preset acupoint template library. Each number corresponds to a template for a specific acupoint. It represents the total number of feature dimensions, derived from the number of dimensions contained in the extracted feature values, and determined by the feature dimensions of the feature waveform set.

[0054] Furthermore, the significance of this formula lies in calculating the marker points in the phase space waveform set of the target patient. Marking points in the preset acupoint template library The distance in the multidimensional feature space is calculated as follows: First, the difference between the feature values ​​of the two points in each feature dimension is calculated. Then, each difference is squared. Finally, all the squared results are added together. The result is the Euclidean distance between the two points. This distance is used as the dimension matching result to measure the similarity between the waveform points of the target patient and the standard template points.

[0055] Furthermore, when the characteristic values ​​of the target patient Eigenvalues ​​of the template library As the difference in a certain feature dimension increases, the squared difference in that dimension also increases, and the sum of the squares of all dimensions increases accordingly, ultimately leading to the Euclidean distance. An increase indicates a decrease in the matching degree between the two points. When and As the differences across all feature dimensions decrease, the sum of squares decreases, and the Euclidean distance decreases. A decrease indicates an improved match between the two points. When... and When the differences across all feature dimensions are zero, the sum of squares is zero, and the Euclidean distance is zero. A value of zero indicates that the two points are a perfect match.

[0056] In summary, reconstructing the phase space of the feature waveform set to obtain the phase space waveform set can transform one-dimensional waveform features into a multi-dimensional point set, comprehensively showcasing the spatial distribution of acupoint features and providing a multi-dimensional foundation for subsequent matching.

[0057] In summary, using the Euclidean distance between each point in the phase space waveform set and the corresponding point in the preset template as the dimensional matching result can accurately measure the degree of feature similarity through quantified distance, providing an objective basis for matching degree determination.

[0058] In summary, mapping the dimensional matching results to bioelectric conduction path markers yields a set of initial acupoint markers. This allows for the determination of the initial position by combining the bioelectric conduction laws, ensuring that the marker set conforms to both waveform matching results and physiological characteristics. This improves the accuracy of initial positioning and lays the foundation for subsequent projection correction.

[0059] In summary, separating the feature waveform set into dynamic component subsets and static component subsets can distinguish between dynamic features that change with physiological activities and stable static features in the waveform, providing a well-defined basis of data for subsequent mapping.

[0060] In summary, mapping dynamic and static component subsets to spatially extended waveforms corresponding to the physiological dimensions of the target patient enables the characteristics of the two components to accurately correspond to the patient's physiological structure and functional dimensions, forming a spatial distribution waveform that fits the patient's actual situation.

[0061] In summary, fusing the topological rules of human bioelectric conduction with spatially extended waveforms to obtain a phase space waveform set can incorporate the physiological laws of bioelectric conduction, making the waveform set more consistent with the biophysical characteristics of acupoints, improving the accuracy of the phase space waveform set in representing acupoint features, and providing a more reliable waveform basis for subsequent matching.

[0062] S3. Project the initial position marker set of the acupoints onto the body surface area corresponding to the target patient, and perform distortion correction on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient; In this embodiment of the invention, the initial location marker set of the acupoints is projected onto the body surface area corresponding to the target patient, and distortion correction is performed on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient, including: The initial location markers of the acupoints are mapped to the body surface space coordinate system to obtain the original projected coordinate set of the target patient; The original projection coordinate set is spatially calibrated based on the preset acupoint distribution topology. Based on the physical structure parameters of the preset acupoint distribution topology, optical distortion compensation is performed on the calibrated projection coordinate set to obtain the high-precision acupoint coordinate set of the target patient.

[0063] Specifically, the initial position marker point set of acupoints is obtained from the preset acupoint waveform template library. This set contains the position information of each acupoint in the standard coordinate system.

[0064] Furthermore, the transformation relationship between the two coordinate systems is determined by identifying the same anatomical reference points in the standard coordinate system and the constructed body surface space coordinate system, such as the protrusions of specific bones.

[0065] Furthermore, according to this transformation relationship, the position information of each acupoint in the initial position marker set is transformed into the body surface space coordinate system to obtain the specific coordinates of each acupoint in the coordinate system. The set of all these coordinates is the original projected coordinate set of the target patient.

[0066] Furthermore, the preset acupoint distribution topology is based on fixed rules for the relative positions of acupoints determined by human meridians and anatomy, including the distance range, connecting angle, and arrangement order between related acupoints.

[0067] Furthermore, the actual relative position of each acupoint and its surrounding related acupoints in the original projected coordinate set is compared one by one with the corresponding standard relative position in the preset acupoint distribution topology. If there is a deviation between the actual position and the standard position, the specific distance and direction that need to be adjusted are calculated according to the standard relationship in the topology.

[0068] Furthermore, the coordinates of the corresponding acupoints in the original projection coordinate set are moved so that the relative positions of all acupoints after adjustment conform to the preset distribution rules, thus obtaining the calibrated projection coordinate set.

[0069] Furthermore, the physical structural parameters of the preset acupoint distribution topology include physical characteristics that affect optical imaging, such as skin thickness, subcutaneous tissue density, and surrounding bone morphology in the acupoint area. Optical distortion can cause deviations such as edge stretching or center compression between the projected coordinates and the actual position.

[0070] Furthermore, based on these physical structural parameters, the degree of distortion in different regions is determined. For example, areas with prominent bones have less distortion, while areas with thicker soft tissue have more distortion, for each acupoint in the calibrated projection coordinate set.

[0071] Furthermore, a compensation value is calculated based on the degree of distortion in the area where the acupoint is located. The coordinates of the acupoint in the body surface spatial coordinate system are adjusted according to the compensation value so that the adjusted coordinates accurately reflect the actual position of the acupoint on the patient's body surface. The set of these adjusted coordinates is the high-precision acupoint coordinate set of the target patient.

[0072] In summary, mapping the initial set of acupoint location markers to the body surface spatial coordinate system yields the original projected coordinate set, which can transform the template markers to the patient's specific coordinate system, achieving spatial correspondence and providing a unified benchmark for subsequent correction.

[0073] In summary, calibrating the original projection coordinate set based on the preset acupoint distribution topology can adjust the coordinates according to the inherent distribution rules of acupoints, correct deviations caused by individual differences, and improve coordinate accuracy.

[0074] In summary, by performing optical distortion compensation on the calibrated coordinate set based on preset topological physical structure parameters, a high-precision coordinate set can be obtained. This can eliminate optical distortion, optimize accuracy, provide reliable data for subsequent processing, and ensure accurate positioning.

[0075] S4. Map the high-precision acupoint coordinate set to three-dimensional space to obtain the three-dimensional acupoint model of the target patient; In this embodiment of the invention, mapping the high-precision acupoint coordinate set to three-dimensional space to obtain a three-dimensional acupoint model of the target patient includes: The two-dimensional surface coordinates of the high-precision acupoint coordinate set are extended in the depth direction to obtain the initial three-dimensional acupoint cloud of the target patient; The initial three-dimensional acupoint cloud was spatially corrected based on the distribution pattern of human acupoints. The corrected 3D acupoint cloud is topologically connected to the anatomical landmarks of the target patient to obtain the 3D acupoint model of the target patient.

[0076] Specifically, the two-dimensional surface coordinates of the high-precision acupoint coordinate set record the front-back and left-right positions of each acupoint on the patient's body surface, based on the subcutaneous depth data corresponding to each acupoint in human anatomy, such as the depth of acupoints in the muscle layer being greater than that of acupoints in the subcutaneous fat layer.

[0077] Furthermore, corresponding depth and shallowness values ​​are added to each two-dimensional coordinate, transforming the coordinates of each acupoint from a two-dimensional form containing front-back and left-right to a three-dimensional form containing front-back, left-right, and depth. The set of all these three-dimensional coordinate points is the initial three-dimensional acupoint cloud of the target patient.

[0078] Furthermore, the distribution pattern of acupoints in the human body includes fixed rules such as the arrangement order of acupoints along the meridians, the correspondence between acupoints and the gaps between bones, and the changes in the depth gradient of adjacent acupoints. The three-dimensional coordinates of each acupoint in the initial three-dimensional acupoint cloud are compared with these rules one by one.

[0079] Furthermore, if a certain acupoint deviates from the direction of its corresponding meridian, its front-back or left-right coordinates are adjusted according to the spatial path of the meridian; if the depth value of a certain acupoint does not conform to the gradient law of the area, its depth coordinates are corrected by referring to the depth values ​​of adjacent acupoints. After adjustment, the spatial position of all acupoints conforms to the distribution law of human acupoints, and a corrected three-dimensional acupoint cloud is obtained.

[0080] Furthermore, the anatomical landmarks of the target patients include anatomical structures with fixed spatial locations such as the spinous processes of the spine, the edges of the ribs, and the inter-joint spaces. The three-dimensional coordinates of these points have been obtained through previous scans, based on the correlation between acupoints and anatomical landmarks in traditional Chinese medicine theory.

[0081] For example, if an acupoint is located "1.5 cun lateral to the spinous process of the third lumbar vertebra", the spatial distance and direction of the connection between each acupoint and the corresponding anatomical landmark in the corrected three-dimensional acupoint cloud are calculated. The acupoints and anatomical landmarks are then connected in three-dimensional space according to these distances and directions to form a network structure that includes the location of the acupoints and the association between the acupoints and the anatomical structures. This structure is the three-dimensional acupoint model of the target patient.

[0082] In summary, extending the two-dimensional surface coordinates to obtain the initial three-dimensional acupoint cloud can supplement subcutaneous depth information, lay the foundation for the three-dimensional model, and make the acupoint description more consistent with the three-dimensional structure of the human body.

[0083] In summary, correcting the initial 3D point cloud based on the distribution patterns of acupoints can correct positional deviations, ensure that the spatial location of acupoints conforms to the human physiological structure, and improve data accuracy.

[0084] In summary, a three-dimensional acupoint model is obtained by topologically connecting the corrected point cloud with anatomical landmarks. This model can associate acupoints with fixed human structures, forming a three-dimensional model containing the location and interrelationships of acupoints. This provides a structurally complete foundation for subsequent processing and improves the reliability of positioning.

[0085] S5. Based on the real-time physiological state of the target patient, perform real-time parameter registration on the three-dimensional acupoint model to obtain the standard spatial parameter set of the three-dimensional acupoint model; In this embodiment of the invention, the three-dimensional acupoint model is registered in real time based on the real-time physiological state of the target patient to obtain a standard spatial parameter set of the three-dimensional acupoint model, including: Spatial coordinate acquisition is performed on the high-precision acupoint coordinate set to obtain the initial three-dimensional coordinate set of the target patient; The initial three-dimensional coordinate set is physically integrated with the structured coordinate topology to obtain the standard spatial parameter set of the three-dimensional acupoint model.

[0086] Specifically, spatial coordinates are acquired from a high-precision acupoint coordinate set to obtain the initial three-dimensional coordinate set of the target patient. A real-time three-dimensional scanning device is used to dynamically scan the acupoint area of ​​the target patient. This device can capture the spatial position of acupoints under different physiological states (such as breathing, slight changes in body position) for each acupoint in the high-precision acupoint coordinate set.

[0087] Furthermore, the coordinate values ​​of the acupoint in the front-back, left-right, and depth directions at different times during the scanning process are recorded. After removing obvious outliers caused by noise, the average coordinates of each acupoint are used as the real-time three-dimensional coordinates of that acupoint. The set formed by combining the real-time three-dimensional coordinates of all acupoints is the initial three-dimensional coordinate set of the target patient.

[0088] Furthermore, the initial three-dimensional coordinate set is physically integrated with the structured coordinate topology to obtain the standard spatial parameter set of the three-dimensional acupoint model. The structured coordinate topology is a spatial relationship framework for acupoints constructed based on the standard anatomical structure of the human body, which includes fixed spatial parameters such as standard distances, angles, and hierarchical distributions between acupoints. The coordinates of each acupoint in the initial three-dimensional coordinate set are compared with the standard coordinates of the corresponding acupoints in the structured coordinate topology.

[0089] Furthermore, the deviations between the two in the front-back, left-right, and depth directions are calculated. Based on the deviation values, the coordinates of the acupoints in the initial three-dimensional coordinate set are adjusted so that the spatial relationships such as the distance and angle between the acupoints after adjustment are consistent with the standard parameters in the structured coordinate topology.

[0090] Furthermore, by integrating and adjusting the coordinates of all acupoints and their spatial parameters, a set of parameters containing a unified spatial standard is formed, which is the standard spatial parameter set of the three-dimensional acupoint model.

[0091] In summary, by acquiring spatial coordinates from a high-precision acupoint coordinate set to obtain an initial three-dimensional coordinate set, the dynamic spatial position of acupoints under the real-time physiological state of the target patient can be captured, ensuring that the coordinate data matches the patient's current physiological state and providing real-time basic data for parameter registration.

[0092] In summary, physically integrating the initial 3D coordinate set with the structured coordinate topology to obtain a standard spatial parameter set enables the real-time acquired coordinate data to be integrated with the coordinate framework based on standard anatomical structures. This unifies the standard of acupoint spatial parameters, eliminates the influence of individual physiological fluctuations on model parameters, ensures the consistency and standardization of the spatial parameters of the 3D acupoint model, provides standardized parameter support for subsequent visualization overlay, and improves the stability and accuracy of acupoint positioning.

[0093] S6. The standard spatial parameter set is superimposed onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient.

[0094] In this embodiment of the invention, the standard spatial parameter set is superimposed onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient, including: The standard spatial parameter set is projected onto the patient coordinate system of the three-dimensional acupoint model to obtain the initial projection mark point set of the target patient; The initial set of projected marker points is compared with the spatial location of the target patient to obtain the deviation dataset of the spatial location; The projection angle is adjusted based on the spatial position deviation dataset to obtain the calibrated projection parameters for the target patient; The calibration projection parameters are projected onto the target patient, resulting in a visualized acupoint view of the target patient.

[0095] Based on the spatial position deviation dataset, the projection angle is adjusted to obtain the calibrated projection parameters for the target patient, including: Analyze the deviation direction and magnitude data of the projection angle in the spatial position deviation data set; The spatial pointing parameters of the projection are dynamically corrected based on the deviation direction and the deviation magnitude data. The corrected spatial pointing parameters are registered and verified with the surface anatomical coordinate system of the target patient to obtain the calibration projection parameters of the target patient.

[0096] Specifically, the standard spatial parameter set includes the standard spatial position and interrelationship parameters of each acupoint in the three-dimensional acupoint model, and the patient coordinate system of the three-dimensional acupoint model is a three-dimensional coordinate system established based on the anatomical structure of the target patient's body surface.

[0097] Furthermore, each parameter in the standard spatial parameter set is projected one-to-one into the patient's coordinate system according to its corresponding acupoint location, and a corresponding spatial point is marked for each acupoint in the coordinate system. The set of all these marked points is the initial projection mark point set for the target patient.

[0098] Furthermore, the initial projection mark point set records the projection position of each acupoint in the patient coordinate system. The spatial position of the target patient refers to the three-dimensional space of the actual body parts of the patient. The coordinates of each mark point in the initial projection mark point set are compared with the coordinates of the corresponding acupoint in the actual spatial position of the patient.

[0099] Furthermore, the differences between the two in the three directions of front and back, left and right, and depth are calculated. These differences of all marked points are organized according to the corresponding acupoints, and the resulting dataset containing the positional deviation of each acupoint is the spatial position deviation dataset.

[0100] Furthermore, each deviation value in the spatial position deviation dataset contains the direction and magnitude of the deviation. The direction of the projection angle adjustment is determined based on the deviation direction, and the adjustment range is determined based on the deviation magnitude. The projection angle of the projection device is adjusted according to the determined direction and range. The adjusted projection angle, projection range, and other parameters are the calibration projection parameters for the target patient.

[0101] Furthermore, using a projection device according to the calibrated projection parameters, the marked patterns of each acupoint are projected onto the corresponding positions on the body surface of the target patient, ensuring that each marked pattern accurately covers the actual position of the corresponding acupoint during the projection process.

[0102] Furthermore, after projection is completed, the combination of acupoint marker patterns formed on the patient's body surface that can be directly observed constitutes the visual acupoint view of the target patient.

[0103] Specifically, to analyze the deviation direction and magnitude of the projection angle in the spatial position deviation dataset, it is necessary to extract the deviation record of each acupoint projection in the dataset one by one, and determine whether the deviation direction corresponding to each record is forward, backward, left, right, deeper or shallower.

[0104] Furthermore, the deviation distance values ​​in each direction are extracted simultaneously, such as how much the projection point of a certain acupoint deviates forward and how much it deviates to the left. These directional information and distance values ​​are then categorized and organized by acupoint to form a clear list of deviation directions and a list of deviation magnitudes.

[0105] Furthermore, the spatial pointing parameters of the projection are dynamically corrected based on the deviation direction and deviation magnitude data. The spatial pointing parameters include the horizontal rotation angle, vertical tilt angle and projection depth parameters of the projection device, and the correction direction is determined according to the deviation direction.

[0106] Furthermore, if there is a forward deviation, the horizontal rotation angle will be adjusted backward; if there is a leftward deviation, the vertical tilt angle will be adjusted to the right. The correction range is determined according to the deviation range. The larger the deviation distance, the larger the angle adjustment range in the corresponding direction. By gradually fine-tuning, the pointing of the projection device is gradually brought closer to the target position until the adjustment requirements corresponding to the deviation direction and magnitude are fully compensated.

[0107] Furthermore, the corrected spatial pointing parameters are registered and verified with the target patient's surface anatomical coordinate system to obtain the target patient's calibration projection parameters. The surface anatomical coordinate system is established with the patient's skeletal landmarks, joint positions, etc., as reference points.

[0108] Furthermore, the corrected spatial pointing parameters are input into the projection device to project a set of verification markers, and the actual positions of these markers in the body surface anatomical coordinate system are measured.

[0109] Furthermore, the coordinate system is compared with the standard position of the corresponding acupoint. If all verification markers completely coincide with the standard position, the current spatial pointing parameter is the calibration projection parameter. If there are non-coincident markers, the correction and verification process is repeated until all markers coincide with the standard position.

[0110] In summary, projecting the standard spatial parameter set onto the patient coordinate system of the three-dimensional acupoint model to obtain the initial projection mark point set can make the standard parameters consistent with the patient's own coordinate system, providing a basis for subsequent deviation comparison and ensuring the initial correspondence between the projection marks and the model.

[0111] In summary, by comparing the initial set of projected markers with the spatial location of the target patient to obtain a deviation dataset, the difference between the projected markers and the actual location can be accurately identified, the direction and magnitude of adjustment can be clearly defined, and a basis for adjusting the projection angle can be provided.

[0112] In summary, adjusting the projection angle based on the spatial position deviation dataset yields calibrated projection parameters, which can specifically correct deviations, making the projection parameters fit the patient's actual spatial position and improving projection accuracy.

[0113] In summary, projecting the calibration projection parameters onto the target patient to obtain a visualized acupoint view can intuitively present precise acupoint information on the patient's body surface, achieving clear visualization of acupoints, providing accurate and intuitive positioning references for clinical applications, and improving the practicality and reliability of acupoint positioning.

[0114] In summary, analyzing the deviation direction and magnitude data in the spatial position deviation dataset can accurately pinpoint the root cause of the projection angle deviation, providing a clear basis for subsequent adjustments and ensuring the accuracy of the correction direction.

[0115] In summary, dynamically correcting the spatial pointing parameters of the projection based on deviation direction and amplitude data can specifically adjust the projection angle, making the projection parameters match the patient's actual spatial position and effectively reducing projection deviation.

[0116] In summary, registering and verifying the corrected spatial pointing parameters with the target patient's surface anatomical coordinate system allows for further verification of the accuracy of the projection parameters using anatomical benchmarks. This ensures that the final calibrated projection parameters accurately correspond to the patient's acupoint locations, providing reliable parameter support for the generation of visualized acupoint views and improving the accuracy of acupoint location.

[0117] like Figure 2 The diagram shown is a functional block diagram of an acupoint intelligent positioning system based on multimodal imaging provided in an embodiment of the present invention.

[0118] The acupoint intelligent positioning system 100 based on multimodal imaging described in this invention can be installed in an electronic device. Depending on the functions implemented, the acupoint intelligent positioning system 100 based on multimodal imaging may include a feature decoupling module 101, a waveform matching module 102, a projection correction module 103, a three-dimensional acupoint model construction module 104, a parameter registration module 105, and a visualization overlay module 106. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0119] In this embodiment, the functions of each module / unit are as follows: The feature decoupling module 101 is used to perform dual-modal feature decoupling on the infrared thermal imaging data and ultrasound elastography data of the target patient to obtain the feature waveform set of the target patient. The waveform matching module 102 is used to match the feature waveform set with the corresponding acupoint initial position marker point set in the preset acupoint waveform template library; The projection correction module 103 is used to project the initial position marker set of the acupoints onto the body surface area corresponding to the target patient, and to perform distortion correction on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient. The three-dimensional acupoint model construction module 104 is used to map the high-precision acupoint coordinate set to three-dimensional space to obtain the three-dimensional acupoint model of the target patient. The parameter registration module 105 is used to perform real-time parameter registration on the three-dimensional acupoint model based on the real-time physiological state of the target patient, so as to obtain the standard spatial parameter set of the three-dimensional acupoint model. The visualization overlay module 106 is used to overlay the standard spatial parameter set onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient.

[0120] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0121] The modules described as separate components may or may not be physically separate. The components shown as modules 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 embodiment according to actual needs.

[0122] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0123] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0124] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0125] Finally, 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 it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for intelligent acupoint localization based on multimodal imaging, characterized in that, The method includes: S1. Perform dual-modal feature decoupling on the infrared thermal imaging data and ultrasound elastography data of the target patient to obtain the characteristic waveform set of the target patient; S2. Match the feature waveform set with the corresponding acupoint initial position marker point set in the preset acupoint waveform template library; S3. Project the initial position marker set of the acupoints onto the body surface area corresponding to the target patient, and perform distortion correction on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient; S4. Map the high-precision acupoint coordinate set to three-dimensional space to obtain the three-dimensional acupoint model of the target patient; S5. Based on the real-time physiological state of the target patient, perform real-time parameter registration on the three-dimensional acupoint model to obtain the standard spatial parameter set of the three-dimensional acupoint model; S6. The standard spatial parameter set is superimposed onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient.

2. The acupoint intelligent positioning method based on multimodal imaging as described in claim 1, characterized in that, The infrared thermal imaging data and ultrasound elastography data of the target patient are decoupled in two modes to obtain the characteristic waveform set of the target patient, including: Wavelet transform decomposition is performed on the infrared thermal imaging data to obtain the frequency domain energy distribution feature set of the infrared thermal imaging data; Time-frequency joint analysis of ultrasonic elastography data is performed to obtain the spatial gradient tensor set of ultrasonic elastography data; The frequency domain energy distribution feature set is coupled with the spatial gradient tensor set in the tensor domain to obtain the feature waveform set of the target patient. The tensor domain coupling calculation formula is as follows: ; In the formula, For the characteristic waveform set, It is a set of frequency domain energy distribution features. For the spatial gradient tensor set, It is the transpose of the spatial gradient tensor. is the coupling coefficient.

3. The acupoint intelligent positioning method based on multimodal imaging as described in claim 1, characterized in that, Matching the feature waveform set with the corresponding initial position marker point set of acupoints in a preset acupoint waveform template library includes: The phase space of the feature waveform set is reconstructed to obtain the phase space waveform set of the feature waveform set; The Euclidean distance between each point in the phase space waveform set and the corresponding waveform marker point in the preset acupoint waveform template library is used as the dimensionality matching result of the preset acupoint waveform template library. The Euclidean distance calculation formula is as follows: ; In the formula, For the dimension matching results, For the characteristic values ​​of the target patient, Indexed by feature dimensions, For phase space waveform point indexing, These are the marker points for the phase space waveform set. For the marker points in the preset acupoint template library, For acupoint template index, The total number of feature dimensions. The feature values ​​of the template library; The dimensional matching results are mapped to bioelectric conduction path markers to obtain the initial location marker set of acupoints for the target patient.

4. The acupoint intelligent positioning method based on multimodal imaging as described in claim 3, characterized in that, Phase space reconstruction is performed on the feature waveform set to obtain the phase space waveform set of the feature waveform set, including: The feature waveform set is separated into a dynamic component subset and a static component subset; Map the dynamic component subset and the static component subset to the spatial extension waveform corresponding to the physiological dimension of the target patient; The phase space waveform set of the feature waveform set is obtained by fusing the topological rules of human bioelectric conduction with the spatially extended waveform.

5. The acupoint intelligent positioning method based on multimodal imaging as described in claim 1, characterized in that, The initial location marker set of the acupoints is projected onto the body surface area corresponding to the target patient, and distortion correction is performed on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient, including: The initial location markers of the acupoints are mapped to the body surface space coordinate system to obtain the original projected coordinate set of the target patient; The original projection coordinate set is spatially calibrated based on the preset acupoint distribution topology. Based on the physical structure parameters of the preset acupoint distribution topology, optical distortion compensation is performed on the calibrated projection coordinate set to obtain the high-precision acupoint coordinate set of the target patient.

6. The acupoint intelligent positioning method based on multimodal imaging as described in claim 1, characterized in that, Mapping the high-precision acupoint coordinate set to three-dimensional space yields a three-dimensional acupoint model of the target patient, including: The two-dimensional surface coordinates of the high-precision acupoint coordinate set are extended in the depth direction to obtain the initial three-dimensional acupoint cloud of the target patient; The initial three-dimensional acupoint cloud was spatially corrected based on the distribution pattern of human acupoints. The corrected 3D acupoint cloud is topologically connected to the anatomical landmarks of the target patient to obtain the 3D acupoint model of the target patient.

7. The acupoint intelligent positioning method based on multimodal imaging as described in claim 1, characterized in that, Based on the real-time physiological state of the target patient, the three-dimensional acupoint model is registered in real-time to obtain a standard spatial parameter set for the three-dimensional acupoint model, including: Spatial coordinate acquisition is performed on the high-precision acupoint coordinate set to obtain the initial three-dimensional coordinate set of the target patient; The initial three-dimensional coordinate set is physically integrated with the structured coordinate topology to obtain the standard spatial parameter set of the three-dimensional acupoint model.

8. The acupoint intelligent positioning method based on multimodal imaging as described in claim 1, characterized in that, The standard spatial parameter set is superimposed onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient, including: The standard spatial parameter set is projected onto the patient coordinate system of the three-dimensional acupoint model to obtain the initial projection mark point set of the target patient; The initial set of projected marker points is compared with the spatial location of the target patient to obtain the deviation dataset of the spatial location; The projection angle is adjusted based on the spatial position deviation dataset to obtain the calibrated projection parameters for the target patient; The calibration projection parameters are projected onto the target patient, resulting in a visualized acupoint view of the target patient.

9. The acupoint intelligent positioning method based on multimodal imaging as described in claim 8, characterized in that, Based on the spatial position deviation dataset, the projection angle is adjusted to obtain the calibrated projection parameters for the target patient, including: Analyze the deviation direction and magnitude data of the projection angle in the spatial position deviation data set; The spatial pointing parameters of the projection are dynamically corrected based on the deviation direction and the deviation magnitude data. The corrected spatial pointing parameters are registered and verified with the surface anatomical coordinate system of the target patient to obtain the calibration projection parameters of the target patient.

10. An intelligent acupoint positioning system based on multimodal imaging, characterized in that, The system includes: The feature decoupling module is used to perform dual-modal feature decoupling on the infrared thermal imaging data and ultrasound elastography data of the target patient to obtain the feature waveform set of the target patient; The waveform matching module is used to match the feature waveform set with the corresponding acupoint initial position marker point set in the preset acupoint waveform template library; The projection correction module is used to project the initial position marker set of the acupoints onto the body surface area corresponding to the target patient, and to perform distortion correction on the acupoint projection to obtain a high-precision acupoint coordinate set of the target patient. A three-dimensional acupoint model construction module is used to map the high-precision acupoint coordinate set to three-dimensional space to obtain a three-dimensional acupoint model of the target patient. The parameter registration module is used to perform real-time parameter registration of the three-dimensional acupoint model based on the real-time physiological state of the target patient, so as to obtain the standard spatial parameter set of the three-dimensional acupoint model. The visualization overlay module is used to overlay the standard spatial parameter set onto the three-dimensional acupoint model to obtain a visualized acupoint view of the target patient.

Images